The conversation on this page has been archived and is no longer active.

iMac G5 cooling and design

Message #1 - Posted 2007/05/23 - G.T.

Michelle Steiner wrote:

Previously, G.T. wrote:

Give it a rest already. Your hardware design suggestions reek of cluelessness and naivetÈ.

So do his software design suggestions.

Thankfully I haven't paid attention to those.

Greg

The ticketbastard Tax Tracker:
http://www.ticketmastersucks.org/tracker.html

Dethink to survive - Mclusky

Message #2 - Posted 2007/05/24 - The New Guy

So they should run cool. But the one I saw in the store, actually all 3 INTEL iMacs, (17", 20" and 24") ran HOT. And they were idling. So my point stands. Lousy cooling, lousy design, unless you're some person that gets thrilled on functionless form of course. Then they're just peachy.

My Intel iMac Core Duo runs 24/7 for weeks and even months on end, without any hint of overheating.

Those highly paid hardware and case designers at Apple know a hell of a lot more than you do, and that is evident if for no other reason than by your demonstrated ignorance.

I realize that software is your area of expertise Michelle. In that area you are most specific. Here you are not. When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific. Like I was. Point/counterpoint, if you disagree. And you seem to most definitely disagree. :)

Message #3 - Posted 2007/05/24 - The New Guy

What's an iMac? A computer for non-thinking buyers that enjoy almost zero configurability options and lousy cooling? An iMac should be an all in one computer but that doesn't have to mean almost zero options.

An iMac is a computer with just about everything that's needed built in, so people who are interested in just using the damn machine can do so simply by plugging it in and turning it on.

And that is a wonderful concept, except in the execution of that premise, the buyer is sorely removed from the wonderful world of "choice". Unless you consider the ability to put another stick of ram in "choice". Its a very limiting design. My point was that it didn't have to be. My point was that with better design, Apple could have had a product that appeals not only to the person too lazy and complacent to demand anything better than mediocrity, but to people with a heavy focus in one particular area. And that encompasses a lot of people, whether they want performance in general speed, video or audio, or the size of the monitor. I talk to so many people that wouldn't consider an iMac because of its lack of choice. My point was that it didn't need to be that way. And in the future, maybe they will open their eyes. It could be a product that appeals to the most basic of user as well as the most specific of user.

The only things "missing" are the ability to have more than two monitors, and the ability to have more than one each internal hard disk and optical disk. And since there are easy ways to have external disks, the lack of additional internal ones is a minor issue at best. Even more minor is the third-monitor issue.

Third monitor issue? Who said that?

Message #4 - Posted 2007/05/24 - The New Guy

Now, why he's concerned about the design of a model that was superseded in early 2006 is a puzzlement to me.

Superseded? In what way?

Um, by the fact that Apple stopped making them then and replaced them with the Intel iMacs. You did hear about that, didn't you? Oh, and, for the record, the Intel processors run far cooler than the G5s.

So they should run cool. But the one I saw in the store, actually all 3 INTEL iMacs, (17", 20" and 24") ran HOT. And they were idling. So my point stands. Lousy cooling, lousy design, unless you're some person that gets thrilled on functionless form of course. Then they're just peachy.

You still don't get the concept of efficient heat transfer. Components in a computer generate heat - this is undisputed. The whole point of *any* cooling system is to transfer heat from the components to the air outside of the case - this is also undisputed. The more heat you transfer, and the faster you transfer it, the greater the difference in temperature there is between the hot air coming out of the case and the ambient air outside the case. You seem to think that, instead, the air coming out the case should be closer to room temperature, which would mean the heat isn't being transferred out of the case efficiently and is instead stuck inside, warming the ambient air and other parts of the interior (this is actually quite common for the majority of Wintel PCs, BTW). That's not efficient heat transfer.

If something is being cooled well, the air coming off it is not hot. Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise. In a good design, the heat sink is big enough that it doesn't get hot because the heat is spread over a larger area. That larger area should be cooled by a large fan (120 mm or larger) and since its large it doesn't need to rotate as fast to move the same amount of air. Hence its inaudible or close to the ambient noise of a typical quiet room. With a proper heat sink, the air coming off that device will never be hot. So the air coming out of the case will never be hot. That's good design. My point also stressed that the heat coming out of the iMac was hot and it wasn't even doing anything plus the ambient temperature was not high. The computer was just idling. That's bad design. Style takes paramount importance at Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

Message #5 - Posted 2007/05/24 - Jeffrey Goldberg

On Wed, 23 May 2007, The New Guy wrote:

[...] Its amazing to think that if they had made the cabinet a couple of inches deeper they could have used full height PCI/PCI-E cards (so people that need good video and audio would buy it), and could have used 120 mm (inaudible when running at 5 volts) fans on top and bottom to pull air out of the enclosure so its basically a normal, upgradable machine empowering the buyer and ensuring long term upgradability and scalability. [...]

And for total logic the monitor could have been interchangeable so the audio user who doesn't need a 24" screen could save money and go 17", while the graphics nut could be happy with built on audio while maxing out to the largest screen available.

Well, there is a dirty little secret I'll let you in on. If you want that kind of hardware configurability, modularity, and upgradability don't buy an iMac. It wasn't designed for you.

Now that you know this secret, you will join the rest of the world who have -- somehow -- already managed to discover it.

-j

Jeffrey Goldberg http://www.goldmark.org/jeff/ Relativism is the triumph of authority over truth, convention over justice I rarely read top-posted, over-quoting or HTML postings.

Message #6 - Posted 2007/05/24 - Ockham's Razor

Previously, Michelle Steiner wrote:

Previously, The New Guy wrote:

What's an iMac? A computer for non-thinking buyers that enjoy almost zero configurability options and lousy cooling? An iMac should be an all in one computer but that doesn't have to mean almost zero options.

An iMac is a computer with just about everything that's needed built in, so people who are interested in just using the damn machine can do so simply by plugging it in and turning it on.

The only things "missing" are the ability to have more than two monitors, and the ability to have more than one each internal hard disk and optical disk. And since there are easy ways to have external disks, the lack of additional internal ones is a minor issue at best. Even more minor is the third-monitor issue.

don't you recognize spam?

"When fascism comes to America, it will be wrapped in the flag and carrying a cross."
Sinclair Lewis

Message #7 - Posted 2007/05/24 - Mike Rosenberg

The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

You keep saying that, but it's absurd. It's AIR COOLED!

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #8 - Posted 2007/05/24 - Mike Rosenberg

The New Guy wrote:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #9 - Posted 2007/05/24 - Jeffrey Goldberg

On Thu, 24 May 2007, The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

Where does the heat go then? I must confess that I didn't do particularly well in my thermodynamics course back in college a long time ago, so I don't claim to be an authority, but I've heard the rumor that you can't cool one thing without making something else hotter. Am I wrong about these subtle and tricky ways of the universe?

On a related note, I don't understand how motion detectors work -- the kind used for automatic doors and the like. Therefore they must be misdesigned and broken.

-j

Jeffrey Goldberg http://www.goldmark.org/jeff/ Relativism is the triumph of authority over truth, convention over justice I rarely read top-posted, over-quoting or HTML postings.

Message #10 - Posted 2007/05/24 - Mike Rosenberg

Ockham39;s Razor wrote:

don't you recognize spam?

I think you mean "troll" since it's definitely not spam.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #11 - Posted 2007/05/24 - The New Guy

If something is being cooled well, the air coming off it is not hot.

You keep saying that, but it's absurd. It's AIR COOLED!

Try to reread what I wrote. Again. And also its just simple netiquette not to delete something you're commenting on.

I guess you've never worked with heat sinks. The concept is simple. The larger the heat sink, the lower its temperature cooling a device. Think of a large pot vs. a small pot on your stove. If you put a huge, thick pot on the stove and turn it up high, it will take a while for the water in it to boil. If you put a tiny, thin pot on the stove the water will boil faster as there isn't as much metal for the stove to heat before the heat is transferred to the water. This may not be the best analogy I could have drawn from of course.....:)

Another analogy: you place your warm hand around a small piece of cold, thin metal and it instantly is warmed by your hand vs. placing your warm hand around a cold, solid pipe which is not going to get warm at all. The larger the device is that is absorbing the heat, the cooler the temperature of that device will be.

That's why good cooling designs employ large heat sinks. To those of you who's feelings have been hurt by a critical statement being leveled at your most revered, hallowed, most high esteemed company, take heart. The Mac Pro has good CPU cooling. You'll notice they've used a proven design similar (or maybe its a genuine Scythe!) used by Scythe in their Ninjz for years. Its the highest rated heat sink in the world. But this little post was about the iMac and its possibilities. Sorry buddies......:)

Message #12 - Posted 2007/05/24 - Mike Rosenberg

The New Guy wrote:

Try to reread what I wrote. Again.

So I can laugh again?

And also its just simple netiquette not to delete something you're commenting on.

I didn't delete anything. Your post is still right there in its entirety and archived for anyone to read whenver they want to. My response cited only the portions I was responding to, which is good netiquette.

Anyway, you can keep repeating your contention over and over, but that doesn't make it any less absurd. Showing reports of all those G5 iMacs that have failed from overheating, on the other hand, would be convincing evidence that they have "lousy cooling." Show us the data.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #13 - Posted 2007/05/24 - The New Guy

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard? Its going to vastly shorten the lifespan of its components. And anybody in electronics knows that heat is not your friend. Generally, the cooler things run, the longer they last.

Talking about a solution:
I wonder if anyone has run their iMac without the back cover to see if it runs significantly cooler? I've never seen the interior of one, nor do I know how the hot air is conducted. So it would be interesting to see if it makes a difference. A brief search at Flickr didn't bring up anything. But Googling, I did find someone who found an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

Message #14 - Posted 2007/05/24 - The New Guy

Try to reread what I wrote. Again.

So I can laugh again?

And also its just simple netiquette not to delete something you're commenting on.

I didn't delete anything. Your post is still right there in its entirety and archived for anyone to read whenver they want to. My response cited only the portions I was responding to, which is good netiquette.

Anyway, you can keep repeating your contention over and over, but that doesn't make it any less absurd. Showing reports of all those G5 iMacs that have failed from overheating, on the other hand, would be convincing evidence that they have "lousy cooling." Show us the data.

They may not have failed en mass but my point was that the lower the temperature inside your computer, the longer its lifespan. Its a worthy aim of any electronics engineer to focus on heat dissipation.

Message #15 - Posted 2007/05/24 - John McWilliams

The New Guy wrote:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard? Its going to vastly shorten the lifespan of its components. And anybody in electronics knows that heat is not your friend. Generally, the cooler things run, the longer they last.

Generally, the more an anonymous poster keeps pounding on the same subject, the wronger he is.

lsmft

Message #16 - Posted 2007/05/24 - Mike Rosenberg

The New Guy wrote:

Generally, the cooler things run, the longer they last.

Uh huh. Let's see how low we need to reduce your core temperature before you rethink that statement.

Seriously, everything has an optimal operating temperature _range_. Stay within that range, everything is fine. Go outside that range, in either direction, not so fine anymore.

Please provide evidence that a G5 iMac runs outside its optimal range find another topic to run on and on about.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #17 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

Wrong. It is being cooled by the air heating up and carrying the heat away.

Its usually because of the heat sink size.

The heat sink merely stores the heat away from a component--the CPU and/or memory in this case. The heat sink can hold only so much heat, just like a kitchen sink can hold only so much water. That heat has to be removed from the sink. It is removed by transferring the heat to air. The hot air then moves out of the enclosure (either strictly by convection or by aid of a fan), thus reducing the heat level inside the enclosure as cooler air moves in to replace it.

Once again, you have demonstrated that you don't know what you're talking about.

Support the troops: Bring them home ASAP.

Message #18 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

I guess you've never worked with heat sinks.

You obviously haven't.

That's why good cooling designs employ large heat sinks. To those of you who's feelings have been hurt by a critical statement being leveled at your most revered, hallowed, most high esteemed company, take heart.

Our feelings haven't been hurt, but our opinion of you has plummeted to even lower depths. If you knew ten times what you actually know, you'd know one tenth of what you think you know.

Support the troops: Bring them home ASAP.

Message #19 - Posted 2007/05/24 - The New Guy

If something is being cooled well, the air coming off it is not hot.

It is being cooled by the air heating up and carrying the heat away.

True. But if a large enough heat sink is used, the heat sink itself won't get hot. Ideally it should get warm and have enough airflow around it to dissipate that warmth. But if the heat sink is too small it will get hot thereby requiring more airflow not to endanger the device being cooled.

Is this so difficult to comprehend?

Its usually because of the heat sink size.

The heat sink merely stores the heat away from a component--the CPU and/or memory in this case. The heat sink can hold only so much heat, just like a kitchen sink can hold only so much water.

Exactly. And the amount of heat that heat sink can hold is proportional to its heat absorbing size. If you put a 10' x 10' x 10' heat sink on your little cpu, do you think it would get hot? If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size. The bigger it is, more area it has to dissipate that stored heat. That's why some modern fanless computers can still cool their components adequately because of large heat sinks - and the heat pipes bring the heat to them of course. Here they have opted for this Zalman case to radiate the heat so it in itself becomes a huge heat sink:
http://www.devhardware.com/c/a/Computer-Cases/Zalman-TNN-500A-Fanless-P C-Case-Review/

That heat has to be removed from the sink. It is removed by transferring the heat to air. The hot air then moves out of the enclosure (either strictly by convection or by aid of a fan), thus reducing the heat level inside the enclosure as cooler air moves in to replace it.

Yeah.....I think we got that part.........:) The sky is blue too.

Message #20 - Posted 2007/05/24 - Michelle Steiner

It is bad netiquette not to reference the message to which you are replying.

Previously, The New Guy wrote:

An iMac is a computer with just about everything that's needed built in, so people who are interested in just using the damn machine can do so simply by plugging it in and turning it on.

And that is a wonderful concept, except in the execution of that premise, the buyer is sorely removed from the wonderful world of "choice".

The buyer has a choice; he can choose to buy an iMac or to buy another computer.

Its a very limiting design. My point was that it didn't have to be.

It is not limiting at all; it has just about everything built in.

My point was that with better design, Apple could have had a product that appeals not only to the person too lazy and complacent to demand anything better than mediocrity, but to people with a heavy focus in one particular area.

Your point is that you will fabricate whatever issues you can dream up, no matter how inane, to take pot shots at Apple.

I talk to so many people that wouldn't consider an iMac because of its lack of choice.

So those people choose another computer. Big frigging deal. The point is that Apple puts just about everything into the iMac that people may want or need. And for some people, there are things that they don't want or need--like Bluetooth and Wireless.

You have yet to show what is missing from the iMac.

Support the troops: Bring them home ASAP.

Message #21 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Those highly paid hardware and case designers at Apple know a hell of a lot more than you do, and that is evident if for no other reason than by your demonstrated ignorance.

I realize that software is your area of expertise Michelle. In that area you are most specific. Here you are not.

I have been very specific; you chose to ignore that.

Be specific. Like I was.

You haven't been specific at all. You said that the iMac doesn't offer any choices other than to add RAM. You haven't said what choices the iMac doesn't offer.

Support the troops: Bring them home ASAP.

Message #22 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

Stop stalling. Show us where these problems actually exist.

Speaking of hot air, you're full of it.

Support the troops: Bring them home ASAP.

Message #23 - Posted 2007/05/24 - The New Guy

Those highly paid hardware and case designers at Apple know a hell of a lot more than you do, and that is evident if for no other reason than by your demonstrated ignorance.

I realize that software is your area of expertise Michelle. In that area you are most specific. Here you are not.

I have been very specific; you chose to ignore that.

Be specific. Like I was.

You haven't been specific at all. You said that the iMac doesn't offer any choices other than to add RAM. You haven't said what choices the iMac doesn't offer.

Well that was mentioned in the original post, eons ago, or so it seems.

I'm curious Michelle, you replied to this post way after a far more recent post. ?? Perhaps you're sorting your messages by subject, not by date? That would explain it perhaps. Or maybe the news server is posting the messages not by date? Could that be possible? Any usenet experts here today?

Message #24 - Posted 2007/05/24 - Ockham's Razor

Previously, Mike Rosenberg wrote:

Ockham39;s Razor wrote:

don't you recognize spam?

I think you mean "troll" since it's definitely not spam.

Correct. I had been reading about Spam in the NYT earlier and just carried it over.

"When fascism comes to America, it will be wrapped in the flag and carrying a cross."
Sinclair Lewis

Message #25 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

It is being cooled by the air heating up and carrying the heat away.

True. But if a large enough heat sink is used, the heat sink itself won't get hot.

Actually, the heat sink will have the same amount of heat, just spread over a larger volume. Eventually that heat will have to be carried out of the enclosure--and that will be via air flow. The greater the volume of air and the faster it moves, the cooler the air will be. But the temperature of the air is not relevant.

That heat has to be removed from the sink. It is removed by transferring the heat to air. The hot air then moves out of the enclosure (either strictly by convection or by aid of a fan), thus reducing the heat level inside the enclosure as cooler air moves in to replace it.

Yeah.....I think we got that part.........:) The sky is blue too.

No, you didn't get that part, as evidenced by your rant.

Support the troops: Bring them home ASAP.

Message #26 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

I'm curious Michelle, you replied to this post way after a far more recent post. ?? Perhaps you're sorting your messages by subject, not by date?

Yup, I sort by subject; that way most messages on the same subject are grouped together.

Support the troops: Bring them home ASAP.

Message #27 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 11:30:42 -0500, Jeffrey Goldberg said:

On Thu, 24 May 2007, The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

Where does the heat go then? I must confess that I didn't do particularly well in my thermodynamics course back in college a long time ago, so I don't claim to be an authority, but I've heard the rumor that you can't cool one thing without making something else hotter. Am I wrong about these subtle and tricky ways of the universe?

No, you're right of course. The fool just doesn't get it.

On a related note, I don't understand how motion detectors work -- the kind used for automatic doors and the like. Therefore they must be misdesigned and broken.

LOL...

JR

Message #28 - Posted 2007/05/24 - Daniel Packman

Previously, The New Guy wrote:

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

There is no way for you to know what the interior temperature is from this information. If there is lots of cool air being expelled, the cpu could be out of the air stream and melting. You are making a poor assumption in tying temperature of expelled air to interior temperature.

..... But Googling, I did find someone who found an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

This is the way to do it --- actually measure internal temperatures. But his experience with overheating seems atypical.

Message #29 - Posted 2007/05/24 - G.T.

The New Guy wrote:

Now, why he's concerned about the design of a model that was superseded in early 2006 is a puzzlement to me.

Superseded? In what way?

Um, by the fact that Apple stopped making them then and replaced them with the Intel iMacs. You did hear about that, didn't you? Oh, and, for the record, the Intel processors run far cooler than the G5s.

So they should run cool. But the one I saw in the store, actually all 3 INTEL iMacs, (17", 20" and 24") ran HOT. And they were idling. So my point stands. Lousy cooling, lousy design, unless you're some person that gets thrilled on functionless form of course. Then they're just peachy.

You still don't get the concept of efficient heat transfer. Components in a computer generate heat - this is undisputed. The whole point of *any* cooling system is to transfer heat from the components to the air outside of the case - this is also undisputed. The more heat you transfer, and the faster you transfer it, the greater the difference in temperature there is between the hot air coming out of the case and the ambient air outside the case. You seem to think that, instead, the air coming out the case should be closer to room temperature, which would mean the heat isn't being transferred out of the case efficiently and is instead stuck inside, warming the ambient air and other parts of the interior (this is actually quite common for the majority of Wintel PCs, BTW). That's not efficient heat transfer.

If something is being cooled well, the air coming off it is not hot.

Wow, you're still a complete FUCKING MORON. Heat doesn't magically disappear, it just gets moved around.

Greg

The ticketbastard Tax Tracker:
http://www.ticketmastersucks.org/tracker.html

Dethink to survive - Mclusky

Message #30 - Posted 2007/05/24 - G.T.

The New Guy wrote:

If something is being cooled well, the air coming off it is not hot.

You keep saying that, but it's absurd. It's AIR COOLED!

Try to reread what I wrote. Again. And also its just simple netiquette not to delete something you're commenting on.

I guess you've never worked with heat sinks. The concept is simple. The larger the heat sink, the lower its temperature cooling a device. Think of a large pot vs. a small pot on your stove. If you put a huge, thick pot on the stove and turn it up high, it will take a while for the water in it to boil. If you put a tiny, thin pot on the stove the water will boil faster as there isn't as much metal for the stove to heat before the heat is transferred to the water. This may not be the best analogy I could have drawn from of course.....:)

You're a fucking idiot polluting Usenet. You can't even see that what say in that last paragraph either contradicts or has nothing to do with your line: "If something is being cooled well, the air coming off it is not hot." The better the heat transfer the hotter the air is going to be. IT IS REALLY REALLY SIMPLE. How you spread the resulting heat is another matter altogether.

In a case with little room to work that heat is going to be squirting out small vents and it's going to be HOT. BECAUSE THE HEAT SINKS AND FANS ARE DOING THEIR JOBS but there are limited opportunities to get rid of that heat.

Greg

The ticketbastard Tax Tracker:
http://www.ticketmastersucks.org/tracker.html

Dethink to survive - Mclusky

Message #31 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 09:23:53 -0500, The New Guy said:

If something is being cooled well, the air coming off it is not hot.

You are confusing heat concentration for cooling efficiency. Bad idea. Don't do that. The two are largely unrelated.

Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise.

While this is common in Windows PCs, it's not common in Macs.

In a good design, the heat sink is big enough that it doesn't get hot because the heat is spread over a larger area.

It doesn't matter how large a heat sink is, or hot a heat sink gets, as long as the air flow is substantial enough to transfer heat off the heat sink and out of the case.

That larger area should be cooled by a large fan (120 mm or larger) and since its large it doesn't need to rotate as fast to move the same amount of air. Hence its inaudible or close to the ambient noise of a typical quiet room.

Are you suggesting the fans in today's Macs are too loud? please stay on topic - we are talking about cooling system efficiency, not fan noise.

With a proper heat sink, the air coming off that device will never be hot.

Air flow is what transfers heat from sinks. Increasing heat sink size exposes more of the sink to more air. If the air doesn't move, ambient temperature rises, and eventually, you will be unable to transfer heat off the sink. It's air flow that matters. Put a heat sink on any component with zero air flow in a confined area, and if the ambient temperature gets too high, you're toast.

So the air coming out of the case will never be hot. That's good design.

You seem to think the heat just disappears inside the case! Reality is the heat sink is not transferring enough heat from the component into the air, so the heat is likely increasing ambient inner-case temperature, and worse, increasing the temperature of surfaces and other components inside the case.

My point also stressed that the
heat coming out of the iMac was hot and it wasn't even doing anything plus the ambient temperature was not high. The computer was just idling. That's bad design.

Again, you are confusing heat concentration for cooling efficiency. Space is tight in iMacs, so the air flow is constricted - heat being transferred is concentrated as a result. None of this equates to bad design. The computer was running, whether idle or not, so electricity was flowing through components. Therefore heat was being generated. When Mac OS X is idle, it adjusts the variable-speed fans to balance heat transfer with quiet operation. When you start pushing the CPU and other components, as heat increases, Mac OS X increases fan speed to match so component temperature stays within spec.

Style takes paramount importance at
Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

You clearly don't understand Apple's target market.

JR

Message #32 - Posted 2007/05/24 - Jeffrey Goldberg

On Thu, 24 May 2007, The New Guy wrote:

I guess you've never worked with heat sinks. The concept is simple. The larger the heat sink, the lower its temperature cooling a device.

Yes. We all understand that. But eventually the heat has to leave the box altogether, no matter what the size of any heat sinks within it.

Are you saying that a better design would be the transfer the heat by moving more air (so that the same amount of heat would be transfered but the heated air temperature would be lower)? If so, you don't express yourself very well.

If that is what you are saying, then say it. Then we can argue about what design works best with the form factor that Apple was shooting for. But since you've already said that you'd be happy to see the iMac a couple inches thicker, I don't see much point in actually discussing with you the trade offs Apple had to make.

My first reaction when I saw one of these iMac's up close was "How the hell do they cool this thing?" That was almost two years ago. My wife bought one at that time. We live in Texas and we don't like running the A/C more than we have to. So her iMac with the thin little vent has been doing just fine. One week last summer or the summer before, I actually monitored the CPU temperature. I was impressed to see how steady it was. I don't remember the actual results, but it certainly boosted by confidence in the Apple's design.

-j

Jeffrey Goldberg http://www.goldmark.org/jeff/ Relativism is the triumph of authority over truth, convention over justice I rarely read top-posted, over-quoting or HTML postings.

Message #33 - Posted 2007/05/24 - P. Sture

Previously, The New Guy wrote:

But Googling, I did find someone who found
an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

Oops. The author clearly states that he doesn't know if his mod voids the warranty:

"It is unknown if this will void your warranty or not, so if you do this on your own please keep this in mind."

Going back to your earlier comment:

I wonder if anyone has run their iMac without the back cover to see if it runs significantly cooler?

This can be a dangerous thing to do as it could create spots where air no longer circulates at all.

Paul Sture

Message #34 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 11:59:15 -0500, The New Guy said:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

The fan speed increases based on the temperature, which keeps the component temperature within spec, dummy.

JR

Message #35 - Posted 2007/05/24 - Adrian

Mike Rosenberg wrote:

Ockham39;s Razor wrote:

don't you recognize spam?

I think you mean "troll" since it's definitely not spam.

Yes, New Guy is worthy of Lord of the Rings ....

Adrian

Message #36 - Posted 2007/05/24 - Adrian

The New Guy wrote:

If something is being cooled well, the air coming off it is not hot. Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise. In a good design, the heat sink is big enough that it doesn't get hot because the heat is spread over a larger area. That larger area should be cooled by a large fan (120 mm or larger) and since its large it doesn't need to rotate as fast to move the same amount of air. Hence its inaudible or close to the ambient noise of a typical quiet room. With a proper heat sink, the air coming off that device will never be hot. So the air coming out of the case will never be hot. That's good design. My point also stressed that the heat coming out of the iMac was hot and it wasn't even doing anything plus the ambient temperature was not high. The computer was just idling. That's bad design. Style takes paramount importance at Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

Keep going ... if you say it often enough maybe someone other than yourself will begin to believe it!

You could save some typing by just cutting and pasting from that earlier thread (the last time you pursued this fruitless cooling argument, remember?)

Adrian

Message #37 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 13:40:24 -0500, Daniel Packman said:

Previously, The New Guy wrote:

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

There is no way for you to know what the interior temperature is from this information. If there is lots of cool air being expelled, the cpu could be out of the air stream and melting. You are making a poor assumption in tying temperature of expelled air to interior temperature.

His flawed assumption is that Macs are built like most PCs.

Most PC cases don't move air directly over components and don't separate air flow into targeted zones. Instead, they rely on ambient air inside the case to absorb heat from heat sinks and components, and then they pull that heated ambient air out of the case with a couple fans in the rear.

Today's Macs, on the other hand, have contoured and channeled air pathways inside of the case that funnel air from outside directly over components and through heat sinks attached to components, and that warmed air is expelled out of the case in a concentrated manner.

..... But Googling, I did find someone who found an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

This is the way to do it --- actually measure internal temperatures.

A while back, when The New Guy was on this very same rant, I took actual temperature measurements inside the case of my G5 tower. The measurements showed that the cooling system is quite efficient. Here's a recap, for those interested:

I happen to have a portable Radio Shack digital thermometer handy. : )

Note: I'll have to rest the sensor end of the wire on the bare metal floor of some compartments, but hopefully it won't pick up too much heat from the metal. We'll just say the temperature readings could be off by +/- a couple degrees.

In each measurement, I'll place the sensor with preference to placements where the sensor is dangling freely in the ambient air of the compartment, will completely close the case up again, and will wait 5 minutes for the temperature reading to settle.

Ambient office temperature: 77.6Àö F

Inside G5 tower (from top to bottom):

• CD/DVD/Hard drive section (front): 80.5˚ F • CD/DVD/Hard drive section (rear): 88.7˚ F

Difference: 8.2Àö F

• PCI/AGP/video card section (front): 80.4˚ F • PCI/AGP/video card section (rear): 85.2˚ F

Difference: 4.8Àö F

• RAM/CPU section (front): 79.1˚ F • RAM/CPU section (rear): 88.3˚ F

Difference: 9.2Àö F

Since the air is constantly moving from front to back in an efficient manner, there is a wide difference between the temperature in the front of the case and the temperature in the back of the case, as you would expect with a constant, efficient air flow. If the air inside the case were stagnant, as it is in most PC cases, this difference would not be nearly as significant. This shows the cooling system is doing a good job.

As far as your hypothesis that much of the heat is rising towards the top of the case rather than going straight out the back in an efficient manner, here is a recap of the difference in ambient front temperatures:

• CD/DVD/Hard drive section (front): 80.5˚ F • PCI/AGP/video card section (front): 80.4˚ F • RAM/CPU section (front): 79.1˚ F

So total bottom-to-top ambient internal temperature difference is a minimal 1.4Àö F. This shows that upwards heat radiation is kept to a minimum by the cooling system.

JR

Message #38 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 12:02:15 -0500, The New Guy said:

Try to reread what I wrote. Again.

So I can laugh again?

And also its just simple netiquette not to delete something you're commenting on.

I didn't delete anything. Your post is still right there in its entirety and archived for anyone to read whenver they want to. My response cited only the portions I was responding to, which is good netiquette.

Anyway, you can keep repeating your contention over and over, but that doesn't make it any less absurd. Showing reports of all those G5 iMacs that have failed from overheating, on the other hand, would be convincing evidence that they have "lousy cooling." Show us the data.

They may not have failed en mass but my point was that the lower the temperature inside your computer, the longer its lifespan. Its a worthy aim of any electronics engineer to focus on heat dissipation.

So far you have given zero evidence that Apple hasn't focused on heat dispersion.

JR

Message #39 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 12:39:48 -0500, The New Guy said:

If something is being cooled well, the air coming off it is not hot.

It is being cooled by the air heating up and carrying the heat away.

True. But if a large enough heat sink is used, the heat sink itself won't get hot. Ideally it should get warm and have enough airflow around it to dissipate that warmth. But if the heat sink is too small it will get hot thereby requiring more airflow not to endanger the device being cooled.

Is this so difficult to comprehend?

Again, you are making the flawed assumption that air in Macs is stagnant and that stagnant air is absorbing the heat from the sink, increasing the ambient air temperature inside the case. This is how most PCs today are designed. In such an environment, what you say is true, because the heat-absorbing capacity of the ambient air has limits, so the smaller the heat sink, the less heat will transfer to the ambient air around it. But that is *not* how Macs are designed today. Today's Macs have custom compartments and channels that direct cool incoming air directly over components, through heat sinks, and out of the case. In this type of environment, heat sink size does not directly correspond to the temperature of exiting air.

Its usually because of the heat sink size.

The heat sink merely stores the heat away from a component--the CPU and/or memory in this case. The heat sink can hold only so much heat, just like a kitchen sink can hold only so much water.

Exactly. And the amount of heat that heat sink can hold is proportional to its heat absorbing size. If you put a 10' x 10' x 10' heat sink on your little cpu, do you think it would get hot?

Whether it would get hot depends completely on AIR FLOW, as I have said so many times I am growing tired of typing it.

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

JR

Message #40 - Posted 2007/05/24 - The New Guy

If something is being cooled well, the air coming off it is not hot.

It is being cooled by the air heating up and carrying the heat away.

True. But if a large enough heat sink is used, the heat sink itself won't get hot.

Actually, the heat sink will have the same amount of heat, just spread over a larger volume.

Yes. And it won't get hot. If it doesn't get hot, the air inside the case won't get hot. Every case will have SOME airflow. Better cases have better airflow. So air escapes. And if the heat sink doesn't get hot, that air escaping won't get hot. So the temperature of the air leaving the case IS indicative of good design and cooling.

Eventually that heat will have to be carried out of the enclosure--and that will be via air flow. The greater the volume of air and the faster it moves, the cooler the air will be. But the temperature of the air is not relevant.

See above.

Message #41 - Posted 2007/05/24 - The New Guy

Previously, Michelle Steiner wrote:

Previously, The New Guy wrote:

I'm curious Michelle, you replied to this post way after a far more recent post. ?? Perhaps you're sorting your messages by subject, not by date?

Yup, I sort by subject; that way most messages on the same subject are grouped together.

Thanks. That makes sense. Now for my next rant......:)

Message #42 - Posted 2007/05/24 - The New Guy

Now, why he's concerned about the design of a model that was superseded in early 2006 is a puzzlement to me.

Superseded? In what way?

Um, by the fact that Apple stopped making them then and replaced them with the Intel iMacs. You did hear about that, didn't you? Oh, and, for the record, the Intel processors run far cooler than the G5s.

So they should run cool. But the one I saw in the store, actually all 3 INTEL iMacs, (17", 20" and 24") ran HOT. And they were idling. So my point stands. Lousy cooling, lousy design, unless you're some person that gets thrilled on functionless form of course. Then they're just peachy.

You still don't get the concept of efficient heat transfer. Components in a computer generate heat - this is undisputed. The whole point of *any* cooling system is to transfer heat from the components to the air outside of the case - this is also undisputed. The more heat you transfer, and the faster you transfer it, the greater the difference in temperature there is between the hot air coming out of the case and the ambient air outside the case. You seem to think that, instead, the air coming out the case should be closer to room temperature, which would mean the heat isn't being transferred out of the case efficiently and is instead stuck inside, warming the ambient air and other parts of the interior (this is actually quite common for the majority of Wintel PCs, BTW). That's not efficient heat transfer.

If something is being cooled well, the air coming off it is not hot.

Wow, you're still a complete MORON. Heat doesn't magically disappear, it just gets moved around.

I guess in a poor design it will just move around and around inside the case. In a good design, the heat sink won't get hot, thereby not heating the air in the case. You're forgetting that every case has some airflow. Better cases have better airflow. So as soon as the slightest increase of temperature occurs inside the case, that heated air rises and is drawn out of the case. That heated air will be warm at best, but never hot, if its designed well.

Message #43 - Posted 2007/05/24 - The New Guy

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

There is no way for you to know what the interior temperature is from this information. If there is lots of cool air being expelled, the cpu could be out of the air stream and melting. You are making a poor assumption in tying temperature of expelled air to interior temperature.

But the expelled air IS the interior air being vented. So its one and the same. If its hot when idling, its going to get a whole lot hotter when its being stressed by a demanding task. A well designed case expels warm air, not hot air. A really well designed case expels air just slightly above the ambient room temperature. That's hard to do, granted. Expelling hot air when idling is just not good.

..... But Googling, I did find someone who found an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

This is the way to do it --- actually measure internal temperatures. But his experience with overheating seems atypical.

Could be. Like I said before, just because your system isn't shutting down because of heat issues doesn't mean the design is good. Cool components just last longer. Probably a lot longer. To me it seems a worthy design goal. For others, when their first component dies a premature death and is replaced by the latest and greatest, perhaps not. Its an excuse to upgrade in a way.

Message #44 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 15:28:41 -0500, The New Guy said:

In a good design, the heat sink won't get hot, thereby not heating the air in the case.

ROFLMAO

JR

Message #45 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 15:28:41 -0500, The New Guy said:

You're forgetting that every case has
some airflow. Better cases have better airflow. So as soon as the slightest increase of temperature occurs inside the case, that heated air rises and is drawn out of the case.

LOL. A good cooling system doesn't allow heat to rise to begin with - it expels the heated air ASAP - this is exactly what today's Macs do.

JR

Message #46 - Posted 2007/05/24 - The New Guy

If something is being cooled well, the air coming off it is not hot.

You keep saying that, but it's absurd. It's AIR COOLED!

Try to reread what I wrote. Again. And also its just simple netiquette not to delete something you're commenting on.

I guess you've never worked with heat sinks. The concept is simple. The larger the heat sink, the lower its temperature cooling a device. Think of a large pot vs. a small pot on your stove. If you put a huge, thick pot on the stove and turn it up high, it will take a while for the water in it to boil. If you put a tiny, thin pot on the stove the water will boil faster as there isn't as much metal for the stove to heat before the heat is transferred to the water. This may not be the best analogy I could have drawn from of course.....:)

You can't even see that what you say in that last paragraph either contradicts or has nothing to do with
your line: "If something is being cooled well, the air coming off it is not hot." The better the heat transfer the hotter the air is going to be. IT IS REALLY REALLY SIMPLE. How you spread the resulting heat is another matter altogether.

But don't you see? The heat sink is doing that exact thing. Spreading the heat. That's its role. A large heat sink with lots of radiating area spreads the heat fast so the air that moves through the case expels it before the temperature gets hot.

Maybe the problem here is that some of you are viewing the case as a sealed enclosure where every bit of heat generated is retained. But that is the opposite of a computer case which attempts to expel warm air as rapidly as possible. You always have airflow. Some cases just do it better than others. But it all boils down to the heat sink. And since both aluminum and copper (the best heat transfer materials) are expensive for manufacturers to purchase, so they will always try to use a minimum of it. Its way cheaper to use tiny, cheap fans than large heat sinks.

In a case with little room to work that heat is going to be squirting out small vents and it's going to be HOT.

That's just a bad design of a case! In the ultimate case, every heat generator is cooled and that warmed air from that heat generator is expelled immediately so not to raise the interior temperature of the case. Most computers have just the CPU, video card and hard drives producing most of the heat. Those are our heat challenges. Ram, chipsets and optical devices (DVD burners when they are being used of course) produce a litte more as well.

BECAUSE THE HEAT SINKS AND
FANS ARE DOING THEIR JOBS but there are limited opportunities to get rid of that heat. Greg

Greg, in a well designed case, heat is expelled much faster than in a poorly designed case. The faster the heat is expelled, the lower the interior temperature of the case. The lower the interior temperature, the lower the temperature of the expelled air will be and those components inside that case will last longer.

Message #47 - Posted 2007/05/24 - G.T.

"The New Guy" <replytogroup@here.thanks> wrote in message news:replytogroup-3A61E4.15450324052007@news.lga.highwinds-media.com...

Greg, in a well designed case, heat is expelled much faster than in a poorly designed case.

No fucking shit but that is completely orthogonal to "If something is being cooled well, the air coming off it is not hot." Anything that gets hot will create hot air. It's BASIC THERMODYNAMICS.

Greg

"What have you got in that paper bag?
Is it a dose of Vitamin C?
Ain't got no time for Western medicine
I am Damo Suzuki" - Mark E Smith

Message #48 - Posted 2007/05/24 - G.T.

"Jolly Roger" <jollyroger@R.E.M.O.V.E.pobox.com> wrote in message news:2007052415344031729-jollyroger@REMOVEpoboxcom...

On 2007-05-24 15:28:41 -0500, The New Guy said:

In a good design, the heat sink won't get hot, thereby not heating the air in the case.

ROFLMAO

No kidding. He clearly isn't putting two and two together, or he doesn't know how to express himself.

Greg

Ticketbastard tax tracker:
http://ticketmastersucks.org/tracker.html

Dethink to survive - Mclusky

Message #49 - Posted 2007/05/24 - The New Guy

Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise.

While this is common in Windows PCs, it's not common in Macs.

What's the difference between a Windows machine and a Mac these days? Not the CPU, hard drives, optical drives. Just the motherboard. The CPU in both is the primary heat generator. That, along with the video card and hard drives generate most of the heat in both machines. Look at the large heat sinks in a Mac Pro. Then look at a Scythe Ninja heat sink. They look very close. Maybe it is a Scythe Ninja.....I don't know. They're both computers with the same design challenges.

In a good design, the heat sink is big enough that it doesn't get hot because the heat is spread over a larger area.

It doesn't matter how large a heat sink is, or hot a heat sink gets, as long as the air flow is substantial enough to transfer heat off the heat sink and out of the case.

But you can only get lots of airflow with high rpm fans that produce lots of noise. Using a well designed heat sink doesn't require such a large fan. Remember there is always airflow in a computer case, whether its a Mac or PC. Now here we have a problem: The depth of the iMac case. Remember in my original post I mentioned that if only they had made the case a couple of inches deeper, they could have used 120 mm or larger fans running at 5 volts. Those produce reasonable amounts of airflow and are inaudible. But they are only effective if the heat sink is good. That means on the large side.

That larger area should be cooled by a large fan (120 mm or larger) and since its large it doesn't need to rotate as fast to move the same amount of air. Hence its inaudible or close to the ambient noise of a typical quiet room.

Are you suggesting the fans in today's Macs are too loud? please stay on topic - we are talking about cooling system efficiency, not fan noise.

Nope - my original post lamented that the iMacs cooling system didn't work effectively, hence the hot air being expelled when idling. Whether you want to change that with loud fans, or better heat sinks is up to the designer. I'd opt for large heat sinks with inaudible fans, but that's just me. I hate noise. When you're battling noise, you'll soon learn to just about give up on any fan smaller than 120 mm especially since 120 mm fans are dirt cheap nowadays.

With a proper heat sink, the air coming off that device will never be hot.

Air flow is what transfers heat from sinks. Increasing heat sink size exposes more of the sink to more air. If the air doesn't move, ambient temperature rises, and eventually, you will be unable to transfer heat off the sink. It's air flow that matters.

Agreed. And all cases have airflow. Some of you are thinking that we're dealing with heat generation in a sealed box with no airflow.

Put a heat sink on any component with zero air flow in a confined area, and if the ambient temperature gets too high, you're toast.

But that's a sealed box - and that's the mindset some of you are in. A computer case should be the opposite of a sealed box. It should expel heated air as quickly as possible so it doesn't build up. If it does, that expelled air will not be hot. It will be warm at best.

So the air coming out of the case will never be hot. That's good design.

You seem to think the heat just disappears inside the case!

No, a good design expels that heat immediately so the interior temperature hardly rises at all from the ambient room temperature. You seem to think the computer case is sealed.

Reality is
the heat sink is not transferring enough heat from the component into the air, so the heat is likely increasing ambient inner-case temperature, and worse, increasing the temperature of surfaces and other components inside the case.

Yes - in a bad design, that is exactly what is happening. And that is exactly what happens when you feel hot air being expelled from a case design like the iMac! And when its happening when the unit is idling, its further proof that things are very wrong inside that box. Hopefully Michelle and her iMac enjoy air conditioning on High this summer. An interesting solution would be to vent air from an air conditioner directly to the input vent(s) of your computer. If moisture wasn't a problem, that would be great. Its kind of silly trying to cool something with hot, interior case air. I once talked to a guy who vented outside air (below 0 F in winter) directly to his computer. He said it never ran so well. I tried it with my G4 tower and it worked great. Though, granted, moisture can be a problem so you have to be vigilant.

My point also stressed that the
heat coming out of the iMac was hot and it wasn't even doing anything plus the ambient temperature was not high. The computer was just idling. That's bad design.

Again, you are confusing heat concentration for cooling efficiency. Space is tight in iMacs, so the air flow is constricted - heat being transferred is concentrated as a result.

No. The case design is bad in iMacs so the air flow is constricted. Its just an example of designers sacrificing function (cooling) for style. If they'd made the case just a little deeper, cooling would have been effortless. Granted, it most definitely wouldn't have looked so sleek and pretty.

Style takes paramount importance at
Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

You clearly don't understand Apple's target market.

My original post was that they could have adjusted the style, by not much, and got most of the present target market, as well as tons of PC users that are tired of a very flawed OS and unending virus infection. After all, at about 5% of the market, their market penetration depends on wooing PC users over. Linux users have found a virus free, stable OS that runs far more efficiently than Windows on cheap hardware, so they're certainly not coming over anytime soon. Once Apple got their hardware and pricing on a serious level, PC users would start coming to Apple outlets to buy their computers. Once there, they could be shown the superiority of the Apple OS. THEN you have a convert. Very few people are converted without seeing and feeling the difference. I look at potential.

Message #50 - Posted 2007/05/24 - Dave Balderstone

Previously, The New Guy wrote:

My original post was that they could have adjusted the style, by not much, and got most of the present target market, as well as tons of PC users that are tired of a very flawed OS and unending virus infection. After all, at about 5% of the market, their market penetration depends on wooing PC users over. Linux users have found a virus free, stable OS that runs far more efficiently than Windows on cheap hardware, so they're certainly not coming over anytime soon. Once Apple got their hardware and pricing on a serious level, PC users would start coming to Apple outlets to buy their computers. Once there, they could be shown the superiority of the Apple OS. THEN you have a convert. Very few people are converted without seeing and feeling the difference. I look at potential.

I find it amazing that you obviously know so much more about modern computer design than Apple.

What was *your* gross income last quarter? You must be astoundingly wealthy.

Have you applied to work at Apple, advising the design and marketing teams? I'm certain they would have a place for you.

Message #51 - Posted 2007/05/24 - The New Guy

I guess you've never worked with heat sinks. The concept is simple. The larger the heat sink, the lower its temperature cooling a device.

Yes. We all understand that. But eventually the heat has to leave the box altogether, no matter what the size of any heat sinks within it.

Ideally, the heat leaves immediately and doesn't just sit there raising the interior temperature.

Are you saying that a better design would be the transfer the heat by moving more air (so that the same amount of heat would be transfered but the heated air temperature would be lower)?

Not really, though moving more air across the heat sink helps the cooling process. The trouble with moving more air is the noise. I suggested to look to the source which is the heat sink itself since we can't very much control the CPU heat without underclocking it which destroys performance. A better designed heat sink doesn't require as much airflow. You can use 120 mm fans (dirt cheap) at lower voltages for inaudible cooling. And if your device is being cooled well and you can't hear the cooling process, you have most definitely reached your cooling goal. And the air exiting that case will not be hot, especially when its idling.

If that is what you are saying, then say it. Then we can argue about what design works best with the form factor that Apple was shooting for. But since you've already said that you'd be happy to see the iMac a couple inches thicker, I don't see much point in actually discussing with you the trade offs Apple had to make.

Do most of you think that the iMac would be far less popular if the case was just a couple of inches thicker? Remember that when you're in front of the computer you can't even see the depth of the case, whether it be 2" or 4". Though, when people are making their buying decision, whether it be in person or online, they are looking at it form the side as well.

Message #52 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Actually, the heat sink will have the same amount of heat, just spread over a larger volume.

Yes. And it won't get hot. If it doesn't get hot, the air inside the case won't get hot. Every case will have SOME airflow. Better cases have better airflow. So air escapes. And if the heat sink doesn't get hot, that air escaping won't get hot. So the temperature of the air leaving the case IS indicative of good design and cooling.

Your mind is made up; don't confuse you with facts.

Support the troops: Bring them home ASAP.

Message #53 - Posted 2007/05/24 - The New Guy

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

The fan speed increases based on the temperature, which keeps the component temperature within spec, dummy.

But the quality of the heat sink dictates its ability to cool which controls its temperature, smartypants. :)

Message #54 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Maybe the problem here is that some of you are viewing the case as a sealed enclosure where every bit of heat generated is retained.

No, we're not. We're viewing it as a properly designed container that keeps its interior's temperature within specifications.

The point is that to make the exiting air cooler, you have to move more air in the same time frame, or move the same amount of air in a shorter time frame. That means you need larger vents or a faster fan.

However, and this is the point that you either refuse to understand or are incapable of understanding, getting the exiting air cooler is not necessary; all that is necessary is to keep the interior temperature within specifications. And the iMac does just that.

In short, you do not know what you are talking about again. That in and of itself is not a bad thing; what makes it a bad thing is that you refuse to learn anything.

Support the troops: Bring them home ASAP.

Message #55 - Posted 2007/05/24 - Michelle Steiner

Previously, Jolly Roger wrote:

His flawed assumption is that Macs are built like most PCs.

His flawed assumption is that he knows something.

Support the troops: Bring them home ASAP.

Message #56 - Posted 2007/05/24 - RubyTuesday

Previously, The New Guy wrote:

And that is a wonderful concept, except in the execution of that premise, the buyer is sorely removed from the wonderful world of "choice". Unless you consider the ability to put another stick of ram in "choice". Its a very limiting design. My point was that it didn't have to be. My point was that with better design, Apple could have had a product that appeals not only to the person too lazy and complacent to demand anything better than mediocrity, but to people with a heavy focus in one particular area. And that encompasses a lot of people, whether they want performance in general speed, video or audio, or the size of the monitor. I talk to so many people that wouldn't consider an iMac because of its lack of choice. My point was that it didn't need to be that way. And in the future, maybe they will open their eyes. It could be a product that appeals to the most basic of user as well as the most specific of user.

So people who don't have a heavy focus in one particular area are lazy and complacent? People who want performance in general speed, video or audio, aren't getting performance in speed, video or audio, with an iMac? If someone has an iMac with a 24" built-in monitor he's too lazy and complacent to demand anything bettor?

Ruby

"Rarely is the question asked: is our children learning?" -- George W. Bush

Message #57 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 16:31:38 -0500, The New Guy said:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

The fan speed increases based on the temperature, which keeps the component temperature within spec, dummy.

But the quality of the heat sink dictates its ability to cool which controls its temperature, smartypants. :)

Agreed. And your point is?

JR

Message #58 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Wow, you're still a complete MORON. Heat doesn't magically disappear, it just gets moved around.

I guess in a poor design it will just move around and around inside the case.

And the iMac doesn't do that, as you have told us yourself.

Support the troops: Bring them home ASAP.

Message #59 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Now for my next rant......:)

Please spare us any more evidence of your abysmal ignorance.

Support the troops: Bring them home ASAP.

Message #60 - Posted 2007/05/24 - The New Guy

But Googling, I did find someone who found
an externally mounted fan lowered temperatures of both the CPU and hard drive significantly:
http://www.legitreviews.com/article.php?aid=227 This link is all over the place so its being favorably talked about a lot.

Oops. The author clearly states that he doesn't know if his mod voids the warranty:
"It is unknown if this will void your warranty or not, so if you do this on your own please keep this in mind."

Yes - but is the warranty really that long? Of course, it would be prudent to wait til the warranty expired. If you buy an extended warranty, that might be a harder decision to make.

Going back to your earlier comment:

I wonder if anyone has run their iMac without the back cover to see if it runs significantly cooler?

This can be a dangerous thing to do as it could create spots where air no longer circulates at all.

Well - you could always just feel the components to see how they are running. Its not difficult. If the back cover has ducting attached to it, then its removal might very well increase temperatures. Does it? If not, it should lower them. But I'm just guessing here. We need someone to venture forth into that uncharted netherworld of Mac disassembly.......shudder. Jolly, are you up for it? You certainly did a fine job of temperature logging. Though I wonder why others have just different results. Maybe there's are faulty.

Message #61 - Posted 2007/05/24 - The New Guy

If something is being cooled well, the air coming off it is not hot.

It is being cooled by the air heating up and carrying the heat away.

True. But if a large enough heat sink is used, the heat sink itself won't get hot. Ideally it should get warm and have enough airflow around it to dissipate that warmth. But if the heat sink is too small it will get hot thereby requiring more airflow not to endanger the device being cooled.

Is this so difficult to comprehend?

Again, you are making the flawed assumption that air in Macs is stagnant and that stagnant air is absorbing the heat from the sink, increasing the ambient air temperature inside the case. This is how most PCs today are designed. In such an environment, what you say is true, because the heat-absorbing capacity of the ambient air has limits, so the smaller the heat sink, the less heat will transfer to the ambient air around it. But that is *not* how Macs are designed today. Today's Macs have custom compartments and channels that direct cool incoming air directly over components, through heat sinks, and out of the case. In this type of environment, heat sink size does not directly correspond to the temperature of exiting air.

Sadly, if that were true, the temperature of the exiting air would never be hot. And certainly never hot when idling.

Its usually because of the heat sink size.

The heat sink merely stores the heat away from a component--the CPU and/or memory in this case. The heat sink can hold only so much heat, just like a kitchen sink can hold only so much water.

Exactly. And the amount of heat that heat sink can hold is proportional to its heat absorbing size. If you put a 10' x 10' x 10' heat sink on your little cpu, do you think it would get hot?

Whether it would get hot depends completely on AIR FLOW, as I have said so many times I am growing tired of typing it.

We all know that all computer cases have airflow. Quality and design of the heat sink dictates how much airflow is needed to cool the component. Its the heat sink that is important.

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

By variable you mean that the fan goes faster (louder, noisier) when it needs to. In a good design, you never hear the fan, even under load because the heat sink is cooling so well you don't need tons of airflow to keep it cool.

Message #62 - Posted 2007/05/24 - The New Guy

You're forgetting that every case has
some airflow. Better cases have better airflow. So as soon as the slightest increase of temperature occurs inside the case, that heated air rises and is drawn out of the case.

LOL. A good cooling system doesn't allow heat to rise to begin with - it expels the heated air ASAP - this is exactly what today's Macs do.

If that were true, expelled air off those idling iMacs in the store I was at wouldn't have been so hot. Even in my Mini the exiting air is almost never hot. I have it positioned vertically, near the coolest area of the room (the floor) without the cover on, wih all cooling ducting in place. When its stressed, then it gets quite warm. But when its idling, never.

Message #63 - Posted 2007/05/24 - The New Guy

Greg, in a well designed case, heat is expelled much faster than in a poorly designed case.

But that is completely orthogonal to "If something is being cooled well, the air coming off it is not hot." Anything that gets hot will create hot air. It's BASIC THERMODYNAMICS.

Yes - but airflow moves that air so it doesn't raise the interior temperature. You're thinking that the case is sealed. Air is moving all the time. Its being expelled all the time. Now my point was that a better designed heat sink requires less airflow for the same degree of cooling, than a poorly designed one. IF you have more radiating area on the heat sink, you simply don't need as much airflow to achieve the same degree of cooling.

Message #64 - Posted 2007/05/24 - The New Guy

Previously, RubyTuesday wrote:

Previously, The New Guy wrote:

And that is a wonderful concept, except in the execution of that premise, the buyer is sorely removed from the wonderful world of "choice". Unless you consider the ability to put another stick of ram in "choice". Its a very limiting design. My point was that it didn't have to be. My point was that with better design, Apple could have had a product that appeals not only to the person too lazy and complacent to demand anything better than mediocrity, but to people with a heavy focus in one particular area. And that encompasses a lot of people, whether they want performance in general speed, video or audio, or the size of the monitor. I talk to so many people that wouldn't consider an iMac because of its lack of choice. My point was that it didn't need to be that way. And in the future, maybe they will open their eyes. It could be a product that appeals to the most basic of user as well as the most specific of user.

So people who don't have a heavy focus in one particular area are lazy and complacent?

Those were a poor choice of words. We're all complacent in areas of our lives that don't really interest us much. For some, a computer is just a tool to communicate and learn from. Nothing more. Sorry about that.

Message #65 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 16:23:14 -0500, The New Guy said:

Do most of you think that the iMac would be far less popular if the case was just a couple of inches thicker?

Abso-fucking-lutely it would be less popular. Mac users wouldn't settle for yet another one of these ugly-ass beasts sitting on their desks:

<http://gizmodo.com/assets/2006/06/allinonepc.jpg> <http://www.semicron.com/jiva-pl.gif> <http://www.theregister.co.uk/media/1266.jpg> <http://www.theregister.co.uk/media/1267.jpg>

And, hey - why stop at a 4-inch thick iMac? I mean hell, go all the way and make it 7 inches, and you've got yourself a standard PC case turned sideways. Wooo! *drool* Sexy! </sarcasm>

JR

Message #66 - Posted 2007/05/24 - The New Guy

Previously, Jolly Roger wrote:

On 2007-05-24 16:31:38 -0500, The New Guy said:

When people make sweeping statements, they really don't target anything much. Nor are they useful. Be specific.

Um, the very fact that there's no overheating problems with G5 iMacs tell us there's no "lousy cooling." If you want to demonstrate your point, show us where your research has shown there's an actual problem.

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

The fan speed increases based on the temperature, which keeps the component temperature within spec, dummy.

But the quality of the heat sink dictates its ability to cool which controls its temperature, smartypants. :)

Agreed. And your point is?

My point was that if one uses a well designed heat sink, the airflow doesn't have to be so high, the noise is minimal and the exiting air, of course will never be hot.

Message #67 - Posted 2007/05/24 - Tim Streater

Previously, The New Guy wrote:

Greg, in a well designed case, heat is expelled much faster than in a poorly designed case.

But that is completely orthogonal to "If something is being cooled well, the air coming off it is not hot." Anything that gets hot will
create hot air. It's BASIC THERMODYNAMICS.

Yes - but airflow moves that air so it doesn't raise the interior temperature. You're thinking that the case is sealed. Air is moving all the time. Its being expelled all the time. Now my point was that a better designed heat sink requires less airflow for the same degree of cooling, than a poorly designed one. IF you have more radiating area on the heat sink, you simply don't need as much airflow to achieve the same degree of cooling.

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Message #68 - Posted 2007/05/24 - The New Guy

Do most of you think that the iMac would be far less popular if the case was just a couple of inches thicker?

Absolutely it would be less popular. Mac users wouldn't settle for yet another one of these ugly-ass beasts sitting on their desks: <http://gizmodo.com/assets/2006/06/allinonepc.jpg> <http://www.semicron.com/jiva-pl.gif> <http://www.theregister.co.uk/media/1266.jpg> <http://www.theregister.co.uk/media/1267.jpg>

Looking at those, I realize that the base adds or takes away a lot from the design. I'm into function so height adjustability is crucial but putting height adjustability into a object that weighs that much would be difficult. Still having a sleek base really doesn't detract from strength and looks much better. But if you're really into style, a monitor arm is surely the way to go. And they have them for iMacs now. A little pricey of course. But it weighs more than a monitor so thats to be expected.

And, hey - why stop at a 4-inch thick iMac? I mean hell, go all the way and make it 7 inches, and you've got yourself a standard PC case turned sideways. Wooo! *drool* Sexy! </sarcasm>

Well you haven't lost your sense of humor, that's for sure! :)

Message #69 - Posted 2007/05/24 - The New Guy

Previously, Tim Streater wrote:

Previously, The New Guy wrote:

Greg, in a well designed case, heat is expelled much faster than in a poorly designed case.

But that is completely orthogonal to "If something is being cooled well, the air coming off it is not hot." Anything that gets hot will
create hot air. It's BASIC THERMODYNAMICS.

Yes - but airflow moves that air so it doesn't raise the interior temperature. You're thinking that the case is sealed. Air is moving all the time. Its being expelled all the time. Now my point was that a better designed heat sink requires less airflow for the same degree of cooling, than a poorly designed one. IF you have more radiating area on the heat sink, you simply don't need as much airflow to achieve the same degree of cooling.

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling. Until they find a cheap alternative for copper and aluminum, manufacturers will have challenges.

Message #70 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 16:11:54 -0500, The New Guy said:

Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise.

While this is common in Windows PCs, it's not common in Macs.

What's the difference between a Windows machine and a Mac these days? Not the CPU, hard drives, optical drives. Just the motherboard.

Ummm... you forgot the most important differentiator with respect to the topic at-hand: the case design and cooling system.

The CPU in both is the primary heat generator. That, along with the video card and hard drives generate most of the heat in both machines. Look at the large heat sinks in a Mac Pro. Then look at a Scythe Ninja heat sink. They look very close. Maybe it is a Scythe Ninja.....I don't know. They're both computers with the same design challenges.

And yet the Scythe Ninja, more often than not, sits in your average, ordinary PC case where air inside the case remains largely stagnant resulting in heat radiation and ambient temperature rise internally - compared to a Mac case where air is carefully, thoughtfully, and successfully directed and tunneled so that it passes efficiently over components and heat sinks before being expelled - the heat sinks in Macs get fed a steady stream of air at all times.

In a good design, the heat sink is big enough

that it doesn't get hot because the heat is spread over a larger area.

It doesn't matter how large a heat sink is, or hot a heat sink gets, as long as the air flow is substantial enough to transfer heat off the heat sink and out of the case.

But you can only get lots of airflow with high rpm fans that produce lots of noise.

And yet we don't hear masses of Mac users complaining that fan noise is too loud. Go figure. Are you suggesting Mac users never push their computers so the fans always run at their slowest speed?

Using a well designed heat sink doesn't require such a large fan. Remember there is always airflow in a computer case, whether its a Mac or PC.

There is not nearly as much air flow in a standard PC case, since it's just one large open compartment where air is relatively static when compared to air in Macs.

Now here we have a problem: The depth of
the iMac case. Remember in my original post I mentioned that if only they had made the case a couple of inches deeper, they could have used 120 mm or larger fans running at 5 volts.

You seem to think big expanses of empty space inside of a computer is a good thing, when in fact, it usually leads to slower air flow.

Those produce reasonable
amounts of airflow and are inaudible. But they are only effective if the heat sink is good. That means on the large side.

You can get effective cooling with a small heat sink. This is not up for debate.

With a proper heat sink, the air coming off that device will never be hot.

Air flow is what transfers heat from sinks. Increasing heat sink size exposes more of the sink to more air. If the air doesn't move, ambient temperature rises, and eventually, you will be unable to transfer heat off the sink. It's air flow that matters.

Agreed. And all cases have airflow. Some of you are thinking that we're dealing with heat generation in a sealed box with no airflow.

No, but PCs in general use poorly-ventilated cases compared to Macs.

Put a heat sink on any component with zero air flow in a confined area, and if the ambient temperature gets too high, you're toast.

But that's a sealed box - and that's the mindset some of you are in. A computer case should be the opposite of a sealed box. It should expel heated air as quickly as possible so it doesn't build up. If it does, that expelled air will not be hot. It will be warm at best.

And that's one of the things that sets Macs apart from the majority of the PC industry - rather than relying largely on passive or ambient cooling, they push air through targeted pathways to transfer heat away from components in a quick, efficient manner.

So the air coming out of the case will never be hot. That's good design.

You seem to think the heat just disappears inside the case!

No, a good design expels that heat immediately so the interior temperature hardly rises at all from the ambient room temperature. You seem to think the computer case is sealed.

Expels it *where*? If not straight out the back of the case, then *where*? You keep saying the exhausted air shouldn't be hot - but if the air isn't hot, where's the heat really going? Don't try to suggest the components somehow generate less heat - components generate the same amount of heat in a Mac as they do in any PC - these are mostly the same components, remember. Think about it. The heat must go *somewhere* - if not as much heat is expelled out of the case, then it's an indication the heat is staying inside the case, which means the cooling system is not as efficient.

Reality is
the heat sink is not transferring enough heat from the component into the air, so the heat is likely increasing ambient inner-case temperature, and worse, increasing the temperature of surfaces and other components inside the case.

Yes - in a bad design, that is exactly what is happening. And that is exactly what happens when you feel hot air being expelled from a case design like the iMac!

Nope. The inside of an iMac isn't a big empty chamber, so the air you are feeling coming out the vent is traveling along a sealed path directly over and through heat sinks before being expelled. It's not like your ordinary PC case where the air largely sits in a convection current in a big empty chamber waiting to eventually be exhausted out the back - that's poor circulation and the sign of a badly-designed cooling system.

And when its happening when the unit is idling, its further proof that things are very wrong inside that box.

Nope. Wrong. Simply wrong.

Hopefully Michelle and her iMac enjoy air conditioning on High this summer. An interesting solution would be to vent air from an air conditioner directly to the input vent(s) of your computer. If moisture wasn't a problem, that would be great. Its kind of silly trying to cool something with hot, interior case air.

The iMac has no large expanse of internal case air. You've so obviously never seen one's innards.

Style takes paramount importance at
Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

You clearly don't understand Apple's target market.

My original post was that they could have adjusted the style, by not much,

2 inches is a lot when it comes to style.

and got most of the present target market, as well as tons of PC users that are tired of a very flawed OS and unending virus infection. After all, at about 5% of the market, their market penetration depends on wooing PC users over.

Guess what - Apple's getting PC users anyway lately. And Apple's never really been *that* interested in gaining market share. Gaining market share is not their primary objective by any stretch of the imagination, which is actually why they are so successful.

Linux users have found a virus free, stable
OS that runs far more efficiently than Windows on cheap hardware, so they're certainly not coming over anytime soon.

Bullshit. I'm in the tech industry, and I work for one of the largest semiconductor manufacturers on the planet. I know plenty of Linux users who have purchased Macs lately, and they love them. I see highly technical Linux, Windows, Solaris users get new Macs daily. It's an undeniable trend.

Once Apple got their
hardware and pricing on a serious level, PC users would start coming to Apple outlets to buy their computers.

Like I said, PC users are already buying Macs.

Once there, they could be
shown the superiority of the Apple OS. THEN you have a convert. Very few people are converted without seeing and feeling the difference. I look at potential.

Windows users assume the Mac experience can't be much better than their current Windows experience. Apple making uglier-yet-more-expandable iMacs won't change that.

JR

Message #71 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

My point was that if one uses a well designed heat sink, the airflow doesn't have to be so high, the noise is minimal and the exiting air, of course will never be hot.

Apple has a well designed system, which should be the goal. Instead of obsessing over one component, it is better to focus on the end result.

Support the troops: Bring them home ASAP.

Message #72 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 16:41:39 -0500, The New Guy said:

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

By variable you mean that the fan goes faster (louder, noisier) when it needs to. In a good design, you never hear the fan, even under load because the heat sink is cooling so well you don't need tons of airflow to keep it cool.

Show me such a computer please.

JR

Message #73 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

If the heat sink radiates, all it does is heat up other components and/or the case. Where do you think that the radiated heat goes?

Yet again, you demonstrate that you do not know what you're talking about.

Support the troops: Bring them home ASAP.

Message #74 - Posted 2007/05/24 - Daniel Packman

Previously, The New Guy wrote:

If a computer is expelling hot air when its idling, what's the interior temperature going to be when its stressed and working hard?

There is no way for you to know what the interior temperature is from this information. If there is lots of cool air being expelled, the cpu could be out of the air stream and melting. You are making a poor assumption in tying temperature of expelled air to interior temperature.

But the expelled air IS the interior air being vented. So its one and the same.

All you know from the air being vented is the heat flow (proportional to temperature rise and volume of air per unit time). In a steady state, this corresponds to the heat dissipation inside the case. There is no substitute for measuring the temperature directly on the components of interest.

Message #75 - Posted 2007/05/25 - Bjarne Bäckström

The New Guy wrote:

Previously, Tim Streater wrote:

[...]

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

True -- and the outgoing air gets hotter. To keep the cooled object at the same temperature, you have to transport away the same amount of power (heat.) Power is in this case represented by the temperature gradient (difference between incoming and outgoing) times the mass flow (m**3 per hour) of the air.

It's your choice: If you want lower air mass flow, then you have to make the heat sink more efficient, giving a higher outgoing air temperature, and vice versa. Simple as that, really.
--

Message #76 - Posted 2007/05/24 - The New Guy

Its usually because of the heat sink size. If you put a typical CPU with a standard size heat sink that heat sink will get hot. Its usually paired with some cheap, tiny fan, whirring away at high speed making tons of noise.

While this is common in Windows PCs, it's not common in Macs.

What's the difference between a Windows machine and a Mac these days? Not the CPU, hard drives, optical drives. Just the motherboard.

Ummm... you forgot the most important differentiator with respect to the topic at-hand: the case design and cooling system.

The CPU in both is the primary heat generator. That, along with the video card and hard drives generate most of the heat in both machines. Look at the large heat sinks in a Mac Pro. Then look at a Scythe Ninja heat sink. They look very close. Maybe it is a Scythe Ninja.....I don't know. They're both computers with the same design challenges.

And yet the Scythe Ninja, more often than not, sits in your average, ordinary PC case where air inside the case remains largely stagnant resulting in heat radiation and ambient temperature rise internally - compared to a Mac case where air is carefully, thoughtfully, and successfully directed and tunneled so that it passes efficiently over components and heat sinks before being expelled - the heat sinks in Macs get fed a steady stream of air at all times.

Do you know the Scythe Ninja? Its one of the most expensive non-water cooled systems out there. You'll never see it on any mainstream system. Most computer stores won't touch it because its too expensive. If the air is stagnant, its because of bad fan placement. That's all. This is pretty simple really. Though you'd never know it by looking at most case designs.

Some of you people keep on harping on the wonderful design of the iMac. Well if it was so wonderful the air exiting it would not be hot. But I'm kind of beating a dead horse here and I'm sure most of you are getting tired of this.

In a good design, the heat sink is big enough

that it doesn't get hot because the heat is spread over a larger area.

It doesn't matter how large a heat sink is, or hot a heat sink gets, as long as the air flow is substantial enough to transfer heat off the heat sink and out of the case.

But you can only get lots of airflow with high rpm fans that produce lots of noise.

And yet we don't hear masses of Mac users complaining that fan noise is too loud. Go figure. Are you suggesting Mac users never push their computers so the fans always run at their slowest speed?

I've read a lot of complaints of precisely that. Now whether there was something wrong with the heat sensor in their systems or not, I can't be sure.

And of course - the system needs to adjust to the user's demands.

Using a well designed heat sink doesn't require such a large fan. Remember there is always airflow in a computer case, whether its a Mac or PC.

There is not nearly as much air flow in a standard PC case, since it's just one large open compartment where air is relatively static when compared to air in Macs.

But that's a lousy case design. Most modern cases do have separate air channels - at least the better designed ones do. The Antec P180 is an example of a step in the right direction here. But this is not a PC vs Mac discussion. And I don't want to sound like I'm attacking Apple nor defending the PC camp. It was originally posted to point out iMac problems in hoping that they will be corrected in future designs.

Now here we have a problem: The depth of
the iMac case. Remember in my original post I mentioned that if only they had made the case a couple of inches deeper, they could have used 120 mm or larger fans running at 5 volts.

You seem to think big expanses of empty space inside of a computer is a good thing, when in fact, it usually leads to slower air flow.

Well I could have expressed that point better. But the original post seemed to be getting a little long winded so I shortened things up. I opted for a 4" deep case or whatever would be the absolute minimum height for a PCI/PCI-E card simply because that would be the approximate height of a 120 mm fan. 4" isn't particularly spacious! But with fans pulling the heat out in vertical channels (hard drives on one air channel, video card and CPU on another) I think the airflow would be great and effortless.

Those produce reasonable
amounts of airflow and are inaudible. But they are only effective if the heat sink is good. That means on the large side.

You can get effective cooling with a small heat sink. This is not up for debate.

OK - name one highly rated small heat sink used by performance enthusiasts. I only wish it was true. Maybe its out there but every time I look at performance heat sinks they get bigger and bigger. Some of the newer ones won't even fit in some moderately wide cases which just makes life more difficult.

With a proper heat sink, the air coming off that device will never be hot.

Air flow is what transfers heat from sinks. Increasing heat sink size exposes more of the sink to more air. If the air doesn't move, ambient temperature rises, and eventually, you will be unable to transfer heat off the sink. It's air flow that matters.

Agreed. And all cases have airflow. Some of you are thinking that we're dealing with heat generation in a sealed box with no airflow.

No, but PCs in general use poorly-ventilated cases compared to Macs.

Let's not have a PC vs Mac thing happening. I respect both and see huge areas of improvement needed in both.

Put a heat sink on any component with zero air flow in a confined area, and if the ambient temperature gets too high, you're toast.

But that's a sealed box - and that's the mindset some of you are in. A computer case should be the opposite of a sealed box. It should expel heated air as quickly as possible so it doesn't build up. If it does, that expelled air will not be hot. It will be warm at best.

And that's one of the things that sets Macs apart from the majority of the PC industry - rather than relying largely on passive or ambient cooling, they push air through targeted pathways to transfer heat away from components in a quick, efficient manner.

You're comparing Macs with lousy PC cases. And that's fair in a way because most PC cases are thoughtless, cookiecutter designs. But I'm just looking at design, free of any software/hardware restrictions.

So the air coming out of the case will never be hot. That's good design.

You seem to think the heat just disappears inside the case!

No, a good design expels that heat immediately so the interior temperature hardly rises at all from the ambient room temperature. You seem to think the computer case is sealed.

Expels it *where*? If not straight out the back of the case, then *where*? You keep saying the exhausted air shouldn't be hot - but if the air isn't hot, where's the heat really going?

I wonder if this was not understood by others as well. Sorry about that. I could have been clearer in this area. I meant that as soon as the heat is generated its sucked or ducted out of the case as soon as possible. For the Antec P180, they put the power supply on the bottom so the CPU heatsink is at the top. That heat is sucked out the top through a top mounted fan blowing up. Many people use ducting which works very well too. Apple uses channels in their Mac Pro and that works in the same principle as well. I'd love to find some sort of flexible material I could bend and install in a case so I could route the hot air out wherever I want. Some people use PVC I think but its not very flexible. Perhaps some sort of very thin walled PVC?

Don't try to suggest
the components somehow generate less heat - components generate the same amount of heat in a Mac as they do in any PC - these are mostly the same components, remember. Think about it. The heat must go *somewhere* - if not as much heat is expelled out of the case, then it's an indication the heat is staying inside the case, which means the cooling system is not as efficient.

No argument there!

Reality is
the heat sink is not transferring enough heat from the component into the air, so the heat is likely increasing ambient inner-case temperature, and worse, increasing the temperature of surfaces and other components inside the case.

Yes - in a bad design, that is exactly what is happening. And that is exactly what happens when you feel hot air being expelled from a case design like the iMac!

Nope. The inside of an iMac isn't a big empty chamber, so the air you are feeling coming out the vent is traveling along a sealed path directly over and through heat sinks before being expelled. It's not like your ordinary PC case where the air largely sits in a convection current in a big empty chamber waiting to eventually be exhausted out the back - that's poor circulation and the sign of a badly-designed cooling system.

We're back to the PC vs Mac thing again.....:)
The Antec P180 and the Mac Pro both address cooling in new and better ways. Hopefully this will catch on to their other systems.

Hopefully Michelle and her iMac enjoy air conditioning on High this summer. An interesting solution would be to vent air from an air conditioner directly to the input vent(s) of your computer. If moisture wasn't a problem, that would be great. Its kind of silly trying to cool something with hot, interior case air.

The iMac has no large expanse of internal case air. You've so obviously never seen one's innards.

Yes - I realize its pretty packed in there. But if the air was being vented well, the expelled air would be cool, which it wasn't on the ones I inspected. Maybe yours is the exception. Maybe all 3 of those in a store were faulty. But I doubt it. On a side note they had a 30" Apple monitor there and that was running pretty hot as well. There's almost no ventilation holes in the case. I wonder why they don't opt for more holes? That's a very expensive monitor which people could use for many years to come if it ran cool. Or maybe the components inside it are less prone to heat sensitivity than hard drives and the other components in a computer. It just seems so easy to have more ventilation.

Style takes paramount importance at
Apple. Its such a shame. Such a waste of an OS. As I said, if they had opted for a design that was not quite so sleek, (a little deeper by a couple of inches) they could have had a world of possibilities. Instead function lost to style - again.

You clearly don't understand Apple's target market.

My original post was that they could have adjusted the style, by not much,

2 inches is a lot when it comes to style.

Yes, I think you're definitely right there.

and got most of the present target market, as well as tons of PC users that are tired of a very flawed OS and unending virus infection. After all, at about 5% of the market, their market penetration depends on wooing PC users over.

Guess what - Apple's getting PC users anyway lately. And Apple's never really been *that* interested in gaining market share. Gaining market share is not their primary objective by any stretch of the imagination, which is actually why they are so successful.

Well the shareholders would sure jump for joy if they got a 2 digit market share. When you sell a DVD for $129 and it costs pennies to produce (once R&D costs are paid0 you start making the kind of money that Microsoft makes. But that's a different topic.

Linux users have found a virus free, stable
OS that runs far more efficiently than Windows on cheap hardware, so they're certainly not coming over anytime soon.

I'm in the tech industry, and I work for one of the largest semiconductor manufacturers on the planet. I know plenty of Linux users who have purchased Macs lately, and they love them. I see highly technical Linux, Windows, Solaris users get new Macs daily. It's an undeniable trend.

Why do you think most of the LInux users are coming over to OS X? Software choice? Ease of use? Both Linux and OS X are stable and virus free. What's the attraction?

Once Apple got their
hardware and pricing on a serious level, PC users would start coming to Apple outlets to buy their computers.

Like I said, PC users are already buying Macs.

I meant a lot of them......:)

Once there, they could be
shown the superiority of the Apple OS. THEN you have a convert. Very few people are converted without seeing and feeling the difference. I look at potential.

Windows users assume the Mac experience can't be much better than their current Windows experience. Apple making uglier-yet-more-expandable iMacs won't change that.

Most Windows users that try my system like it. I think that most of them think that "If its this easy right away, imagine what I can do if I really apply myself!". The challenge is how to get them to try it in the first place.

Message #77 - Posted 2007/05/24 - The New Guy

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all
in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

By variable you mean that the fan goes faster (louder, noisier) when it needs to. In a good design, you never hear the fan, even under load because the heat sink is cooling so well you don't need tons of airflow to keep it cool.

Show me such a computer please.

Most anything using an Antec P180 or other smartly designed case.

Actually its simple. Put the power supply outside the box with the fan at 5 volts pointing in instead of pointing out. Positive air cools better than negative air apparently. Now the power supply is silent.

Then pick a large heat sink for your CPU with a large fan. Then mount a fan right beside it pulling the hot air out. Unless you're doing something crazy with high end video that should do it. Just balance the positive and negative airflow using large fans running at 5 or 7 volts.

If you're in a really hot environment, probably water cooling would be best. Does anyone know if humidity makes any difference to cooling? We certainly feel hotter when the humidity rises on warm days but I wonder if it makes a difference to electronic equipment?

Message #78 - Posted 2007/05/24 - The New Guy

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

If the heat sink radiates, all it does is heat up other components and/or the case.

An intelligent design expels the heated air right away. It doesn't heat up anything at all. Its not a sealed box. There is airflow. A really good design would use separate air ducts to cool the CPU, video card and hard drives. Then more ducts to remove that warmed air from each and immediately expel it. Ironically perhaps the easiest design to attain this would be one where the air flows vertically, and there is a separate channel for all 3 devices. The only problem area might be the video card but all that entails is for the case design to accommodate that. A perfect shape for that case might be the same dimensions as a large, widescreen monitor. Oh the irony.

Message #79 - Posted 2007/05/24 - The New Guy

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

True -- and the outgoing air gets hotter.

Actually the heat sink just gets warm, doesn't even get hot. So that increased temperature is far more easily expelled quickly.

To keep the cooled object
at the same temperature, you have to transport away the same amount of power (heat.) Power is in this case represented by the temperature gradient (difference between incoming and outgoing) times the mass flow (m**3 per hour) of the air.

Wow - sounds fancy. If that's the science behind this, fine.

It's your choice: If you want lower air mass flow, then you have to make the heat sink more efficient, giving a higher outgoing air temperature, and vice versa. Simple as that, really.

I'm not sure what you mean by efficient but I would hazard a guess that it means its conducting more heat away from the device without getting as hot as a less efficient one?

Like I keep on saying, if you use a good heat sink you don't need as much airflow. If you don't need as much airflow, you can use lower rpm fans which are inaudible.

The manufacturers don't want to use large heat sinks because they are expensive, because raw materials (aluminum and copper) are expensive.

Message #80 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Like I keep on saying, if you use a good heat sink you don't need as much airflow.

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories. And whether those calories are spread out over a larger volume or concentrated in a smaller one, they are the same number of calories.

Support the troops: Bring them home ASAP.

Message #81 - Posted 2007/05/25 - Bjarne Bäckström

The New Guy wrote:

Its not radiative cooling, its convection cooling. The heat sink heats up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I know nothing about the iMac, not talking about that), or a fan might be required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

True -- and the outgoing air gets hotter.

Actually the heat sink just gets warm, doesn't even get hot. So that increased temperature is far more easily expelled quickly.

Sorry, I don't follow. You say that the heat sink doesn't even get hot, and then you say that the increased (!) temperature "is far more easily expelled quickly," and above you say that "you need less airflow." You can't fool Mother Nature that easily. See below, please.

To keep the cooled object
at the same temperature, you have to transport away the same amount of power (heat.) Power is in this case represented by the temperature gradient (difference between incoming and outgoing) times the mass flow (m**3 per hour) of the air.

--
Message #82 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 20:52:43 -0500, The New Guy said:

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

By variable you mean that the fan goes faster (louder, noisier) when it needs to. In a good design, you never hear the fan, even under load because the heat sink is cooling so well you don't need tons of airflow to keep it cool.

Show me such a computer please.

Most anything using an Antec P180 or other smartly designed case.

So the Antec case has only *two* separate air cavities (from bottom to top):

<http://www.silentpcreview.com/files/images/p180/p180open.jpg>

1. the lower cavity where air is pulled across lower hard drives, through the power supply, and out the computer
2. the upper cavity where upper CD/DVD drives, hard drives, video/PCI cards, and RAM & CPU all share the same air

* a small exception is a strange air duct contraption that tries to feed air from the *back* of the case (where hot air is being exhausted, mind you) to the video card

The Mac G5 and Pro towers from Apple have four separate air cavities (from bottom to top):

<http://jollyroger.kicks-ass.org/jollyroger/G5-cooling.jpg>

1. the power supply
2. the RAM & CPU
3. the video/PCI cards (being fed cool air from the *front* of the case) 4. the CD/DVD and hard drives

Are you seriously suggesting the Antec case is somehow a better design?

JR

Message #83 - Posted 2007/05/24 - Michelle Steiner

Previously, The New Guy wrote:

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

If the heat sink radiates, all it does is heat up other components and/or the case.

An intelligent design expels the heated air right away. It doesn't heat up anything at all.

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

A really good design would use separate air ducts to cool the CPU, video card and hard drives. Then more ducts to remove that warmed air from each and immediately expel it.

You just described what the iMac does.

Support the troops: Bring them home ASAP.

Message #84 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 20:47:02 -0500, The New Guy said:

There is not nearly as much air flow in a standard PC case, since it's just one large open compartment where air is relatively static when compared to air in Macs.

But that's a lousy case design. Most modern cases do have separate air channels

No, no... most modern PC cases certainly do not have separate air channels.

- at least the better designed ones do.

Well Most != Better - you'll have to pick one.

The Antec P180
is an example of a step in the right direction here.

The Antec P180 case is nowhere near as good as the Apple G5/Pro case.

JR

Message #85 - Posted 2007/05/24 - Dave Balderstone

Previously, The New Guy wrote:

An intelligent design expels the heated air right away. It doesn't heat up anything at all. Its not a sealed box. There is airflow.

Like, um, say... the iMac case? Where air passes through channels and immediately moves heat out of the case? By exhausting warm air?

Naw. You're much smarter than that.

Message #86 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 20:47:02 -0500, The New Guy said:

I'm in the tech industry, and I work for one of the largest semiconductor manufacturers on the planet. I know plenty of Linux users who have purchased Macs lately, and they love them. I see highly technical Linux, Windows, Solaris users get new Macs daily. It's an undeniable trend.

Why do you think most of the LInux users are coming over to OS X? Software choice? Ease of use? Both Linux and OS X are stable and virus free. What's the attraction?

It's all about the desktop.

JR

Message #87 - Posted 2007/05/24 - Jolly Roger

On 2007-05-24 20:47:02 -0500, The New Guy said:

Once there, they could be
shown the superiority of the Apple OS. THEN you have a convert. Very few people are converted without seeing and feeling the difference. I look at potential.

Windows users assume the Mac experience can't be much better than their current Windows experience. Apple making uglier-yet-more-expandable iMacs won't change that.

Most Windows users that try my system like it. I think that most of them think that "If its this easy right away, imagine what I can do if I really apply myself!". The challenge is how to get them to try it in the first place.

Yep!

JR

Message #88 - Posted 2007/05/24 - The New Guy

Previously, Jolly Roger wrote:

On 2007-05-24 20:52:43 -0500, The New Guy said:

If you put a tiny chipset heat sink on a cpu it gets hot very fast. Its all in the size.

That is true ONLY if air flow is constant, which is not the case in today's Macs. Today's Macs have variable air flow rates controlled by the operating system based on direct component temperature readings.

By variable you mean that the fan goes faster (louder, noisier) when it needs to. In a good design, you never hear the fan, even under load because the heat sink is cooling so well you don't need tons of airflow to keep it cool.

Show me such a computer please.

Most anything using an Antec P180 or other smartly designed case.

So the Antec case has only *two* separate air cavities (from bottom to top):

<http://www.silentpcreview.com/files/images/p180/p180open.jpg>

1. the lower cavity where air is pulled across lower hard drives, through the power supply, and out the computer
2. the upper cavity where upper CD/DVD drives, hard drives, video/PCI cards, and RAM & CPU all share the same air

I haven't used this case or examined it myself. I gave it as an example because its highly rated, though its mostly rated for low noise. I'm not sure if the side panel enables any ducting directly to the CPU heatsink. Even if it didn't one could always cut out a hole for a fan and attach a tube to do just that.

From the Wiki - something I've been lamenting for a while: There is a fairly large performance gap between the Mac Pro and Apple's most powerful consumer machine, the iMac. A particular sore point for many is that the Mac Pro is the only machine in Apple's lineup that can easily change its graphics card; other machines in the lineup use integrated graphics or hard-to-replace expansion cards, while the Mac Pro uses the industry standard PCIe slots. However, buying a workstation platform just to allow for upgradable graphics is something many people reviewers have complained about. This has led to calls for an "xMac"; a smaller machine with more limited expansion capabilities but retaining a single PCIe slot.[7]

I'm getting off track - sorry.

* a small exception is a strange air duct contraption that tries to feed air from the *back* of the case (where hot air is being exhausted, mind you) to the video card

The Mac G5 and Pro towers from Apple have four separate air cavities (from bottom to top):

<http://jollyroger.kicks-ass.org/jollyroger/G5-cooling.jpg>

1. the power supply
2. the RAM & CPU
3. the video/PCI cards (being fed cool air from the *front* of the case) 4. the CD/DVD and hard drives

Are you seriously suggesting the Antec case is somehow a better design?

Nope. This thread is about the iMac, remember? But you got me reading and I was here reading about the hard drive cooling problems in the Mac Pro.
http://www.amug.org/amug-web/html/amug/reviews/articles/intel/macpro/ A solution is just to run the hard drive cooling fan a little higher which is easily fixed. No biggee. What is surprising is that bay 1 and 4 are the hottest whereas 2 and 3 are cooler. Apparently there is no direct air to bay 1.

On a side note, designers never seem to opt for fans at a 45 degree angle, but that might solve a dilemma when you want to use a larger fan but don't have the height or width. Mounting it at an angle will still direct airflow in the direction wanted but would require far less space. How much less airflow would occur would have to be factored in of course. Ideas like that might allow stylish devices like thin iMacs to use larger, quieter fans with better airflow without bastardizing that stylish, sleek case some of you have grown so attached to.

Message #89 - Posted 2007/05/24 - The New Guy

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

If the heat sink radiates, all it does is heat up other components and/or the case.

An intelligent design expels the heated air right away. It doesn't heat up anything at all.

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :)
Right away means heat from one area doesn't go to the other area. I've never seen any case that uses external air to cool either the cpu, video cards, hard drives, ram, etc, then expels that same area immediately. Even the Mac Pro doesn't do that. So everyone is guilty. I've been looking for a diagram for the Mac Pro cooling design but didn't find it. Jolly Roger's G5 diagram was interesting but I'm not sure if its the same for the later design of the Mac Pro. But I'm getting off topic again.....this is about the iMac.

A really good design would use separate air ducts to cool the CPU, video card and hard drives. Then more ducts to remove that warmed air from each and immediately expel it.

You just described what the iMac does.

Can someone post an URL describing the cooling of the iMac please? I just gotta see this. Wouldn't it be ironic if the iMac does it right? Still wouldn't explain the idling hot air being expelled but I still would have a big foot in my mouth! LOL. :)

Message #90 - Posted 2007/05/24 - The New Guy

Its not radiative cooling, its convection cooling. The heat sink heats
up the air, and the hot air then has to leave the area, carrying the heat away with it. Pure convection (no fan) might be able to do this (I
know nothing about the iMac, not talking about that), or a fan might be
required.

Whatever its called, a larger radiating area means you need less airflow to achieve the same cooling.

True -- and the outgoing air gets hotter.

Actually the heat sink just gets warm, doesn't even get hot. So that increased temperature is far more easily expelled quickly.

Sorry, I don't follow. You say that the heat sink doesn't even get hot, and then you say that the increased (!) temperature "is far more easily expelled quickly," and above you say that "you need less airflow." You can't fool Mother Nature that easily. See below, please.

You're thinking that we're dealing in a sealed space I guess. If something doesn't get very warm, the slightest airflow will stop the temperature from rising in the case. If it gets hot, you'll need more airflow. The idea is to need the least amount of airflow to get that cooling job done. The fact that well designed heat sinks are large and have large radiating areas thanks to numerous fins probably factors into the mathematical equation you mention below. All I know is that good designs don't get hot as easily. So they're easier to cool. Some do it without any direct fan at all. But that only happens when they're running a cooler running processor and decent airflow in the case. Most of us won't risk that.

To keep the cooled object
at the same temperature, you have to transport away the same amount of power (heat.) Power is in this case represented by the temperature gradient (difference between incoming and outgoing) times the mass flow (m**3 per hour) of the air.

Message #91 - Posted 2007/05/25 - Mike Rosenberg

The New Guy wrote:

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #92 - Posted 2007/05/25 - Bjarne Bäckström

The New Guy wrote:

[...]

Sorry, I don't follow. You say that the heat sink doesn't even get hot, and then you say that the increased (!) temperature "is far more easily expelled quickly," and above you say that "you need less airflow." You can't fool Mother Nature that easily. See below, please.

You're thinking that we're dealing in a sealed space I guess.

No, I'm thinking "a more efficient heat sink, everything else (air flow, etc.) equal, means that the outgoing air will get hotter." Why? Simply because a more efficient heat sink will transfer more heat from the device to the air. Thus, the device will run cooler.

IIRC, somewhere at the beginning of this thread (I've not read the whole thread) you thought that the air output of some Mac model was too hot, and you suggested a more efficient heat sink; that would allow less air flow, and the outgoing air flow would be cooler. In fact, both of these changes would make the outgoing air hotter.

The fact that well designed heat sinks are large and have large radiating areas thanks to numerous fins probably factors into the mathematical equation you mention below.

Yes, there's nothing magical about it.

All I know is that good designs don't get hot as easily.

Good designs have *adequate* cooling, and they get as hot as the designer permits them to be.
--

Message #93 - Posted 2007/05/25 - Jolly Roger

On 2007-05-24 23:43:12 -0500, The New Guy said:

All I know is that good designs don't get hot as easily.

Well then by your definition judging from the direct component temperature readings in today's Macs, they all are of good design, because the components don't get hot easily due to the way the systems manage the air flow.

JR

Message #94 - Posted 2007/05/25 - Dave Balderstone

Previously, Mike Rosenberg wrote:

The New Guy wrote:

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

Oh, this is too sweet. He's been ranting on and on, and now admits he doesn't have a clue?

I love it.

Message #95 - Posted 2007/05/25 - Bjarne Bäckström

Michelle Steiner wrote:

Previously, The New Guy wrote:

Like I keep on saying, if you use a good heat sink you don't need as much airflow.

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories.

Stricly speaking, this is not entirely correct either, but it's in the right direction. :-) Calories is /energy/, that is power times time (a finite amount of work (power)). "Ongoing" /power/ is what we're talking about.
--

Message #96 - Posted 2007/05/25 - George Kerby

On 5/25/07 9:08 AM, in article 1hyof3g.1973bcq1099300N%bskb@m.a.c.com, Bjarne B‰ckstrˆm wrote:

Michelle Steiner wrote:

Previously, The New Guy wrote:

Like I keep on saying, if you use a good heat sink you don't need as much airflow.

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories.

Stricly speaking, this is not entirely correct either, but it's in the right direction. :-) Calories is /energy/, that is power times time (a finite amount of work (power)). "Ongoing" /power/ is what we're talking about.

"Heat" is atoms rubbing their fannies together...
Message #97 - Posted 2007/05/25 - eyalnevo@gmail.com

On May 23, 8:26 pm, The New Guy wrote:

Or lack of it! I was in a store a while back and was amazed at how little ventilation there was in the G5 iMac cases. Hot air was pouring out of it and the thing was idling; the ambient temperature was just average (cool) room temperature.

Has anyone tried operating the thing with the back removed? Or putting in a large fan to blast cool, exterior air to the warmer areas? Or increasing the exiting ventilation holes?

Give it a rest already. Your hardware design suggestions reek of cluelessness and naivet=E9.

Perhaps you would enlighten us perhaps why they are clueless and naive? Or is that a secret? Usually the point of a discussion is to actually say something. Not blurt out random accusations with no support. Sort of like, he made a point, now I'll address that particular point.

I remember seeing just a slit on the top and thinking, "No wonder this thing runs so hot!".

Its amazing to think that if they had made the cabinet a couple of inches deeper they could have used full height PCI/PCI-E cards (so people that need good video and audio would buy it), and could have used 120 mm (inaudible when running at 5 volts) fans on top and bottom to pull air out of the enclosure so its basically a normal, upgradable machine empowering the buyer and ensuring long term upgradability and scalability. Not to mention enabling Apple to sell a slew of PCI/PCI-E upgrade cards! And multiple hard drives (more sales for Apple) running in Raid 0 for performance. And 4 slots of ram which would have allowed 8 gb (2gb each of 667 or 800 mhz ram) for performance users.

And for total logic the monitor could have been interchangeable so the audio user who doesn't need a 24" screen could save money and go 17", while the graphics nut could be happy with built on audio while maxing out to the largest screen available. Especially considering that most LCD users have empty space behind their monitor so its not being used anyway! To take it a step further, the hard drives would be in their own sound insulated, air cooled column so one could run 2, 3 or even 4 10k or 15k 3gb/sec Sata drives in Raid 0 because a widescreen cabinet gives enough width. In fact, imagine all Apple monitors having mounting rails inside the cabinet. When you want to buy an iMac, you could just bolt on the motherboard/hard drive/optical drive to the monitor. The power supply would be a brick by the wall plug, silent and out of sight. Oh the possibilities when eyes are opened and the designer and management don't have to answer to the time constrains of greedy, impatient shareholders!

for one, what processor exactly can use 8gb of ram?
for two, do us all a favor and buy yourself a good old Pentium 4 with ubuntu so you have something to play with. Mac users dont care about computers, or scalability, or options- they want a thoroughly designed machine that does what they ask it to.

Message #98 - Posted 2007/05/25 - eyalnevo@gmail.com

On May 23, 9:47 pm, The New Guy wrote:

What's an iMac? A computer for non-thinking buyers that enjoy almost zero configurability options and lousy cooling?

Well, actually, it _is_ targeted at buyers who aren't much into customizing and upgrading.

No problem. They still could have more options. That's my point. The premise of an all in one computer is sound. My opinion is that you don't have to block all configurability to achieve that goal.

As for "lousy cooling," aren't you getting tired of making those absurd claims by now?

Aren't you getting tired of spouting off, saying absolutely nothing? If you have disagreements with my points, be specific. Discussion is good. Random comments are useless. Especially on usenet.

back at you. I am rubber and you are glue. what an infantile joke you are.

Message #99 - Posted 2007/05/25 - Bjarne Bäckström

George Kerby wrote:

On 5/25/07 9:08 AM, in article 1hyof3g.1973bcq1099300N%bskb@m.a.c.com, Bjarne B‰ckstrˆm wrote:

Michelle Steiner wrote:

Previously, The New Guy wrote:

Like I keep on saying, if you use a good heat sink you don't need as much airflow.

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories.

Stricly speaking, this is not entirely correct either, but it's in the right direction. :-) Calories is /energy/, that is power times time (a finite amount of work (power)). "Ongoing" /power/ is what we're talking about.

"Heat" is atoms rubbing their fannies together...

Woo-hooo! :-)

Message #100 - Posted 2007/05/25 - The New Guy

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

If something doesn't cool well it has lousy cooling. I go from results. Not theory. If it blows hot when idling, something's wrong.

Message #101 - Posted 2007/05/25 - The New Guy

Previously, "eyalnevo@gmail.com wrote:

On May 23, 8:26 pm, The New Guy wrote:

Or lack of it! I was in a store a while back and was amazed at how little ventilation there was in the G5 iMac cases. Hot air was pouring out of it and the thing was idling; the ambient temperature was just average (cool) room temperature.

Has anyone tried operating the thing with the back removed? Or putting in a large fan to blast cool, exterior air to the warmer areas? Or increasing the exiting ventilation holes?

Give it a rest already. Your hardware design suggestions reek of cluelessness and naivetÈ.

Perhaps you would enlighten us perhaps why they are clueless and naive? Or is that a secret? Usually the point of a discussion is to actually say something. Not blurt out random accusations with no support. Sort of like, he made a point, now I'll address that particular point.

I remember seeing just a slit on the top and thinking, "No wonder this thing runs so hot!".

Its amazing to think that if they had made the cabinet a couple of inches deeper they could have used full height PCI/PCI-E cards (so people that need good video and audio would buy it), and could have used 120 mm (inaudible when running at 5 volts) fans on top and bottom to pull air out of the enclosure so its basically a normal, upgradable machine empowering the buyer and ensuring long term upgradability and scalability. Not to mention enabling Apple to sell a slew of PCI/PCI-E upgrade cards! And multiple hard drives (more sales for Apple) running in Raid 0 for performance. And 4 slots of ram which would have allowed 8 gb (2gb each of 667 or 800 mhz ram) for performance users.

And for total logic the monitor could have been interchangeable so the audio user who doesn't need a 24" screen could save money and go 17", while the graphics nut could be happy with built on audio while maxing out to the largest screen available. Especially considering that most LCD users have empty space behind their monitor so its not being used anyway! To take it a step further, the hard drives would be in their own sound insulated, air cooled column so one could run 2, 3 or even 4 10k or 15k 3gb/sec Sata drives in Raid 0 because a widescreen cabinet gives enough width. In fact, imagine all Apple monitors having mounting rails inside the cabinet. When you want to buy an iMac, you could just bolt on the motherboard/hard drive/optical drive to the monitor. The power supply would be a brick by the wall plug, silent and out of sight. Oh the possibilities when eyes are opened and the designer and management don't have to answer to the time constrains of greedy, impatient shareholders!

for one, what processor exactly can use 8gb of ram?

Well several motherboards can take 8 gb's of ram. If its not useful on some this present day, it sure will be soon. Everybody yearns for more ram. The Intel 965 and 975 chipsets generally do. The Mac Pro might be based on the 975 or a server chipset. Not sure.

Message #102 - Posted 2007/05/25 - The New Guy

Should I have originally posted this to the advocacy newsgroup? Or at least a group more specific? comp.sys.mac.system seems to get the most traffic so I rarely look at the others.

Message #103 - Posted 2007/05/25 - G.T.

The New Guy wrote:

Should I have originally posted this to the advocacy newsgroup?

Yes. You'll fit well with the trolls in advocacy.

Greg

The ticketbastard Tax Tracker:
http://www.ticketmastersucks.org/tracker.html

Dethink to survive - Mclusky

Message #104 - Posted 2007/05/25 - Grandpa

The New Guy wrote:

Should I have originally posted this to the advocacy newsgroup? Or at least a group more specific? comp.sys.mac.system seems to get the most traffic so I rarely look at the others.

Given your penchant for adamantly sticking with your feelings in the face of scientific principles and obvious lack of research I would suggest you post in alt.kooks.

Grandpa

Message #105 - Posted 2007/05/25 - The New Guy

Should I have originally posted this to the advocacy newsgroup? Or at least a group more specific? comp.sys.mac.system seems to get the most traffic so I rarely look at the others.

Given your penchant for adamantly sticking with your feelings in the face of scientific principles and obvious lack of research I would suggest you post in alt.kooks.

Duly noted. :)

Message #106 - Posted 2007/05/25 - Michelle Steiner

Previously, "eyalnevo@gmail.com wrote:

back at you. I am rubber and you are glue. what an infantile joke you are.

Grade: C+ (for correct analysis and sentiment). Try again, but make it rhyme; then you'll get an A, e.g., "I'm rubber; you're glue. What an infantile joke are you."

Support the troops: Bring them home ASAP.

Message #107 - Posted 2007/05/25 - Michelle Steiner

Previously, Bjarne B‰ckstrˆm wrote:

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories.

Stricly speaking, this is not entirely correct either, but it's in the right direction. :-) Calories is /energy/, that is power times time (a finite amount of work (power)). "Ongoing" /power/ is what we're talking about.

Isn't heat energy?

Support the troops: Bring them home ASAP.

Message #108 - Posted 2007/05/25 - Michelle Steiner

Previously, The New Guy wrote:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

I've never seen any case that uses external air to cool either the cpu, video cards, hard drives, ram, etc, then expels that same area immediately.

All cases use external air to cool; the external care comes in through a vent, heats up, and moves out of the case. Some do it more efficiently than others. The more efficient it is, the hotter the air, the greater the volume of the air, and/or the faster it moves.

Wouldn't it be ironic if the iMac does it right?

Not at all.

Support the troops: Bring them home ASAP.

Message #109 - Posted 2007/05/25 - Michelle Steiner

Previously, The New Guy wrote:

Should I have originally posted this to the advocacy newsgroup? Or at least a group more specific?

You shouldn't have posted it at all because it is totally wrong. You should have conducted your research, and then you would have found that you would have had no reason to post it in the first place.

Support the troops: Bring them home ASAP.

Message #110 - Posted 2007/05/25 - Michelle Steiner

Previously, The New Guy wrote:

If something doesn't cool well it has lousy cooling. I go from results. Not theory.

That's nice. Next time, try actually doing that.

If it blows hot when idling, something's wrong.

Wrong.

BTW, my 17" iMac Core Duo has been running nonstop for more than 24 hours. (I had to restart it yesterday when experimenting with Boot Camp.) and the airflow out the vent on the top back is barely warm. I don't see where you get this "hot" from in the first place.

Support the troops: Bring them home ASAP.

Message #111 - Posted 2007/05/25 - Mike Rosenberg

The New Guy wrote:

If something doesn't cool well it has lousy cooling.

Duh!

I go from results. Not theory.

That's actually something I can support under the right circumstances.

If it blows hot when idling, something's wrong.

Those, however, are not the right circumstances. You keep saying something is wrong with this, however that's an entirely unsubstantiated statement. You are _not_ going from results by stating that.

<http://designsbymike.biz/macconsultshop.shtml> Mac-themed T-shirts <http://designsbymike.biz/musings.shtml> Humorous/muckraking T-shirts <http://designsbymike.biz/prius.shtml> Prius shirts & bumper stickers <http://cafepress.com/comedancing> Ballroom dance-themed shirts & gift

Message #112 - Posted 2007/05/25 - Sue Rodgers

Mike Rosenberg wrote:

If it blows hot when idling, something's wrong.

Those, however, are not the right circumstances. You keep saying something is wrong with this, however that's an entirely unsubstantiated statement. You are _not_ going from results by stating that.

Besides, he himself blows hot air while idling, yet he doesn't seem at all concerned about it.

Living is easy with eyes closed, misunderstanding all you see...

Message #113 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 12:02:25 -0500, The New Guy said:

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

If something doesn't cool well it has lousy cooling. I go from results. Not theory. If it blows hot when idling, something's wrong.

No. If the *component temperatures* inside the computer are hot, THEN something's wrong. The temperature and concentration of air coming out is meaningless.

JR

Message #114 - Posted 2007/05/25 - The New Guy

BTW, my 17" iMac Core Duo has been running nonstop for more than 24 hours. (I had to restart it yesterday when experimenting with Boot Camp.) and the airflow out the vent on the top back is barely warm. I don't see where you get this "hot" from in the first place.

Interesting. Unless its actually a lot warmer than you say it is, which I doubt. And Jolly Roger's iMac seemed to run well by his temperatures. What I don't understand is why the stores multiple iMacs were running so hot.

I'm going to go back to that place and check again. Not to drag this out forever, but I owe that to you all. Hopefully real soon.

Message #115 - Posted 2007/05/25 - Bjarne Bäckström

Michelle Steiner wrote:

Previously, Bjarne B‰ckstrˆm wrote:

No matter how many times you say it, you're still wrong. The heat has to get out of that box, and the only way it gets out is by air flow. You seem to think that heat is temperature; heat is not temperature, it is calories.

Stricly speaking, this is not entirely correct either, but it's in the right direction. :-) Calories is /energy/, that is power times time (a finite amount of work (power)). "Ongoing" /power/ is what we're talking about.

Isn't heat energy?

Heat is a "difficult" term, and I'm afraid that I would use some English terms incorrectly, if I tried to sort it out. Instead, I'll point you to <http://en.wikipedia.org/wiki/Heat> . --

Message #116 - Posted 2007/05/25 - Jeffrey Goldberg

On Fri, 25 May 2007, The New Guy wrote:

I go from results. Not theory.

This discussion has inspired me to test.

My wife's 20" G5 iMac has been in a room with ambiant temperature at around 79F, when loaded the CPU hasn't exceeded 131F.

If it blows hot when idling, something's wrong.

No matter how many times you say that, it still won't make it true.

-j

Jeffrey Goldberg http://www.goldmark.org/jeff/ Relativism is the triumph of authority over truth, convention over justice I rarely read top-posted, over-quoting or HTML postings.

Message #117 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 16:42:36 -0500, The New Guy said:

BTW, my 17" iMac Core Duo has been running nonstop for more than 24 hours. (I had to restart it yesterday when experimenting with Boot Camp.) and the airflow out the vent on the top back is barely warm. I don't see where you get this "hot" from in the first place.

Interesting. Unless its actually a lot warmer than you say it is, which I doubt. And Jolly Roger's iMac seemed to run well by his temperatures. What I don't understand is why the stores multiple iMacs were running so hot.

I'm going to go back to that place and check again. Not to drag this out forever, but I owe that to you all. Hopefully real soon.

Checking the *feel* of the air coming out isn't enough. Your hands might be more or less sensitive depending on numerous factors outside of your control, and even ignoring that, it's just not relevant. You need to check the temperature of the actual components. Ifyou can download or get them to install MenuMeters or XResourceGraph, you can see the temperature of the components, then jot down the temperatures to take with you.

JR

Message #118 - Posted 2007/05/25 - The New Guy

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

But even in the Mac Pro the heat from the one hard drives hits the other hard drives. They get very uneven cooling. Bay 1 and 4 are always hottest. The ram are "cooled" by warmed CPU heat sink air though the temperature of the ram is probably not very important as FB-DIMMs have heat sinks. The power supply shares the optical drives so that's fine. In the Mini all the air that cools the CPU heatsink has been heated by everything else. So that's a disaster but they do have severe space constraints in that design. And he user can always pop off the top if you want some ventilation. For me it seems to run cooler vertically, but that's just a guess.

Now in fairness, I should mention that those iMacs I was looking at the store (last winter) must have been Intel machines. But perhaps some of you have later revisions. Here they talk about fan noise/cooling issues that have been improved in later revisions. Scroll down a bit to see the fan placement.
http://www.scienceman.com/pgs/00_imacG5.html Here's a good pic of the cooling.
http://www.scienceman.com/images/00_imac_G5_01.jpg Here it looks like the CPU heat flows over the hard drive but its difficult to know without a diagram.

I've never seen any case that uses external air to cool either the cpu, video cards, hard drives, ram, etc, then expels that same area immediately.

All cases use external air to cool; the external care comes in through a vent, heats up, and moves out of the case. Some do it more efficiently than others. The more efficient it is, the hotter the air, the greater the volume of the air, and/or the faster it moves.

I meant it gets external air and that air goes directly to the item to cool without passing around/over/underneath anything else. After its cooled it goes DIRECTLY out - not passing over/around/underneath anything else. I've been fiddling around a lot with heat sinks and cooling the last few months so though it may sound like I'm an idiot at times, I have applied myself a lot in this area, enough to see the failings in every design I've looked closely at. Incidentally, many aspects of the Mac Pro I thought of as improvements for generic PC cases before the Mac Pro came out. So I had partial encouragement for my ideas which was nice.

Wouldn't it be ironic if the iMac does it right?

Not at all.

Well its clear that nobody does it "right". A good design uses external air to cool ONE heat generator and immediately expels that heated air. There's not a computer on earth that does that, at least not that I've come across yet. If you know of one, please send the URL. I'd love to see it. The Mac Pro seems to be the best or one of the best so far though. So that's very encouraging, because good ideas work themselves downwards in the product lineup over time for the benefit of all us poorer wretches.

Message #119 - Posted 2007/05/25 - The New Guy

I go from results. Not theory.

That's actually something I can support under the right circumstances.

If it blows hot when idling, something's wrong.

Those, however, are not the right circumstances. You keep saying something is wrong with this, however that's an entirely unsubstantiated statement. You are _not_ going from results by stating that.

Yeah I was. I've worked with computers for several years and torn many apart. I'm always tinkering and fiddling and cooling has been a passion of mine for a while now. After playing with many designs I can positively say that if a device is being cooled well you will NEVER have hot air coming off it when its idling. When its stressed sometimes, maybe, but never when its idling. The heat generators are, in order of importance: CPU, video card, hard drive(s), ram, chipset and power supply (which can easily be external so its not really relevant).

If it blows hot when idling, something's wrong. I'll check in a store again making note of the exact model number. And more than one machine as well. Also I'll check the Activity Monitor to make sure nothing is stressing it chugging away in the background.

Message #120 - Posted 2007/05/25 - The New Guy

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Message #121 - Posted 2007/05/25 - The New Guy

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

If something doesn't cool well it has lousy cooling. I go from results. Not theory. If it blows hot when idling, something's wrong.

No. If the *component temperatures* inside the computer are hot, THEN something's wrong. The temperature and concentration of air coming out is meaningless.

If the component is being cooled well it will never get hot in the first place. And the way to do that is to cool it with external air. And a good heat sink. And a large, low rpm inaudible fan.

Message #122 - Posted 2007/05/25 - Dave Balderstone

Previously, The New Guy wrote:

Can someone post an URL describing the cooling of the iMac please?

Excuse me, but you're the one who's been making the claim that it has "lousy cooling." So all along you've based this on not having the slightest idea of how it works? And now you want someone to tell you what you should have researched and understood before you ever posted in the first place? Yeah, right.

If something doesn't cool well it has lousy cooling. I go from results. Not theory. If it blows hot when idling, something's wrong.

No. If the *component temperatures* inside the computer are hot, THEN something's wrong. The temperature and concentration of air coming out is meaningless.

If the component is being cooled well it will never get hot in the first place. And the way to do that is to cool it with external air. And a good heat sink. And a large, low rpm inaudible fan.

Buzz, buzz, buzz. </Hamlet>

Message #123 - Posted 2007/05/25 - The New Guy

This discussion has inspired me to test.
My wife's 20" G5 iMac has been in a room with ambiant temperature at around 79F, when loaded the CPU hasn't exceeded 131F.

Sounds a little high but then your ambient temperature is higher than here:
http://www.computerbob.com/guides/guide_cpu_cooling.php

Later he talks about the fans which have been improved in later revisions. Also, why didn't anybody mention you can't operate the iMac without the back cover (the stand attaches to it)? You mean nobody here even knew that? That means nobody has taken the cover off? Probably you could just lean it against something. A bit of a hassle, but next time you need to get in there...........:)

"Conclusion? It would be nice to easily take the cover off to be able to see the culprit processor fan, but unfortunately the whole "midplane" assembly (guts) of the computer must be lifted in order to remove the cover. Is there something rubbing in there? Is it just a poor quality fan? Is it poorly mounted? Who knows - but in my humble opinion this fan need not be any noisier than the other fans in the system. Why use a fan with an annoying buzz? It would be really nice to be a fly on the wall during Apple engineering meetings! ScienceMan has to come to the conclusion that Apple dropped the ball on this one, and as a result has set a very annoying undertone to an otherwise fantastic piece of engineering. If Apple was customer-savvy on this one, they would get themselves some quiet fans and design a kit for processor fan replacement - then start sending those suckers out!" -------------
I wonder if you insert a typical thermometer into the slit of the cabinet at the top, will it read the temperature, even approximately? I know that sounds pretty goofy, but it might work. And most of us have an old thermometer lying around somewhere. Whereas very few of us have a laser or infared thermometer that enables the measurement of a surface temperature from a short distance away. Even that might be difficult given the space constraints in the iMac case.

Message #124 - Posted 2007/05/25 - The New Guy

BTW, my 17" iMac Core Duo has been running nonstop for more than 24 hours. (I had to restart it yesterday when experimenting with Boot Camp.) and the airflow out the vent on the top back is barely warm. I don't see where you get this "hot" from in the first place.

Interesting. Unless its actually a lot warmer than you say it is, which I doubt. And Jolly Roger's iMac seemed to run well by his temperatures. What I don't understand is why the stores multiple iMacs were running so hot.

I'm going to go back to that place and check again. Not to drag this out forever, but I owe that to you all. Hopefully real soon.

Checking the *feel* of the air coming out isn't enough. Your hands might be more or less sensitive depending on numerous factors outside of your control, and even ignoring that, it's just not relevant. You need to check the temperature of the actual components. Ifyou can download or get them to install MenuMeters or XResourceGraph, you can see the temperature of the components, then jot down the temperatures to take with you.

That's a great suggestion! Hey maybe, if its online, I can even email myself a screen shot!

Message #125 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 18:32:53 -0500, The New Guy said:

The Mac Pro seems to be the best or one of
the best so far though.

I disagree. I think the G5 is the best one so far, by a wide margin.

JR

Message #126 - Posted 2007/05/25 - The New Guy

The Mac Pro seems to be the best or one of
the best so far though.

I disagree. I think the G5 is the best one so far, by a wide margin.

Wow.....that was unexpected. How come? Because of the water cooling?

Message #127 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 18:57:07 -0500, The New Guy said:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

JR

Message #128 - Posted 2007/05/25 - Dave Balderstone

Previously, Jolly Roger wrote:

On 2007-05-25 18:57:07 -0500, The New Guy said:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

Yes, but it's WARM when expelled! Someone call Steve Jobs and tell him his designers are idiots! The cool air entering at the bottom should be *cool* when it leaves at the top!

Doesn't Apple understand ANYTHING? If they made the iMac six inches thick they could have cool air EVERYWHERE!

Dolts.

Message #129 - Posted 2007/05/25 - The New Guy

Previously, Dave Balderstone wrote:

Previously, Jolly Roger wrote:

On 2007-05-25 18:57:07 -0500, The New Guy said:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

Yes, but it's WARM when expelled! Someone call Steve Jobs and tell him his designers are idiots! The cool air entering at the bottom should be *cool* when it leaves at the top!

Doesn't Apple understand ANYTHING? If they made the iMac six inches thick they could have cool air EVERYWHERE!

Dolts.

Now THAT was funny! :) Well done....:)

Message #130 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 19:17:36 -0500, The New Guy said:

This discussion has inspired me to test.
My wife's 20" G5 iMac has been in a room with ambiant temperature at around 79F, when loaded the CPU hasn't exceeded 131F.

Sounds a little high

Nope. It's not high at all. The official PowerPC970FX RISC Microprocessor specifications state the recommended thermal diode temperature should be in the range of 0 to 105Àö Celsius. 131Àö F is only 55Àö C, which is clearly well within recommended operating range.

but then your ambient temperature is higher than here: http://www.computerbob.com/guides/guide_cpu_cooling.php

That entire web page is based on AMD Athlon processors, crappy PC single-compartment cases, etc. There's no reason to think any of the number on that page apply to an iMac G5.

Later he talks about the fans which have been improved in later revisions. Also, why didn't anybody mention you can't operate the iMac without the back cover (the stand attaches to it)? You mean nobody here even knew that? That means nobody has taken the cover off?

Nobody has ever *needed* to take the back cover off. You're trying to fabricate a problem out of thin air.

Repeat after me: THERE IS NO COOLING PROBLEM.

Probably you could just lean it against something. A bit of a hassle, but next time you need to get in there...........:)

Yeah, because when I buy Macs the first thing I want to do is rig them up to look like they are broken for no good reason. ; )

"Conclusion? It would be nice to easily take the cover off to be able to see the culprit processor fan, but unfortunately the whole "midplane" assembly (guts) of the computer must be lifted in order to remove the cover. Is there something rubbing in there? Is it just a poor quality fan? Is it poorly mounted? Who knows - but in my humble opinion this fan need not be any noisier than the other fans in the system. Why use a fan with an annoying buzz? It would be really nice to be a fly on the wall during Apple engineering meetings! ScienceMan has to come to the conclusion that Apple dropped the ball on this one, and as a result has set a very annoying undertone to an otherwise fantastic piece of engineering. If Apple was customer-savvy on this one, they would get themselves some quiet fans and design a kit for processor fan replacement - then start sending those suckers out!" -------------
I wonder if you insert a typical thermometer into the slit of the cabinet at the top, will it read the temperature, even approximately? I know that sounds pretty goofy, but it might work.

Why are you so fixated on the temperature of the air exiting the case when what really matters is the temperature of the individual components.

You can read the temperature of various components directly, so why not concentrate on *those* temperatures?

JR

Message #131 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 19:20:56 -0500, The New Guy said:

The Mac Pro seems to be the best or one of
the best so far though.

I disagree. I think the G5 is the best one so far, by a wide margin.

Wow.....that was unexpected. How come? Because of the water cooling?

Unexpected why?

No, I'm talking about the majority of G5 models which are not water cooled.

The G5 has four separate compartments where separate components are cooled with their own separate, fresh supply of cool air. It has direct-read thermo sensors built in that are monitored by the operating system in real time. Based on those readings, the operating systems constantly control fan speed, which, in turn, controls air flow, to keep the various component temperatures within specifications. It works and works very well.

JR

Message #132 - Posted 2007/05/25 - Dave Balderstone

Previously, The New Guy wrote:

Previously, Dave Balderstone wrote:

Previously, Jolly Roger wrote:

On 2007-05-25 18:57:07 -0500, The New Guy said:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

Yes, but it's WARM when expelled! Someone call Steve Jobs and tell him his designers are idiots! The cool air entering at the bottom should be *cool* when it leaves at the top!

Doesn't Apple understand ANYTHING? If they made the iMac six inches thick they could have cool air EVERYWHERE!

Dolts.

Now THAT was funny! :) Well done....:)

Thanks. It was a paraphrase of your posts throughout this thread.

Why aren't you rich designing cooling systems for computers?

Message #133 - Posted 2007/05/25 - The New Guy

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to the other area.

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

And is warmed by those components, especially so as its such a confined space. What I don't understand is why not just have the fan for the CPU, hard drive and whatever else right on top of the device being cooled? Exterior air going right to the heated air and then vented out. Instead, in this design it has to wander around from the bottom and snake its way out the top. And why use such tiny fans? 120 or 140 mm fans could have been placed on the back (in white for those of you sensitive to color clashing), ran at 5 volts and they'd have good airflow, venting out the top with total inaudibility. Maybe in another design soon.

What is interesting is that this may be the only case design that uses a more or less straight vertical movement for the heat. Maybe this is the start of something really good. Heat does rise after all.

Message #134 - Posted 2007/05/25 - The New Guy

This discussion has inspired me to test.
My wife's 20" G5 iMac has been in a room with ambiant temperature at around 79F, when loaded the CPU hasn't exceeded 131F.

Sounds a little high

Nope. It's not high at all. The official PowerPC970FX RISC Microprocessor specifications state the recommended thermal diode temperature should be in the range of 0 to 105Àö Celsius. 131Àö F is only 55Àö C, which is clearly well within recommended operating range.

but then your ambient temperature is higher than here: http://www.computerbob.com/guides/guide_cpu_cooling.php

That entire web page is based on AMD Athlon processors, crappy PC single-compartment cases, etc. There's no reason to think any of the number on that page apply to an iMac G5.

That's a good point. AMD Athlon CPU's run very hot. So getting down to those lower temperatures is quite a feat. Sounds like the iMac should be well below those numbers. But that's just a guess.

Later he talks about the fans which have been improved in later revisions. Also, why didn't anybody mention you can't operate the iMac without the back cover (the stand attaches to it)? You mean nobody here even knew that? That means nobody has taken the cover off?

Nobody has ever *needed* to take the back cover off. You're trying to fabricate a problem out of thin air.

Like I said, I'll check with a store for the air temps.

Repeat after me: THERE IS NO COOLING PROBLEM.

I will later if it turns out there isn't. You deserve it. You've been very patient all through this!

Probably you could just lean it against something. A bit of a hassle, but next time you need to get in there...........:)

Yeah, because when I buy Macs the first thing I want to do is rig them up to look like they are broken for no good reason. ; )

) Yeah, yeah........ I sure you realize that I meant that it would

interesting to see what the operating temps would be like with the back off.

"Conclusion? It would be nice to easily take the cover off to be able to see the culprit processor fan, but unfortunately the whole "midplane" assembly (guts) of the computer must be lifted in order to remove the cover. Is there something rubbing in there? Is it just a poor quality fan? Is it poorly mounted? Who knows - but in my humble opinion this fan need not be any noisier than the other fans in the system. Why use a fan with an annoying buzz? It would be really nice to be a fly on the wall during Apple engineering meetings! ScienceMan has to come to the conclusion that Apple dropped the ball on this one, and as a result has set a very annoying undertone to an otherwise fantastic piece of engineering. If Apple was customer-savvy on this one, they would get themselves some quiet fans and design a kit for processor fan replacement - then start sending those suckers out!" -------------
I wonder if you insert a typical thermometer into the slit of the cabinet at the top, will it read the temperature, even approximately? I know that sounds pretty goofy, but it might work.

Why are you so fixated on the temperature of the air exiting the case when what really matters is the temperature of the individual components.

In my experience, its a very good indication. Every case I've ever worked on exits cool to warm area when its being cooled well. It truly is a good indicator. Once you dismantle these things, and analyse exactly what is producing the heat, then rectify that exact problem, you learn what works, what doesn't, and what the expelled air feels like. I know I'm sounding like a broken record though.........

You can read the temperature of various components directly, so why not concentrate on *those* temperatures?

That's another good point you bring up. On a typical iMac, how many temperature sensors are there and where are they? I need to know that before I go snooping at that poor store I'm going to pester. I'd feel pretty stupid posting that info only to find out later I missed the most important information of all.

Message #135 - Posted 2007/05/25 - The New Guy

The Mac Pro seems to be the best or one of
the best so far though.

I disagree. I think the G5 is the best one so far, by a wide margin.

Wow.....that was unexpected. How come? Because of the water cooling?

Unexpected why?

Because I thought the Mac Pro improved on all the features of the previous model. I've never read about people wishing for the previous model's features. Not that I've done much reading on G5/Mac Pro differences, mind you. But the press has been so gushing over the Mac Pro.......

No, I'm talking about the majority of G5 models which are not water cooled. The G5 has four separate compartments where separate components are cooled with their own separate, fresh supply of cool air. It has direct-read thermo sensors built in that are monitored by the operating system in real time. Based on those readings, the operating systems constantly control fan speed, which, in turn, controls air flow, to keep the various component temperatures within specifications. It works and works very well.

Isn't that the same in the Mac Pro?

Message #136 - Posted 2007/05/25 - The New Guy

Previously, Dave Balderstone wrote:

Previously, The New Guy wrote:

Previously, Dave Balderstone wrote:

Previously, Jolly Roger wrote:

On 2007-05-25 18:57:07 -0500, The New Guy said:

Right, and that's what the iMac does; it expels the heated air right away, before the heat can be radiated. Heat sinks are not supposed to radiate because when they radiate, they transfer the heat to other components and to the case.

Actually - no. :) Right away means heat from one area doesn't go to
the other area.

Yup, it goes right from the component and its heat sink to the air, and
is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

Yes, but it's WARM when expelled! Someone call Steve Jobs and tell him his designers are idiots! The cool air entering at the bottom should be *cool* when it leaves at the top!

Doesn't Apple understand ANYTHING? If they made the iMac six inches thick they could have cool air EVERYWHERE!

Dolts.

Now THAT was funny! :) Well done....:)

Thanks. It was a paraphrase of your posts throughout this thread.

Why aren't you rich designing cooling systems for computers?

In time, my son, in time.....lol. Not that the rich part is important but it often does indicate you can offer something useful to the world. "What's the use of happiness. It can't buy you money." Henny Youngman.

Message #137 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 21:38:37 -0500, The New Guy said:

You can read the temperature of various components directly, so why not concentrate on *those* temperatures?

That's another good point you bring up. On a typical iMac, how many temperature sensors are there and where are they?

Depends on the model, of course. If Apple has a list of all sensors in each model Mac, I'm unaware of it.

FYI: You can use this command to read the temperature sensor values at any time on any current model Mac:

ioreg -n IOHWSensor

You can filter the results down to save yourself some reading (caution - not sure how Usenet and news readers will clobber this) with this single command line:

ioreg -n IOHWSensor | awk '/location/ || /current-value/ || /"type"/' | sed -e 's/[^"]*"//' -e 's/" =//' -e 's/location//' -e 's/type//' -e 's/"//g' | awk '{ d=($2/65536); if ($1=="current-value") print substr(d,1,7) "\t" ; if ($1!="current-value") print $0 }' | sed -e 's/temperature/Celsius/' -e 's/voltage/volts/' -e 's/fanspeed/fan RPM/' -e 's/current/Amps/' -e 's/^temp$//' -e 's/ //' | awk '{ if ((NR % 3) == 0) print $0; else printf $0 " " }'

Anyway, on my G5 tower, I have:

28.5 DRIVE BAY Celsius
46.875 BACKSIDE Celsius
56.125 U3 HEATSINK Celsius
34.75 MLB MAX6690 AMB Celsius
25.5 MLB INLET AMB Celsius
0 SLOT 12V power
0 SLOT 5V power
0 SLOT 3.3V power
0.48451 SLOT COMBINED power
43.25 CPU A AD7417 AMB Celsius
0.00660 CPU A AD7417 AD1 adc
0.00895 CPU A AD7417 AD2 adc
0.00933 CPU A AD7417 AD3 adc
0.00306 CPU A AD7417 AD4 adc
47.25 CPU B AD7417 AMB Celsius
0.00891 CPU B AD7417 AD1 adc
0.00958 CPU B AD7417 AD2 adc
0.00930 CPU B AD7417 AD3 adc
0.00343 CPU B AD7417 AD4 adc

On my mom's iMac G5:

41.5 ODD Temp Celsius
49.5 HD Temp Celsius
57.25 NB Ambient Celsius
80.25 NB Temp Celsius
68.375 GPU Ambient Celsius
80.125 GPU Temp Celsius
0.00144 CPU Current Amps
0.01263 CPU Voltage volts
0.00825 CPU T-Diode temp
32.5 Incoming Air Temp Celsius

Probably easiest just to install a temperature monitor application like one of these:

<http://islayer.com/index.php?op=item&id=22> <http://www.macupdate.com/info.php/id/10220/x-resource-graph>

JR

Message #138 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 21:44:42 -0500, The New Guy said:

The Mac Pro seems to be the best or one of
the best so far though.

I disagree. I think the G5 is the best one so far, by a wide margin.

Wow.....that was unexpected. How come? Because of the water cooling?

Unexpected why?

Because I thought the Mac Pro improved on all the features of the previous model. I've never read about people wishing for the previous model's features. Not that I've done much reading on G5/Mac Pro differences, mind you. But the press has been so gushing over the Mac Pro.......

Well I haven't seen wide-spread complaints about Mac Pros overheating, so...

No, I'm talking about the majority of G5 models which are not water cooled.

The G5 has four separate compartments where separate components are cooled with their own separate, fresh supply of cool air. It has direct-read thermo sensors built in that are monitored by the operating system in real time. Based on those readings, the operating systems constantly control fan speed, which, in turn, controls air flow, to keep the various component temperatures within specifications. It works and works very well.

Isn't that the same in the Mac Pro?

No, Apple moved stuff around in the Mac Pro. Look at them side-by-side:

<http://jollyroger.kicks-ass.org/jollyroger/G5_and_Mac_Pro_Cases.jpg>

In the G5, the hard drives are isolated from the rest of the case and are spaced apart such that air is constantly flowing between them. Also in the G5 they share air only with the CD/DVD drive, which more often than not, isn't generating much heat at all. In the Mac Pro, the hard drives share air space with PCI cards, and aren't spaced apart like they were in the G5.

Because I don't have a Mac Pro yet (my G5s are still going strong), it's unclear to me if the air flow of the processors and RAM is actually kept separate from the hard drives and PCI cards. It's hard to tell from the picture - they seem to have separate air flows, but I'm not sure.

JR

Message #139 - Posted 2007/05/25 - Jolly Roger

On 2007-05-25 21:24:56 -0500, The New Guy said:

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

And is warmed by those components, especially so as its such a confined space.

You have not demonstrated that this causes a problem.

What I don't understand is why not just have the fan for the CPU, hard drive and whatever else right on top of the device being cooled? Exterior air going right to the heated air and then vented out.

Many potential reasons come to mind, like maybe they wanted a sleek case design, and hanging two ugly-ass fans on the back didn't look good. Maybe they simply figured out how to do it with the fans at the top so the components are kept cool enough.

Instead, in this design it has to wander around from the bottom and snake its way out the top.

You haven't demonstrated that this is a problem.

And why use such tiny fans?

Maybe to keep them hidden, for aesthetic appeal, maybe other reasons. Only Apple knows.

120 or 140 mm fans could have been placed on the back (in white for those of you sensitive to color clashing), ran at 5 volts and they'd have good airflow, venting out the top with total inaudibility. Maybe in another design soon.

I hope not. I don't want an ordinary PC case. I want a sexy Apple case.

What is interesting is that this may be the only case design that uses a more or less straight vertical movement for the heat. Maybe this is the start of something really good. Heat does rise after all.

Well, technically, heat rises only when air flow is not sufficient to transfer the heat in another direction or when the air flow is going up anyway.

JR

Message #140 - Posted 2007/05/26 - Eric Lindsay

Previously, The New Guy wrote:

What's the difference between a Windows machine and a Mac these days? Not the CPU, hard drives, optical drives. Just the motherboard. The CPU in both is the primary heat generator. That, along with the video card and hard drives generate most of the heat in both machines.

All of which could explain why an iMac G5 like you complained about has three fans, for CPU, hard drive, and system. If it had a games class graphic card, doubtless there would have been a fourth fan. Each fan is different in design, runs at a different idle speed (1500, 2300 and 1700 rpm) and has a different peak speed. Apple didn't just shove a commodity fan in place, like so many PC cases.

The idle temperature for the CPU is around 52C-55C, and around 50C for the hard disk once it has been running for an half hour or so. The fan ramp up speeds vs temperatures show clear care in getting the heat out, and ensuring the CPU temperature isn't over 75C. As you would be aware, the silicon surface area of a G5 (or a Pentium) and their full power draw ensures that their uncooled temperature exceeds that of a bar radiator (at which point they wouldn't actually be working) so it is very important to ensure temperatures are handled correctly. Any computer that works at all is getting this mostly right. The rest of the treatment is a matter of what the design is intended to do.

I note that with fans at idle, my 2 year old iMac sound levels are below 35dBA, so we can easily conclude that low sound levels were an important design consideration. Especially important in a computer in which the fans are about level with your ears.

While I am certain running the temperatures the iMac uses risks a slightly greater failure rate, the usual heuristic is fourth power of the temperature in degrees Celsius. I am in the tropics, so my summer ambient temperatures are around 30C, rather than the office temperature that Apple probably considered reasonable. n terms of service life, this is lost in the noise. I will have replaced the iMac well before I could expect it to fail due to temperature related faults.

If you are actually seeing excessive temperatures at the air exit of an iMac G5 on idle, then there is a hardware fault. I suggest using X Resource Graph to check your internal temperatures, and a contactless thermometer to check the exit temperatures.

For example, on a iMac G5 ALS just woken from sleep, I am seeing a CPU of 46C, a hard drive of 28C, and an exit air temperature of 26.5C (ambient here is currently 26C - southern hemisphere tropics approaching winter - we don't actually have seasons, except for wet and perfect).

You have probably noted that by using a laptop class CPU in the Intel iMac, Apple dropped internal temperatures considerably. This means that next version of the iMac will be smaller (probably thinner - and run hotter). Pretty much irrelevant. The expected life span will be adequate for an appliance that you expect to be replaced every 3 or 4 years.

Maybe you could let us know what temperature you are seeing on CPU, hard drive and exit air?

Message #141 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

What the fuck are you talking about? The air comes in the bottom, picks up heat as it travels up the case through four well defined pathways (which is what it is supposed to do), and exits out the vents.

If you are a troll, I salute you; you are a very good troll because I can't tell whether you are trolling or are really as stupid and ignorant as you appear to be.

Support the troops: Bring them home ASAP.

Message #142 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

And is warmed by those components, especially so as its such a confined space. What I don't understand is why not just have the fan for the CPU, hard drive and whatever else right on top of the device being cooled?

Because the fan is not supposed to blow on the component; the fan is supposed to suck the hot air out of the case.

Exterior air going right to the heated air and then vented out. Instead, in this design it has to wander around from the bottom and snake its way out the top.

Yup; the hot air is supposed to leave the case. That's how it removes the heat from the inside. If the air simply blows around inside the case, it's not going to cool anything.

Support the troops: Bring them home ASAP.

Message #143 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

After playing with many designs I can positively say that if a device is being cooled well you will NEVER have hot air coming off it when its idling.

I'm sure that you know all about hot air; you're so full of it yourself.

Support the troops: Bring them home ASAP.

Message #144 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

I meant it gets external air and that air goes directly to the item to cool without passing around/over/underneath anything else. After its cooled it goes DIRECTLY out - not passing over/around/underneath anything else.

Holy computer cases, Batman!

I've been fiddling around a lot with heat sinks and cooling the last few months so though it may sound like I'm an idiot at times, I have applied myself a lot in this area, enough to see the failings in every design I've looked closely at.

Well, nothing is perfect; the world is full of trade offs. But if you're such a genius at designing computer cases and at designing cooling systems, why don't you hire yourself out to Apple, Dell, HP/Compaq, and/or the other computer companies as a consultant, and make a bundle of money?

Oh, I know why: you're not the genius you think you are.

Incidentally, many aspects of the Mac Pro I thought of as improvements for generic PC cases before the Mac Pro came out.

I'll just file that as another of your fantasies.

Wouldn't it be ironic if the iMac does it right?

Not at all.

Well its clear that nobody does it "right".

The iMac does it right; it doesn't overheat.

A good design uses external air to cool ONE heat generator and immediately expels that heated air.

Really? And your source for that bit of information is what? Or did you make it up all by your lonesome?

Support the troops: Bring them home ASAP.

Message #145 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

Repeat after me: THERE IS NO COOLING PROBLEM.

I will later if it turns out there isn't. You deserve it. You've been very patient all through this!

All you would have to do is document actual reports of cooling problems. There aren't any such reports; therefore, there is no cooling problem.

Why are you so fixated on the temperature of the air exiting the case when what really matters is the temperature of the individual components.

In my experience, its a very good indication. Every case I've ever worked on exits cool to warm area when its being cooled well.

The actual temperature inside is a better indicator.

Here are the temperatures for my computer, which has been operating for a few days, as measured by the Temperature Monitor application (version 4.11):

SMART Disk WDC WD1600JS-40NGB2 (WD-WCAP01019097): 127.4 ∞F Ambient Air: 93.2 ∞F CPU A Temperature Diode: 123.8 ∞F Graphics Processor Temperature Diode: 161.6 ∞F Hard Drive Bay 1: 122.0 ∞F Memory Controller: 127.4 ∞F Optical Drive: 109.4 ∞F CPU Core 1: 113.0 ∞F CPU Core 2: 111.2 ∞F

All are within specification--well within specification.

Regarding the ambient air reading, I assume that the sensor is measuring the air as it is being expelled from the case.

Support the troops: Bring them home ASAP.

Message #146 - Posted 2007/05/26 - Jolly Roger

On 2007-05-26 11:34:33 -0500, Michelle Steiner said:

Here are the temperatures for my computer

What model Mac is that?

JR

Message #147 - Posted 2007/05/26 - Michelle Steiner

Previously, Jolly Roger wrote:

Here are the temperatures for my computer

What model Mac is that?

iMac Core Duo (first model of the Intel iMac)

Support the troops: Bring them home ASAP.

Message #148 - Posted 2007/05/26 - The New Guy

Wait... According to that picture, the air going to the fans is on its way to be expelled out the top. Cool air enters from the bottom, passes over components, goes through the fans, and then is expelled at the top.

And is warmed by those components, especially so as its such a confined space. What I don't understand is why not just have the fan for the CPU, hard drive and whatever else right on top of the device being cooled?

Because the fan is not supposed to blow on the component; the fan is supposed to suck the hot air out of the case.

Well IF the fan blew right on the component, that same component would be much cooler. That's positive air pressure whereas negative air pressure is sucking the air out of the case. Positive air works better. Also positive air usually uses cooler exterior air instead of warmed interior air to that's obviously a huge benefit to the cooling process.

Message #149 - Posted 2007/05/26 - The New Guy

Yup, it goes right from the component and its heat sink to the air, and is convected out of the case.

As you can see it does nothing of the kind.
http://www.medicalmac.com/mac111.html#3 The air going to the fans is already warmed by the innards of the computer. Unless they have changed the airflow path in a later revision of the Intel iMac (highly doubtful). Once again a design is using warmed interior case air to try to cool a component.

What are you talking about? The air comes in the bottom, picks up heat as it travels up the case through four well defined pathways (which is what it is supposed to do), and exits out the vents.

And "picking up heat" is exactly what you don't want airflow to do (for more than one component if its a hot component like a CPU heatsink.) As computers get hotter and hotter as performance increases and especially as Raid 0 with multiple hard drives attracts more and more users, you'll find that hard drives will be in their own air channel, with sound absorbing material inside minimizing the platter whine, especially if 10k and 15k models gain more popularity.

Anyway, as the air travels up the case it is warmed by the components in that case. Its using warmed interior air to cool a component. Bad design. Sorry to put down your equipment Michelle, but I guess you really haven't experimented much with cooling and computer parts or you wouldn't have asked such a question. For software I salute you, you're one of the most knowledgeable minds I've ever met. For hardware, not so much. Remember that Borat clip where they're at the dinner table and he's talking about the 2 wives and how they'd be accepted in the home country? Same inflection in the "not so much".......:) And what's with the profanity? Surely you're above that.

Message #150 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

Because the fan is not supposed to blow on the component; the fan is supposed to suck the hot air out of the case.

Well IF the fan blew right on the component, that same component would be much cooler.

According to Intel, you're full of hot air. Check out "Thermally Advantaged Chassis"; you'll see what Intel recommends, and compare that to what the Macintosh computers do. You'll find that they're pretty much the same, and nothing like what you say it should be like.

Here is a hint: "As with most computers, the rear fan and power supply fan exhaust, moving hot air away from the computer. This causes a slight depressurization inside the chassis, and requires all other openings to become intake vents. Airflow from the front of the chassis moves around the Chassis Air Guide, allowing the processor fan to only draw air from outside the chassis, providing more effective cooling."

The following is from an article on computer fans: "While in earlier personal computers it was possible to cool most components using convection (passive cooling), more efficient cooling has become a necessity on many components. To cool these components, fans are used to move heated air away from the components and draw cooler air over them."

And from an article on computer cooling: "Desktop computers typically use one or more fans for heat management. Almost all desktop power supplies have at least one fan to exhaust air from the case. Most manufacturers recommend bringing cool, fresh air in at the bottom front of the case, and exhausting warm air from the top rear.

"If there is more air being forced into the system than being pumped out (due to an imbalance in the number of fans), this is referred to as a "positive" airflow, as the pressure inside the unit would be higher than outside. A balanced or neutral airflow is the most efficient, although a slightly positive airflow results in less dust build up if dust filters are used."

That's positive air pressure whereas negative air pressure is sucking the air out of the case. Positive air works better.

Prove that it works better!

Also positive air usually uses cooler exterior air instead of warmed interior air to that's obviously a huge benefit to the cooling process.

They both use exterior air.

Support the troops: Bring them home ASAP.

Message #151 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

And "picking up heat" is exactly what you don't want airflow to do

That is *exactly* what you want it to do. If it doesn't pick up heat and take that heat out of the case, the heat remains in the case, and the case and the components just keep getting hotter and hotter.

Sorry to put down your equipment Michelle, but I guess you really haven't experimented much with cooling and computer parts or you wouldn't have asked such a question.

You have demonstrated that you don't know anything about computer design, cooling design, or anything related to them.

Support the troops: Bring them home ASAP.

Message #152 - Posted 2007/05/26 - Adrian

The New Guy wrote:

And "picking up heat" is exactly what you don't want airflow to do

I don't think it's worth you continuing this nonsense. You are missunderstanding so much it would be best just to leave it there.

Adrian

Message #153 - Posted 2007/05/26 - The New Guy

Previously, Jolly Roger wrote:

On 2007-05-25 21:38:37 -0500, The New Guy said:

You can read the temperature of various components directly, so why not concentrate on *those* temperatures?

That's another good point you bring up. On a typical iMac, how many temperature sensors are there and where are they?

Depends on the model, of course. If Apple has a list of all sensors in each model Mac, I'm unaware of it.

FYI: You can use this command to read the temperature sensor values at any time on any current model Mac:

ioreg -n IOHWSensor

You can filter the results down to save yourself some reading (caution - not sure how Usenet and news readers will clobber this) with this single command line:

ioreg -n IOHWSensor | awk '/location/ || /current-value/ || /"type"/' | sed -e 's/[^"]*"//' -e 's/" =//' -e 's/location//' -e 's/type//' -e 's/"//g' | awk '{ d=($2/65536); if ($1=="current-value") print substr(d,1,7) "\t" ; if ($1!="current-value") print $0 }' | sed -e 's/temperature/Celsius/' -e 's/voltage/volts/' -e 's/fanspeed/fan RPM/' -e 's/current/Amps/' -e 's/^temp$//' -e 's/ //' | awk '{ if ((NR % 3) == 0) print $0; else printf $0 " " }'

Anyway, on my G5 tower, I have:

28.5 DRIVE BAY Celsius
46.875 BACKSIDE Celsius
56.125 U3 HEATSINK Celsius
34.75 MLB MAX6690 AMB Celsius
25.5 MLB INLET AMB Celsius
0 SLOT 12V power
0 SLOT 5V power
0 SLOT 3.3V power
0.48451 SLOT COMBINED power
43.25 CPU A AD7417 AMB Celsius
0.00660 CPU A AD7417 AD1 adc
0.00895 CPU A AD7417 AD2 adc
0.00933 CPU A AD7417 AD3 adc
0.00306 CPU A AD7417 AD4 adc
47.25 CPU B AD7417 AMB Celsius
0.00891 CPU B AD7417 AD1 adc
0.00958 CPU B AD7417 AD2 adc
0.00930 CPU B AD7417 AD3 adc
0.00343 CPU B AD7417 AD4 adc

On my mom's iMac G5:

41.5 ODD Temp Celsius
49.5 HD Temp Celsius
57.25 NB Ambient Celsius
80.25 NB Temp Celsius
68.375 GPU Ambient Celsius
80.125 GPU Temp Celsius
0.00144 CPU Current Amps
0.01263 CPU Voltage volts
0.00825 CPU T-Diode temp
32.5 Incoming Air Temp Celsius

Probably easiest just to install a temperature monitor application like one of these:

<http://islayer.com/index.php?op=item&id=22> <http://www.macupdate.com/info.php/id/10220/x-resource-graph>

Interesting temperatures.

Now compare them with these from this page noting that these people are aiming for noise reduction, not the best cooling possible!!! http://www.silentpcreview.com/forums/viewtopic.php?t=7654&postdays=0&po storder=asc&start=30

"My temps as i write this (idle) are 39C HDD, 31C CPU, 29C case in a 25C room. These temps are WAY better than the "old rig" (now in quiet upgrade land), so you have plenty of upside to still be safe."

"Just got the case yesterday and think it's great. When I set it up the hard drive temps (Maxtor 120GB, 7200rpm) were idling around 39 - 41 deg C. So I just took an old 92mm fan that I had, did the 7 volt mod to it, and then placed it in the drive cage below to blow onto the hdd. Now the hdd idles at 24 - 25 deg.C and maxes out at 28 deg.C (ran a thorough defrag on it that lasted about half an hour)." (An example of how well positive airflow works, Michelle.)

So I've been reading about the Mac Pro. Too bad there are so few used right now. I guess that's a good indication most users are very satisfied with it. People seem to love it except for the hard drive(s) noise. But with Sorbothane or other tricks used on SPCR, that should be able to be remedied. Oh, another review commented on the latency problems of FB-Dimms, but I'm not knowledgeable in that area so it doesn't mean much to me yet. They were saying that if you tend to multi task a lot, you should use 2 pairs of ram sticks (like 4 x 512mb's instead of 2 x 1gb's for 2 gb's) to make use of the quad core technology. Sounds easy enough. And strangely they lament for a non-Xeon, single CPU, non-FB-Dimm machine with eSata.....just what I was moaning about a few weeks ago here. LOL. Yeah, I'm such an idiot.

Message #154 - Posted 2007/05/26 - The New Guy

Wow Eric, what a great post. My replies below.
But ...... I was thinking that it would be interesting to compare the idle/max temperatures of the CPU in the iMac with the same CPU in a PC box. At least then we're comparing apples with apples (CPU's would be identical). Since the CPU is the main heat generator, its logical to focus on that first. I sure wish someone could pop off the back of that iMac to compare tamps!

What's the difference between a Windows machine and a Mac these days? Not the CPU, hard drives, optical drives. Just the motherboard. The CPU in both is the primary heat generator. That, along with the video card and hard drives generate most of the heat in both machines.

The idle temperature for the CPU is around 52C-55C, and around 50C for the hard disk once it has been running for an half hour or so. The fan ramp up speeds vs temperatures show clear care in getting the heat out, and ensuring the CPU temperature isn't over 75C. As you would be aware, the silicon surface area of a G5 (or a Pentium) and their full power draw ensures that their uncooled temperature exceeds that of a bar radiator (at which point they wouldn't actually be working) so it is very important to ensure temperatures are handled correctly. Any computer that works at all is getting this mostly right. The rest of the treatment is a matter of what the design is intended to do.

I note that with fans at idle, my 2 year old iMac sound levels are below 35dBA, so we can easily conclude that low sound levels were an important design consideration. Especially important in a computer in which the fans are about level with your ears.

Thankfully, people are now measuring the frequency of the noise. That is crucial as higher frequencies are usually more distracting and annoying. I type in a higher frequency as some of you have duly noted. ;)

While I am certain running the temperatures the iMac uses risks a slightly greater failure rate, the usual heuristic is fourth power of the temperature in degrees Celsius. I am in the tropics, so my summer ambient temperatures are around 30C, rather than the office temperature that Apple probably considered reasonable. In terms of service life, this is lost in the noise. I will have replaced the iMac well before I could expect it to fail due to temperature related faults.

But a computer should last for many years. That would be ideal.

If you are actually seeing excessive temperatures at the air exit of an iMac G5 on idle, then there is a hardware fault. I suggest using X Resource Graph to check your internal temperatures, and a contactless thermometer to check the exit temperatures.

Yeah - that would be nice if that was available at the store I was at!

) When I go back I'll poke around in Applications and Utilities to

see if it looks like there are any temperature monitoring programs in there. I wonder if System Profiler would give any useful information?

Message #155 - Posted 2007/05/27 - Eric Lindsay

Previously, Michelle Steiner wrote:

BTW, my 17" iMac Core Duo has been running nonstop for more than 24 hours. (I had to restart it yesterday when experimenting with Boot Camp.) and the airflow out the vent on the top back is barely warm. I don't see where you get this "hot" from in the first place.

That "hot" is what I am wondering about also. My iMac G5 ALS has only been running 9 days, and surfing this morning has not raised the fans above idle, so CPU temperature is only 54C and hard drive temperature 52C. The exit air temperature is 40C, at an ambient of 26C. To me, that says the fans are at about the right speed, and the cooling is about right. But 40C isn't what I'd call hot air from the exit.

Message #156 - Posted 2007/05/26 - Jolly Roger

On 2007-05-26 16:07:56 -0500, The New Guy said:

"picking up heat" is exactly what you don't want airflow to do

ROFLMAO!

JR

Message #157 - Posted 2007/05/26 - Jolly Roger

On 2007-05-26 16:07:56 -0500, The New Guy said:

as the air travels up the case it is warmed by the components in that case. Its using warmed interior air to cool a component.

No. The cool external air enters at the bottom and travels along the specifically-designed channel (a channel that does not pass over other components), it remains cool until it gets to the components it is meant to cool, only then does it get warmer and exit the case at the top.

JR

Message #158 - Posted 2007/05/26 - Jolly Roger

On 2007-05-26 19:19:20 -0500, The New Guy said:

But a computer should last for many years. That would be ideal.

Mac *do* last for years. In fact they last and are useful for many more years than the average PC.

JR

Message #159 - Posted 2007/05/26 - Jolly Roger

On 2007-05-26 19:19:20 -0500, The New Guy said:

Yeah - that would be nice if that was available at the store I was at!

) When I go back I'll poke around in Applications and Utilities to

see if it looks like there are any temperature monitoring programs in there. I wonder if System Profiler would give any useful information?

Nope you'll probably have to download it from http://macupdate.com or the like, and install it yourself.

JR

Message #160 - Posted 2007/05/26 - The New Guy

Because the fan is not supposed to blow on the component; the fan is supposed to suck the hot air out of the case.

Well IF the fan blew right on the component, that same component would be much cooler.

According to Intel, you're full of hot air. Check out "Thermally Advantaged Chassis"; you'll see what Intel recommends, and compare that to what the Macintosh computers do. You'll find that they're pretty much the same, and nothing like what you say it should be like.

That's because most computer cases are thoughtless in design. They are catering to the lowest common denominator.

Here is a hint: "As with most computers, the rear fan and power supply fan exhaust, moving hot air away from the computer. This causes a slight depressurization inside the chassis, and requires all other openings to become intake vents. Airflow from the front of the chassis moves around the Chassis Air Guide, allowing the processor fan to only draw air from outside the chassis, providing more effective cooling."

Anyone that has experimented with fans knows that positive air pressure is far more effective than negative air pressure. Not to mention is far more controllable.

The following is from an article on computer fans: "While in earlier personal computers it was possible to cool most components using convection (passive cooling), more efficient cooling has become a necessity on many components. To cool these components, fans are used to move heated air away from the components and draw cooler air over them."

Duh....

And from an article on computer cooling: "Desktop computers typically use one or more fans for heat management. Almost all desktop power supplies have at least one fan to exhaust air from the case. Most manufacturers recommend bringing cool, fresh air in at the bottom front of the case, and exhausting warm air from the top rear.

This is the same garbage design that Jolly Roger was mentioning before. Its just bad PC design. This is going to change in the next year or so. The components and the heat they generate will demand it, not to mention people are tired of cheap, noisy fans.

"If there is more air being forced into the system than being pumped out (due to an imbalance in the number of fans), this is referred to as a "positive" airflow, as the pressure inside the unit would be higher than outside. A balanced or neutral airflow is the most efficient, although a slightly positive airflow results in less dust build up if dust filters are used."

And anybody that is serious about cooling knows that dust filters severely detract from a fan's performance. They should only be used in the worst environments. A few minutes with a vacuum once a month with the correct accessory will keep most computer case interiors nice and dust free, if you have a dust problem. For others it might be every few months.

That's positive air pressure whereas negative air pressure is sucking the air out of the case. Positive air works better.

Prove that it works better!

Duh.....try it. I can't believe you don't know this. Its like saying is the sky blue? Have you never tried to cool to a hard drive with a fan? Try it directly (underneath of course since that is where most of the heat is coming from) and then put the hard drive in a column with the same fan sucking the air out of that column. There's no comparison. You can feel a huge difference just with touching the hard drive, both top, bottom and even the sides are cooler.

Also positive air usually uses cooler exterior air instead of warmed interior air to that's obviously a huge benefit to the cooling process.

They both use exterior air.

Yeah, but almost all designs move that air over warmed components before cooling other components. Thats the problem. Bring the air in, cool the device, vent the air out. That's the goal. Its rarely accomplished because of styling goals.

Message #161 - Posted 2007/05/26 - The New Guy

And "picking up heat" is exactly what you don't want airflow to do

That is *exactly* what you want it to do. If it doesn't pick up heat and take that heat out of the case, the heat remains in the case, and the case and the components just keep getting hotter and hotter.

I guess you don't understand what I said because I didn't express it clearly. I meant you don't want that air that is going to cool, for instance, the CPU heat sink, to "pick up heat" as it passes by other warm components. I'm sure we don't disagree on that point.

Message #162 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

And "picking up heat" is exactly what you don't want airflow to do

That is *exactly* what you want it to do. If it doesn't pick up heat and take that heat out of the case, the heat remains in the case, and the case and the components just keep getting hotter and hotter.

I guess you don't understand what I said because I didn't express it clearly. I meant you don't want that air that is going to cool, for instance, the CPU heat sink, to "pick up heat" as it passes by other warm components. I'm sure we don't disagree on that point.

It all depends on how much heat it picks up and the volume of the air passing the components in a given unit of time.

If you actually knew what you were talking about, you wouldn't be making these asinine statements.

Support the troops: Bring them home ASAP.

Message #163 - Posted 2007/05/26 - Michelle Steiner

Previously, The New Guy wrote:

That's because most computer cases are thoughtless in design. They are catering to the lowest common denominator.

<...>

This is the same garbage design that Jolly Roger was mentioning before.

OK, now I understand. Intel doesn't know anything; Apple doesn't know anything. No one except you knows anything. Only you know what good computer case design is. You are the only expert in the world on this subject.

Support the troops: Bring them home ASAP.

Message #164 - Posted 2007/05/26 - Michelle Steiner

Previously, Jolly Roger wrote:

But a computer should last for many years. That would be ideal.

Mac *do* last for years. In fact they last and are useful for many more years than the average PC.

I bought a Mac 128K in June 1984, and gradually upgraded it to a Mac Plus; I retired it in October 1990. In 1997, I sold it to an engineer (for 1% of its original list price) at Lockheed Martin to use for some kind of monitoring project. So far as I know, he's still using it.

Support the troops: Bring them home ASAP.

Need Help? Have a Question?

Looking for more help, comments, and answers?

Ask your questions on Ask Different. Ask Different is a community of Apple users ready to help.