Intel P965: The Double Mint Twins Gone Wild
by Gary Key on November 9, 2006 8:00 PM EST- Posted in
- Motherboards
General System Performance
We devised a script that would compress our standard test folder consisting of 444 files, ten subfolders, and 602MB worth of data, convert a 137MB High Definition QuickTime movie clip to a 37MB MPEG-4 format, play back the first two chapters of Office Space with PowerDVD, and run our AVG anti-virus program in the background. We stop the script when the file compression and video conversion are complete. This is a very taxing script for the CPU, Memory, and Storage subsystem. We also found it to be a good indicator of system stability during our overclocking testing.
The performance difference basically mirrors our PCMark 2005 tests with the ASUS P965 boards scoring about 3% better overall in our multitasking test. We also see our Biostar 965PT board nipping on the heels of the ASUS boards and showing the results in PCMark 2005 were repeatable in this multitasking test. Both Gigabyte boards finish last which is surprising given their strength in our other benchmarks. As with our PCMark 2005 test both Gigabyte boards had issues with the virus scanning tests. We noticed in this test that our AVG anti-virus scan would stutter severely at various times during the tests.
PCMark 2005, together with two benchmarks that use rendering to test system performance - Cinebench 9.5 and POV-Ray 3.6 - has replaced Winstones for testing general performance. The Cinebench 9.5 and POV-RAY 3.6 benchmarks both heavily stress the CPU subsystem while performing graphics modeling and rendering. We utilize the standard benchmark demos in each program along with their default settings. Cinebench 9.5 features two different benchmarks with one test utilizing a single core and the second test using the power of multiple cores to render the benchmark image. We utilize the dual core test for our results.
We see our collection of Biostar and Gigabyte boards taking top honors in these benchmarks although all of the scores are extremely close. We continue to see a pattern where the P965 boards at stock speeds are scoring slightly better than our 975X and NVIDIA 570SLI offerings.
We devised a script that would compress our standard test folder consisting of 444 files, ten subfolders, and 602MB worth of data, convert a 137MB High Definition QuickTime movie clip to a 37MB MPEG-4 format, play back the first two chapters of Office Space with PowerDVD, and run our AVG anti-virus program in the background. We stop the script when the file compression and video conversion are complete. This is a very taxing script for the CPU, Memory, and Storage subsystem. We also found it to be a good indicator of system stability during our overclocking testing.
The performance difference basically mirrors our PCMark 2005 tests with the ASUS P965 boards scoring about 3% better overall in our multitasking test. We also see our Biostar 965PT board nipping on the heels of the ASUS boards and showing the results in PCMark 2005 were repeatable in this multitasking test. Both Gigabyte boards finish last which is surprising given their strength in our other benchmarks. As with our PCMark 2005 test both Gigabyte boards had issues with the virus scanning tests. We noticed in this test that our AVG anti-virus scan would stutter severely at various times during the tests.
PCMark 2005, together with two benchmarks that use rendering to test system performance - Cinebench 9.5 and POV-Ray 3.6 - has replaced Winstones for testing general performance. The Cinebench 9.5 and POV-RAY 3.6 benchmarks both heavily stress the CPU subsystem while performing graphics modeling and rendering. We utilize the standard benchmark demos in each program along with their default settings. Cinebench 9.5 features two different benchmarks with one test utilizing a single core and the second test using the power of multiple cores to render the benchmark image. We utilize the dual core test for our results.
We see our collection of Biostar and Gigabyte boards taking top honors in these benchmarks although all of the scores are extremely close. We continue to see a pattern where the P965 boards at stock speeds are scoring slightly better than our 975X and NVIDIA 570SLI offerings.
Facebook
Twitter
RSS
23 Comments
View All Comments
Gary Key - Thursday, November 9, 2006 - link
It is coming. We had to retest all of the high-end boards with CrossFire capability since the official 6.10 drivers we used generated measurable differences (sometimes better than 7%) in several games compared to the early beta 6.10 drivers. We did not see this issue with our single card testing.Sho - Thursday, November 9, 2006 - link
Ah, ok :). Rock on.JarredWalton - Friday, November 10, 2006 - link
Gary also neglected to tell you about his latest hard drive "testing", in the which he lost many of his in-the-work articles. I keep telling him that he shouldn't stress test his own hardware, but does he listen? Noooo! I really ought to run RAID 1 or start do more frequent backups, come to think of it....