iXBT Labs :: Computer Hardware In Detail

Platform

Video

Multimedia

Mobile

Other

Refined Cooling for Video Cards

ShareThis

Part 3. Test results, Conclusions


Part 1. The stock cooler of VPU ATI Radeon X800XT, Arctic Cooling ATI Silencer 4, Cooler Master CoolViva, and GlacialTech NorthPole 1000

Part 2. Zalman ZM80D-HP, VF700-AlCu, and VF700-Cu

In order to test thermal efficiency of the VPU coolers under review, we have followed practically the same method principles that are used in our CPU cooler tests. The primary data used for the consequent calculation of thermal resistance is also temperature readings of the built-in thermal diode, only the thermal source is different (now it's VPU ATI X800), the base platform (MSI RX800XT Platinum-VTD256 video card), and a set of test applications.

We have also inspected thermal operating modes of video memory - we have measured the temperature on U58 (front side) and U59 (back side) memory chips with calibrated electronic thermometers, ensuring engineering accuracy of measurement with the consequent statistic analysis of the test data.

The testbed has the following configuration:

  • Motherboard: Fujitsu Siemens Computers D1688
  • Processor: Intel Pentium 4 3.2 GHz (Prescott, HT Technology)
  • Video card: MSI RX800XT Platinum-VTD256
  • OS: Microsoft Windows XP

To simulate the close to maximum VPU thermal load, we used 3DMark03 test package — Battle of Proxycon, Troll's Lair, and Mother Nature scenes in cycle (according to our readings, these scenes reflect the typical thermal load on video cards in modern games). Each VPU cooler was tested in four 60-minute stages, the environment temperature (innards of the half-open test box) was kept at 30°C. And finally, we used the RivaTuner utility to monitor the video chip temperature.

Diagram 1. Temperatures, VPU

Notes
Each cooler was tested with the stock thermal interface
The diagram contains complex results


Diagram 2. Temperatures, video memory



Diagram 3. Thermal resistance

Note
Thermal resistance θja is defined as the relation
θja = (Tj — Ta)/Ph, where Tj is the VPU temperature, Ta is the environment temperature (it's 30°C in this case), Ph is the VPU thermal capacity (in this case it's 33 W).

Finally, at the end of this article we publish the noise measurement results (the method of testing is described in the article Noise characteristics of coolers and the noise measurement method) as well as the efficiency/noise rating of coolers.

Diagram 4. Noise characteristics

Note: Background noise level 18 dBA


Diagram 5. Efficiency/noise ratio

Note
The efficiency/noise ratio (ENR) is calculated as:

ENR = DM*(Rt/TC)/(NL/Rn), where

Rt — reference temperature (it corresponds to the reference thermal resistance θja of the cooling system - 40°C), TC — VPU temperature with the operating cooling system, Rn — reference noise (the reference noise level is 25 dBA), NL — noise level, generated by the cooling system, DM — denominate multiplier (10).

There seems to be no need in any comments here. Let's draw the bottom line!

Conclusions

Probably the most important conclusion for today runs that highly-effective and ergonomic cooling of video cards is a tangible matter. It can be achieved in different ways (it depends on users' preferences and the sum of money), but the results will be positive and immediate.

I'd like to particularly mention GlacialTech NorthPole 1000 — this cooler demonstrates excellent efficiency. By the combination of consumer properties (balanced thermal efficiency and noise characteristics, very simple installation, and low price) it seems to us the most interesting solution for advanced cooling of middle and high end VPU ATI video cards. Arctic Cooling ATI Silencer 4 is also an excellent cooler, it successfully combines high cooling efficiency and the ergonomic noise level. The pair of new models from Zalman (VF700-AlCu and VF700-Cu) fairs quite well — these coolers are notable for excellent universality (they support almost all modern video cards) and can cool video memory in the best way. But Zalman ZM80D-HP and Cooler Master CoolViva turn out in the group of outsiders: though they are technically advanced in terms of construction, they still cannot provide decent thermal efficiency.

So, having examined our today's contenders, we decided that Arctic Cooling ATI Silencer 4 and GlacialTech NorthPole 1000 merit the Original Design award.




And Zalman ZM80D-HP, notable for its rich bundle, gets the Excellent Package award.




That's all for today.

Vitali Crinitsin (vit@ixbt.com)
June 9, 2005.


Article navigation:



  Most Popular Reviews More   RSS  

59 CPU Roundup

Summarizing results obtained with CPU Test Method v4.0.
Mar 12, 2010 · General Platform

Intel H55, H57 Chipsets

Integrated chipsets for lower-end Nehalem processors.
Jan 18, 2010 · Chipsets

E-MU 0404 USB

A top-class external audio interface.
Nov 23, 2006 · ProAudio

How CPU Features Affect CPU Performance, Part 7

Intel Core i7, memory subsystem.
Feb 18, 2010 · General Platform

AMD Athlon II X3 425, 435 Processors

Three cores for the price of two.
Oct 30, 2009 · Processors: AMD
  Latest Reviews More   RSS  

Intel DP55WB "Whitesburg" Motherboard

One of the cheapest LGA1156 solutions.
Mar 19, 2010 · Motherboards

AMD 890GX Chipset Overview

Newer integrated graphics core, Dual Graphics, SATA 3.0 support.
Mar 16, 2010 · Chipsets

ATI Radeon HD 5850/5870/5970 Graphics Cards

From AFOX, Manli, Sapphire, XFX.
Mar 15, 2010 · Video cards: ATI GPUs

59 CPU Roundup

Summarizing results obtained with CPU Test Method v4.0.
Mar 12, 2010 · General Platform

MSI 770-C35 Motherboard

A solution for conservative users.
Mar 11, 2010 · Motherboards
  Latest News More   RSS  
  Useful Links Get listed  

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ·  Compare Prices  ||  Feedback  ·  Advertise at iXBT Labs  ·  About us  ·  Affiliates  ·  Forum


21

Copyright © Byrds Research & Publishing, Ltd., 1997—2010. All rights reserved.