PDA

View Full Version : Zalman's got an AS beater?



Xavior
06-26-2006, 05:17 AM
www.zalman.co.kr/eng/product/view.asp?idx=198

http://www.hi-techreviews.com/Joomla/content/view/170/27/1/2/

http://www.bcchardware.com/index.php?option=com_content&task=view&id=2688&Itemid=40

Anyone tried this stuff before?

granulf
06-26-2006, 05:29 AM
Is that nail painting or something?? :stick:

Smalltimer
06-26-2006, 05:42 AM
I wonder if it beats liquid metal?
I just applied some to my setup and I still haven't picked up my jaw off the floor! - my temps dropped 13c under load :slobber:

Marci
06-26-2006, 05:55 AM
Sounds more like you had a duff CPU block mount prior to applying... no thermal paste can account for a drop that large.

Smalltimer
06-26-2006, 05:59 AM
Nah, I mounted the block plenty of times.
Originally I had:
31c idle and 48 under load with my silent watercooling setup.

Once I whiped and cleaned the AS3 off my block and core and replaced it with a thin film of liquid metal and my temps changed to:
28c idle and 35c under load

it has remained at 35c under load despite raising the voltages also. I moved it up to 1.48v until the meter showed an increase of 2c under load. Otherwise it sits hapilly at 35c

The stuff is simply amazing!

granulf
06-26-2006, 06:03 AM
I'd have to agree with Marci, 13c can't be right! I would love to test it myself, but I haven't found anyone selling it yet...(the liquid metal thing that is)

Marci
06-26-2006, 06:05 AM
Not a debatable point. Something else accounts for the difference. A thermal compound and it's application CANNOT account for 13 deg C load temp drop alone.

Smalltimer
06-26-2006, 06:07 AM
Well those are my results. - sceptism is a moot point since this is what I got out of it.
You can find the stuff here: http://www.performance-pcs.com/catalog/index.php?main_page=product_info&products_id=4373

Marci
06-26-2006, 06:20 AM
But you cannot conclusively account your findings to a variance in TIM and TIM alone. Therefore, your findings are not conclusive, empirical, proven or otherwise especially when it takes a paper of this calibre (http://www.electronics-cooling.com/html/2003_november_a2.html) just to outline the difficulties of measuring the performance of a TIM. Don't tell me you've suddenly cracked how to do it when (the world over) scientists and ThermoDynamics engineers can't manage it accurately or agreeably.

http://www.zalman.co.kr/images/product_img/etc/stg1_graph.gif

Zalman's OWN testing only shows a 2 deg C difference between "High Performance Compounds" (ie: Silver based compounds) and their new stuff... and a 5 deg C drop between "regular compounds" (ie: silicon based white goop) and their new stuff... your old compound applications / block mounts were flawed. Your current mount / application is not. Yes, Zalman's TIM exhibited a drop, but it is NOT the sole cause of a 13 deg C drop, and end-users would be foolish to expect such.

Syranide
06-26-2006, 06:23 AM
Well those are my results. - sceptism is a moot point since this is what I got out of it.
You can find the stuff here: http://www.performance-pcs.com/catalog/index.php?main_page=product_info&products_id=4373

Yes and look at the graphs... I don't see a 13C drop, I see a 6C drop at _best_ in their tests.
I must concur with Marci, 13C drop unless you are talking temperatues of 80C is not plausible, for common computers a 13C drop would mean that you are now below room temperature first of all, almost even for my computer which is running really hot because I want it silent.

I'm guessing as marci, your previous setup was probably a little skew or perhaps some AS was in contact with the area around the core. 13C drop _is not_ possible for computers at sane temperatures, the paste only improves heat transfer, it does not dissipate heat itself.

Smalltimer
06-26-2006, 06:25 AM
Well I didn't think it was a such a big deal, but after reading the importance placed on my accomplishments your plight has driven me to publish a paper and claim a spot for the next annual NP. :toast:

vudoodoodoo
06-26-2006, 06:27 AM
Is that nail painting or something?? :stick:

That's what I thought lol!

Smalltimer
06-26-2006, 06:31 AM
Yes and look at the graphs... I don't see a 13C drop, I see a 6C drop at _best_ in their tests.
I must concur with Marci, 13C drop unless you are talking temperatues of 80C is not plausible, for common computers a 13C drop would mean that you are now below room temperature first of all, almost even for my computer which is running really hot because I want it silent.

I'm guessing as marci, your previous setup was probably a little skew or perhaps some AS was in contact with the area around the core. 13C drop _is not_ possible for computers at sane temperatures, the paste only improves heat transfer, it does not dissipate heat itself.
The only thing I can attribute to my results is that the AS3 was not fully set, or it may not of been fully optimized in terms of amount and spread. However I should point out that this is a common occurence in many cases and that the use of liquid metal eliminates these problems.

So I would say that the liquid metal is responsible for the 13c drop on my setup when compared to the alternative brand.

granulf
06-26-2006, 06:36 AM
But this review surely can't be right?? 8c better than ceramique? :stick:
http://www.bcchardware.com/index.php?full=1&set_albumName=ZM-STG1&id=chart&option=com_gallery&Itemid=&include=view_photo.php

Marci
06-26-2006, 06:44 AM
None of the companies conducting these reviews have the equipment to conduct them and produce empirical results.

A TIM's performance cannot be measured on PC Hardware. Fullstop. All these reviews are pointless other than to create a buzz regarding a product. That is all. Everyone will draw their own conclusions after purchase regardless, but misleading comments and reviews will merely serve to bring about a lot of folks pissed at that product for not living up to it's claims. Solution - make no claims unless they are empirical. The same held true for Waterblock reviews, Radiator reviews, Pump reviews... if it's for use in a cooling system, then generally it's performance cannot be reviewed and summarised on a PC or using PC Hardware due to the inherant variances in that hardware. Ignore any reviews you read on Thermal Compounds unless the testing methodology is scientifically sound.

WeStSiDePLaYa
06-26-2006, 06:49 AM
i think you are all forgetting the swiss and cheddar cheese TIM results.

Marci
06-26-2006, 06:51 AM
Toothpaste beat everything didn't it?? Well reminded - a link to that article may be invaluable here... http://www.dansdata.com/goop.htm


Firing up my Minty Fresh Thermal Transfer Experiment produced, rather hilariously, an excellent score - 0.47°C/W.

That's right - Toothpaste Superior To Arctic Silver 3! Film At Eleven!

.sentinel
06-26-2006, 08:13 AM
Over at OCForums SSS is testing it and trying to talk to Zalman.