CPU Cooler Charts 2008, Part I - Loosing Your Cool?

[EvilHomerINC]

yuur fuunny

I just thought it was amusing that while you were bytching about the editing you didn't make sure your post was edited. :lol:

I'm going to step in for croc and point out that it was a joke, not an oversight...

 

[hannibal]

Nice to see really big comparison! Normally there is only few of them at given time.

And some comments:
ranking: For some people the cooling is most important thing (owerclckers etc) to some another the easy installation and noice. (normal dudes, that want to change the original with littel better.) So I think that its relevant to give relative ranking based all measured things. If you are experienced mounting wizard, you can look only the cooling part and ignore the rest.
The measurement:If they keep the environment same for all pars, it should tell the difference, and thats good enough!
Mounting: Not sure about this. It should have impact, but now we have open installation, so it's not so much matter, because the hot air can go anywere without big effect. In real environment it will have bigger impact. I think that now this set up will test fairly well how efective the cooler draw out the heat from the prosesser.
Temps: As long as they are comparable to each other we can find the best solutions.

 

[IH8U]

The Nirvana FAILS, that has to be a load of BS. It seems to easily keep my 5000+ BE OC'ed to 3.2 at about 30c idle, and below 48c on load. This is with the AMD furnace proc! I had no issues with installation. WTF is so hard about using a flat screwdriver to push it onto the AMD board>? anyone? Buhler, Buhler? My vid card is still the loudest thing in my case (with the Nirvana set to 60%).

<end rant>.

 

[msoltyspl]

Push-down pins are probably the most ridiculous 'invention' I've seen so far, if we talk about heatsink mounting. In most cases very hard to lock, even harder with larger heatsinks (you might ask your friendly strongman for assistance). Due to cheap plastique used I've seen pins popping off after few months of use.

I don't understand how can anyone prefer push-down pins to backplate.

If removing motherboard is considered disadvantage, I'm speechless. Unless someone has a fetish towards changing hetsinks on a weekly basis.

 

[night_wolf_in]

I do not normally complain about anything, but there are some clear over sites here

1. Put it in a damn case(use one with a removable tray for fast swapping and cut a hole in it for backplate mounting.....).
2. Don't face the coolers every witch way(hot air rises, why now fact then up? many cases have a top fan now)
3. Don't complain about installation. you only do it once in most cases(it should have a minimum impact on end score.). and those push pins don't give me a feeling of secure with large heat sinks....
4. Why the hell are your temps SOOOOO high? I mean i can run my 9500AT or AC freezer(a little harder here with the close fins, but i have done it. Thats the $23 wonder heatsink is what it is) passive and get better then some of those(i mean 7-9 volts 55c -57 on the hottest core....52 on the lowest 2....lapping time?).
5. frostytech does kick ass

another thing is.

6-is the processor overclocked? yes or no? reading it tho seems it is no.
7-what is the vcore voltage?

 

[beowulfw]

I registered just so I can post my disgust at the poor quality of this "comprehensive" review, my complaints are similar to what has been voiced above.
Fans blowing downwards on most of the coolers, only the IFX had the fan blowing up to me is a major flaw in the results.

Pls redo it properly.

 

[SpinachEater]

I registered just so I can post my disgust at the poor quality of this "comprehensive" review, my complaints are similar to what has been voiced above.
Fans blowing downwards on most of the coolers, only the IFX had the fan blowing up to me is a major flaw in the results.

Pls redo it properly.

^ Explain how the fan direction is flawing the results when they aren't testing in a case.

To those worried about the "rising hot air"....please explain on how you feel this is affecting the results.

 

[CompTIA_Rep]

It is all relative. As long as each cooler was placed in the same conditions.
However, without a baseline, you can't take these numbers and compare them directly to real world systems.

This test is like pretend land with no magic rainbow leading back home. Dorothy is stuck in OZ, but she doesn't realize it.

 

[skittle]

I prefer the term: unscientific

 

[Grimmy]

Yes, ambient temperature can (and does) play a role here and I don't recall (without going back through the article) if they mentioned it or not. Still, there are too many case designs with too many fan setups to even possibly go through them. And remember, too, that dude is testing 80 of them! Personally, I wouldn't want to even test 5! And someone mentioned using a slide-out MB tray; are you serious? name the case with an opening that big! I do believe that even in cases with slide-out MB trays, the CPU cooler has to be removed.

:lol: . o O (oh... wahhhh.. )

Not my fault that they want to do 80 of them. Not to mention the temps they provide. Are they Tcase based, or Tjunction? I'm guessing they are averaging 4 cores (or maybe hottest core) on the Tjunction temps, since the thermal spec is 71C/G0 (Tcase base) and their tests showed on 6 coolers that went over that. Heh, one of them got as hot as 98.3C??? . o O (that would be close to the Tjunction Max 100C?)

*note - B3 Version of the Q6600, the thermal spec is 62.2C

Also looking at a couple results:

Fan High / with load on CPU:

Zalman 8700 (FAILED) - 71C
Zaward Sylphee - The Ugly Duckling (PASSED) - 89C

Fan Low / with load on CPU:

Zalman 8700 (FAILED) - 80.5C
Zaward Sylphee - The Ugly Duckling (PASSED) - NOT POSSIBLE FAN AUTO?????

Yet... with the Zalman on low (80.5C) is still cooler then 89C with the Sylphee on high and ummm passed???


Even though... it's 80 different coolers they are going to test, they could (gawd I have to give ideas?) make plexi-glass box like cover that would mimic a PC case with fan holes, and create airflow. Using a fan controller, they could also mimic a bad/good/best airflow scenario, without having to change the case.

I mean, if you don't have the time do the test right, THEN DON'T DO IT! . o O (let the other site provide actual data)

. o O (that's my rant... I'm bouncing away now)

Edit:

Just to verify what I was saying about the 2 coolers above.

Edit:

We were disappointed by manufacturers Scythe, Silverstone, Zerotherm and Zaward. Their products failed the test due to insufficient cooling performance, inferior build quality or an ill-conceived installation procedure.

Question... Did those companies claim it would sufficiently cool a quad, with the particular product they chose to test?

 

[SpinachEater]

I prefer the term: unscientific

More so non-analytical.

 

[dark41]

Well here's my 2 cents worth.

I build hundreds of systems every month. Only overclockers are built with aftermarket fans, which is a pretty small percentage. But all 5 of our own systems are overclockers with air cooling.

I've installed the CPU cooler into my case every 3-6 months as I change CPU, motherboard, or case itelf, so installation is indeed important to me. However, just because a motherboard has to be removed from the case shouldn't be grounds for ranking installation. The best cooling options all need to be installed to the motherboard at least partially prior to going into a case. It usually takes me the better part of an hour to strip my motherboard from the case and put it back together (no removeable motherboard tray). I don't consider that a problem. What is a problem is some of these clips and retainer placements that can't be reached with the cooler in place.

I agree with most others here that all of these test figures are useless outside of a case. Case flow, and what a cooler does to add or detract from this flow, is probably the most critical component of any air cooled system.

Add to that the fact that some of these coolers are mounted in ways that would never fit in most cases. (EG: IFX-14's second cooler is mounted to the top: no case that I've ever used for a system would have room for this as my wires are tucked up there for aesthetics and to keep the air flow across the motherboard, front to back. These 2nd cooler fins would be destroyed during installation.)

Then all of the side blowing fans are set up to blow up or down, rather than out the back. Who in their right mind would do such a thing? Blowing top to bottom or visa versa doesn't do a thing for air flow in a case. Even if there is a top fan in the case, these coolers are placed directly under the PSU. (Very few cases have the PSU on the bottom and IMO they're all crap, not worth the money for shipping) The air is forced to move forward, against the natural air flow, or circulate back up or down to the rear exhaust. Moronic installation at best.

CPU coolers are designed to blow down for a very good reason. Especially with todays elaborate cooling heatpipes and fins on motherboards, the air is supposed to help cool the motherboard as the flow reflects off the CPU. Because these test systems are not installed in a case (which would have a higher ambient temp than sitting in the open, and be much more reactive to air flow of the case) none of the side blowing nor down blowing coolers take into consideration what they do to help the motherboard temps or air flow.

I think the idea for this review is great, test as many CPU coolers as possible. But the method leaves a lot to be desired. I could have run all 80 of these tests correctly, in a case, in a week, if that were my full time job. As is, it's not a real life comparison for CPU or motherboard temps.

Of course everyone and their brother will have a cooler that failed on this test and claim it's the best thing since sliced bread. And they may even have a good one, or some trick to setting it up. But chances are they're relying upon Everest or some other 3rd party software program to read sensors in the motherboard, which anyone with a brain knows can't be trusted to be accurate. And the idea of needing a "trick" to set it up is exactly why it should receive a poor score for setup.

TG is showing terrible technical knowledge and misleading results with these setups. I hope this is a one off and not a trend for TG.

Anyway, just one system builder and enthusiast's opinion.

 

[SpinachEater]

It is all relative. As long as each cooler was placed in the same conditions.
However, without a baseline, you can't take these numbers and compare them directly to real world systems.

This test is like pretend land with no magic rainbow leading back home. Dorothy is stuck in OZ, but she doesn't realize it.

Bingo. That is what I have been saying too. The theory of relativity knows no bounds. People are still trying to compare these temps to their home systems. It could be 30C in that room...WTF knows.

 

[Onus]

Context, context, context. As much as I agree with SpinachEater, CompTIA and Dark41, looked at in the context of its flaws, this series of articles will still be useful. It's like laughing at a NewEgg review because it is obvious what the drone did wrong; you still get an idea of the product out of it. If you see why a product did poorly, or very well, then you can figure how those results would or would not apply in your own situation. Remember too, a single sample is hardly representative. What if there was a speck of tissue left behind when the last TIM was cleaned off? What if, what if, what if? I still think it's worth something, even if it could have been better. Take it for what it is, just more information to consider when making a buying decision.

 

[LAN_deRf_HA]

I hope they do that newish Thermaltake V1... has very unique design and a very easy installation.

 

[Grimmy]

Context, context, context. As much as I agree with SpinachEater, CompTIA and Dark41, looked at in the context of its flaws, this series of articles will still be useful. It's like laughing at a NewEgg review because it is obvious what the drone did wrong; you still get an idea of the product out of it. If you see why a product did poorly, or very well, then you can figure how those results would or would not apply in your own situation. Remember too, a single sample is hardly representative. What if there was a speck of tissue left behind when the last TIM was cleaned off? What if, what if, what if? I still think it's worth something, even if it could have been better. Take it for what it is, just more information to consider when making a buying decision.

Pretty much understand what you are saying... but then, you have a link from that article to this forum to discuss.

Are we being to hard on the article?

So should we all just kick back and keep our thoughts to ourselves?

Does the article writer read what we talk about, and perhaps take considerations?

I know.. I'll just go away to my corner....

 

[Phrozt]

....

1) Tests were not done in a case
2) Idiot doesn't know the difference between "lose" and "loose"
3) "Author" sounds like a neurotic "hardcore" 12 year old board troll. 45% fail?? what the ghey...

This article failed EPICALLY within the first 2 pages.

And about the case/no case arguments. Coolers are designed SPECIFICALLY for use in a case. Hence the fact that most user manuals have phrases like, "mount so that air is moving out the back of the case."

What a retarded article and complete waste of time. Funniest part is where he insults all the previous THG cooler articles...

How this [strike]gigantic, nauseating, steaming pile of regurgitated horse shit[/strike] "copy" even passed an editor... and then made it to the front page is beyond me.

 

[TechnologyCoordinator]

I agree with others. The tests should have been done in case. They also should have used a stock HSF as a baseline.

 

[SpinachEater]

....

1) Tests were not done in a case
2) Idiot doesn't know the difference between "lose" and "loose"
3) "Author" sounds like a neurotic "hardcore" 12 year old board troll. 45% fail?? what the ghey...

This article failed EPICALLY within the first 2 pages.

And about the case/no case arguments. Coolers are designed SPECIFICALLY for use in a case. Hence the fact that most user manuals have phrases like, "mount so that air is moving out the back of the case."

What a retarded article and complete waste of time. Funniest part is where he insults all the previous THG cooler articles...

How this [strike]gigantic, nauseating, steaming pile of regurgitated horse shit[/strike] "copy" even passed an editor... and then made it to the front page is beyond me.

You are insulting someone’s intelligence for what is probably a type-o. I think you have opened yourself up to the same criticism by saying “what the ghey” That is technically a type-o on your part now or maybe you don’t know the difference between ghey and gay.

If you hold yourself to the same standards as what you held the author up to, that makes you an idiot now right?

 

[SpinachEater]

I agree with others. The tests should have been done in case. They also should have used a stock HSF as a baseline.

Wait, you don't run your board on a wood plank out in the open?

 

[Grimmy]

What board would be better?

Cherry?
Mahogany?
Plywood?
Oak?

Hmmmm..

 

[Phrozt]

You are insulting someone%u2019s intelligence for what is probably a type-o. I think you have opened yourself up to the same criticism by saying %u201Cwhat the ghey%u201D That is technically a type-o on your part now or maybe you don%u2019t know the difference between ghey and gay.

If you hold yourself to the same standards as what you held the author up to, that makes you an idiot now right?

This is a published article. You shouldn't have any typos... much less one IN THE TITLE!!!

And "ghey" is not a typo. It's either an internet meme or a way to avoid the filter. Obviously, "h" and "e" are no where close to "a," as well as the fact that "he" is two letters while "a" is one. Maybe you should co-author the next article in the series...

Regardless, as I and most others on this thread have pointed out, there are a multitude of reasons to call this "author" an idiot. You don't like the fact that he tries to use big words but fails at small words? Ok... pick from the plethora of other problems he has; like the fact that he doesn't know that you order graphs from best to worst... not necessarily highest number to lowest (temps/speeds/dbS...).

 

[Phrozt]

I agree with others. The tests should have been done in case. They also should have used a stock HSF as a baseline.

Why use a base line that is available to everyone who purchases the retail version of the CPU when you can use another variable... as a constant!!

 

[CompTIA_Rep]

Because, they were not testing inside a case, or in conditions equal to everyone's situation.

I want to know how much better these coolers are to my stock fan... Is it worth the $80 investment? I have no idea, so article fails to answer a simple question...


I'm sure the author is reading this and will correct the article.... Please fix it, please with cherry whipped cream in a standard antec case on top?

 

[Grimmy]

Whelp... Seems as though the majority of those HS do say in the spec for a Quad, except a couple, which I only read C2D series in the specs. But I was looking around for Zaward Vivo. And what ya know, I found a benchmark for it:

Zaward VIVO PCJ004 Exposed Copper Heatpipe CPU Cooler

Now it tests the E6600, not the quad, but look at how this testbed is setup:

Test System

*
Motherboard: Gigabyte GA-P35C-DS3R (Intel P35 chipset) with F4G BIOS
*
System Memory: Crucial PC2-8500 CL5 Ballistix DDR2 RAM
*
Processor: Intel E6600 Core 2 Duo 2.2GHz operating at multiple speeds
*
Cooling: Xigmatek HDT-S1283 Exposed Copper Heatpipe CPU Cooler
*
Video: Foxconn GeForce 8800 GTS 640MB overclocked to 600MHz Core / 900MHz RAM
*
Audio: HT Omega CLARO7.1 C-Media 8788 Oxygen Chipset
*
Hard Disk 1: Western Digital WD360GD 10,000 RPM SATA
*
Optical Drive: Toshiba Samsung SH-S203 SATA 20x DVD R/W
*
Enclosure: Lian Li PC-B20A ATX Case with 2x120mm and 1x90mm cooling fans
*
PSU: ePower Technology EP-1200P10 xScale 1200W PSU
*
Operating System: Windows XP Professional SP-2 (optimized to 16 processes at idle)

What ya know... Lian Li PC-B20A ATX case.

And the graphs indicating ambient temps.. and and.. a STOCK HS BASELINE!!!

Owell... I think I'll go to bed now...