NVIDIA nForce4 Ultra: Biostar's Performance Surprise
by Gary Key on February 16, 2006 12:05 AM EST- Posted in
- Motherboards
Disk Controller Performance
With the variety of disk drive benchmarks available, we needed a means of comparing the true performance of the wide selection of controllers. The logical choice was Anand's storage benchmark first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs. the World. The iPeak test was designed to measure "pure" hard disk performance, and in this case, we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive.
We played back Anand's raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's iPEAK utility was then used to play back the trace file of all IO operations that took place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance differences to the controllers that we were testing, we used the Maxtor MaXLine III 7L300S0 300GB 7200 RPM SATA drive in all tests . The drive was formatted before each test run and a composite average of 5 tests on each controller interface was tabulated in order to ensure consistency in the benchmark.
iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number is meaningless as far as hard disk performance is concerned, as it is just the number of IO operations completed in a second. However, the scores are useful for comparing "pure" performance of the storage controllers in this case.
With the variety of disk drive benchmarks available, we needed a means of comparing the true performance of the wide selection of controllers. The logical choice was Anand's storage benchmark first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs. the World. The iPeak test was designed to measure "pure" hard disk performance, and in this case, we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive.
We played back Anand's raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's iPEAK utility was then used to play back the trace file of all IO operations that took place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance differences to the controllers that we were testing, we used the Maxtor MaXLine III 7L300S0 300GB 7200 RPM SATA drive in all tests . The drive was formatted before each test run and a composite average of 5 tests on each controller interface was tabulated in order to ensure consistency in the benchmark.
iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number is meaningless as far as hard disk performance is concerned, as it is just the number of IO operations completed in a second. However, the scores are useful for comparing "pure" performance of the storage controllers in this case.
Facebook
Twitter
RSS
31 Comments
View All Comments
jamesbond007 - Thursday, February 16, 2006 - link
Haha! Way to go, Gary. You have a fan base! =PGary Key - Thursday, February 16, 2006 - link
Thanks for kind words everyone. I will post a short update to this article in a couple of days as the new bios results are looking promising in resolving some overclock and bios lockup issues.drewintheav - Thursday, February 16, 2006 - link
Gary is awesome! :)Zebo - Thursday, February 16, 2006 - link
I love Gary can't we get him writing articles people will read? Intel/biostar - common.. you'll get 1000 page hits max and 3/4 of them are because Gary wrote it!:PGooger - Thursday, February 16, 2006 - link
Not too bad if you want a P4, but for me I am avoiding nVIDIA Chipsets except when it comes to AMD products. Go Go ULi!DigitalFreak - Thursday, February 16, 2006 - link
Uh, you mean Go Go Nvidia, since they own ULi now...Googer - Thursday, February 16, 2006 - link
Finaly Intel Gets Hypertransport on their chips, like it or not HTT probably is becoming a standard that Intel might have to adopt sooner or later.DigitalFreak - Thursday, February 16, 2006 - link
This has nothing to do with Intel. Nvidia uses HT to communicate between their north and south bridges. They've done it with all their Intel chipsets so far.Googer - Thursday, February 16, 2006 - link
Since the noth bridge has HTT, in theory you could connect an nVIDIA based nFORCE north bridge to a ULi or nVIDIA AMD north bridge and have one of several things:1) A dual CPU system- One Intel Pentium M/4 and One AMD 64 CPU running on the same motherboard simultaniously. The OS might need to be re-written so that multi-threaded applications only use one processor. Linux prehaps?
2) AMD 64 Could get Quad Channel RAM higher.
3) You could ADD a ULi M1567 Southbridge to get True AGP with that PCI-express SLI.
4) You could possibly mix and match chipsets. VIA K8T8xx with one of AMD's north/south bridges and an nFORCE Intel Editon.
You could possibly Connect the AMD 64 Directly (using it's own HTT link) in to the the P4 north bridge with no need to use the chipset designed for the A64.
HTT on Intel means a whole new world of possibilites!
Furen - Thursday, February 16, 2006 - link
Huh? How is Intel getting hypertransport on its chips? HT is a standard but I dont think Intel will ever adopt it because of its pride, more than anything else. It truly doesn't matter though, since HT is just a data transport and using any other data transport gives you the same results as long as it is used in a similar configuration.