After careful consideration I have decided to transfer all hardware review activities to a new domain. I purchased Hardwareasylum.com in 2012 and have been working hard to build a new and improved Ninjalane on that domain. If you are reading this you have reached one of the archived articles, news, projects and/or reviews that were left behind during the site migration.
Please update your bookmarks and be sure to visit the new and improved Ninjalane at Hardwareasylum.com
Introducing the Multi-GPU Index
Author: Dennis Garcia
Published: Thursday, December 02, 2010
Multi GPU Testing
While video games are the primary reasons that drive the availability of Multi-GPU configurations, the realistic limit for the mainstream public is no more than 2 cards. Any more than 2 and you run into a wide variety of issues such as motherboard support, power supply limitations, drivers, particular card brand support, and most importantly cost. The cost of running three and four cards increases considerably after the second card threshold.
So why bother? Two reasons really. Competitive benchmarking and bragging rights. Aside from manufacture driven overclocking competitions like GOOC and MOA, the main force behind competitive benchmarking is HWBot. HWBot (http://www.hwbot.org) is a website dedicated to collecting and maintaining benchmark results for the purpose of ranking those scores against other submissions. As with any competition, the person with the best score overall wins. It may not sound like much, but the whole process is rather addicting, especially when you set out to overtake a position and do whatever it takes to make it happen.
We are going to run a couple of popular benchmarks found on HWBot and see how the different motherboard configurations affect the scores. To help keep the results as fair as possible each of the systems were tuned to have similar CPU scores (unless otherwise noted below)
So why bother? Two reasons really. Competitive benchmarking and bragging rights. Aside from manufacture driven overclocking competitions like GOOC and MOA, the main force behind competitive benchmarking is HWBot. HWBot (http://www.hwbot.org) is a website dedicated to collecting and maintaining benchmark results for the purpose of ranking those scores against other submissions. As with any competition, the person with the best score overall wins. It may not sound like much, but the whole process is rather addicting, especially when you set out to overtake a position and do whatever it takes to make it happen.
We are going to run a couple of popular benchmarks found on HWBot and see how the different motherboard configurations affect the scores. To help keep the results as fair as possible each of the systems were tuned to have similar CPU scores (unless otherwise noted below)
The system(s) as they were tested:
All of the tests were run using the following:
Western Digital Raptor X 150BG SerialATA 10000RPM
LG 20x Super Multi GSA-H55L
Corsair HX 850 Watt PSU
Windows 7 Ultimate 64bit
Drivers and Cards:
Radeon HD 4850 512MB Crossfire (Catalyst 10.10)
Western Digital Raptor X 150BG SerialATA 10000RPM
LG 20x Super Multi GSA-H55L
Corsair HX 850 Watt PSU
Windows 7 Ultimate 64bit
Drivers and Cards:
Radeon HD 4850 512MB Crossfire (Catalyst 10.10)
CPU Configuration
Core i7 920 @ 3.0Ghz – Used on all X58 based motherboards
Core i5 750 @ 4.0Ghz – Used on all P55 based motherboards
Core 2 Duo E6400 @ 3.4Ghz – Used on all X48 based motherboards (inaccurate cpu score)
Phenom II X6 1090T @ 3.6Ghz – Used on 890FX based motherboards
Phenom II X3 720 @ 2.8Ghz – Used on 790FXB (inaccurate cpu score)
Opteron 148 @ 2.2Ghz – Used on NFPIK8AA (inaccurate cpu score)
Core i5 750 @ 4.0Ghz – Used on all P55 based motherboards
Core 2 Duo E6400 @ 3.4Ghz – Used on all X48 based motherboards (inaccurate cpu score)
Phenom II X6 1090T @ 3.6Ghz – Used on 890FX based motherboards
Phenom II X3 720 @ 2.8Ghz – Used on 790FXB (inaccurate cpu score)
Opteron 148 @ 2.2Ghz – Used on NFPIK8AA (inaccurate cpu score)
