Gigabyte SINXP1394 (SiS 655): More Dual Channel DDR for P4
by Evan Lieb on February 11, 2003 3:04 AM EST- Posted in
- Motherboards
Gigabyte SINXP1394: Stress Testing
We managed to test the SINXP1394 in several different areas and configurations, including:
1. Chipset and motherboard stress testing was conducted by running the FSB at 164MHz.
2. Memory stress testing was conducted by running RAM at 333MHz and 400MHz in dual DDR operation at the most aggressive timings possible.
Front Side Bus Stress Test Results:
As usual we ran a large load of stress tests and benchmarks to ensure the SINXP1394 was absolutely stable at each overclocked FSB speed we experimented with. We ran our usual array of stress tests, including Prime95 torture tests, which were run in the background for a total of 48 hours. Just as we did with Gigabyte’s 8INXP, we ran lots of other tasks such as data compression, various DX8 games, and light apps like Word and Excel while Prime95 was running. Finally, we reran our entire benchmark suite, which includes Sysmark 2002, Quake3 Arena, Jedi Knight II, Unreal Tournament 2003, SPECviewperf 7.0, and XMPEG. In the end, 164MHz FSB (163.86MHz FSB) was the highest overclock we were able to achieve without encountering any reliability problems.
Memory Stress Test Results:
In our experience, the best motherboards on the market are usually the ones that can handle high memory speeds, not just high FSB speeds. As far as actual memory speeds are concerned, some the best motherboards are able to run memory at 333MHz with all memory banks filled. Here were the timings we were able to achieve in that scenario:
Stable Dual DDR333 Timings |
|
Clock
Speed:
|
166MHz
|
Timing
Mode:
|
N/A
|
CAS
Latency:
|
2.0
|
Bank
Interleave:
|
N/A
|
Precharge to Active:
|
2T
|
Active
to Precharge:
|
6T
|
Active
to CMD:
|
2T
|
Command Rate:
|
N/A
|
Interestingly enough we were unable to achieve CAS2/2T/5T/2T at any memory speed with any amount of memory banks filled. That is, even with only one memory module installed and running at 266MHz, CAS2/2T/6T/2T was the most aggressive timings we were able to reach. We experienced this “issue” with three different BIOSes, including M01, M04 and M05. The M05 BIOS was definitely the fastest BIOS out of the three, but did not remedy this odd timing “issue.” We put the word “issue” in quotes because the performance delta between an Active to Precharge setting of 5T and 6T is very small. However, we note this experience because we rarely encounter a motherboard that can’t do CAS2/2T/5T/2T yet can do CAS2/2T/6T/2T.
In the coming months you’ll be hearing even more ruckus about DDR400, mostly from Intel and their partners. This is obviously because Intel will be officially supporting DDR400. Intel is already working with JEDEC and other industry bodies and companies on ratifying a DDR400 standard. Anyway, Gigabyte seems confident that their SINXP1394 can support DDR400 quite well, and that indeed held true in our tests. Here were the timings we were able to achieve with four identical Corsair XMS PC3200 modules running at 400MHz:
Stable Dual DDR400 Timings |
|
Clock
Speed:
|
200MHz
|
Timing
Mode:
|
N/A
|
CAS
Latency:
|
2.0
|
Bank
Interleave:
|
N/A
|
Precharge to Active:
|
2T
|
Active
to Precharge:
|
6T
|
Active
to CMD:
|
2T
|
Command Rate:
|
N/A
|
Being able to run CAS2/2T/6T/2T with four banks filled and running at 400MHz is no small feat by any means. MSI’s 655 Max board was able to achieve slightly better, but both Gigabyte and MSI’s memory performance at 400MHz is simply excellent.
As usual, we ran several memory stress tests and general apps to make sure all these timings were stable. We started off by running Prime95 torture tests; a grand total of 24 hours of Prime95 was successfully run at the timings listed in the above charts. We also ran Sciencemark (memory tests only) and Super Pi. Neither stress test was able to bring the SINXP1394 to its knees.
0 Comments
View All Comments