Gigabyte GA-965P-DS3: Lab Testing Update
by Gary Key on January 4, 2007 9:00 AM EST- Posted in
- Motherboards
Overclocking
Our combination of components resulted in a maximum stable overclock of 7x505 FSB with the F9 BIOS. This is an improvement of 50FSB over the F6 BIOS and 30FSB over the F8 BIOS release. We were also able to boot and enter Windows XP at 7x505 with our memory set at 4-4-4-12, but we had to change our memory settings to 5-5-5-15 for the system to complete the entire benchmark suite (including eight hours of dual Prime95). The performance penalty from running the relaxed memory latencies was minimal in our benchmark testing but still resulted in a one to two percent decrease in several of the benchmarks. Overall, the improvements in this BIOS release are somewhat amazing considering our previous test results. However, these results mean that the DS3 just now equals other boards in the same price range. We still value stability and overall system performance over high FSB rates that will only be used by a minority of users but this board now offers the best of both worlds to a certain degree.
We optimized the MCH (+.2V) , FSB (+.2V), and Memory (+.5V) voltages and were able to run the majority of our benchmark test suite at the 7X520 FSB setting but at very relaxed 5-6-5-18 settings. In order to complete the dual Prime95 testing we had to further relax our memory timings to 5-7-6-20, which resulted in the majority of our benchmark scores being worse than our 7x505 results. Our ASUS P5B-E 1.02G motherboard will run our entire benchmark test suite at 4-4-4-12 memory timings at the same 7x520FSB,so we feel Gigabyte still has some memory and MCH tuning to complete in the next BIOS release. Considering the fact that Gigabyte has already stated the motherboard needs a resistor change to exceed the 520FSB level, we believe the BIOS engineers have maxed out the board from an overclocking viewpoint and should concentrate on memory performance now. We have the revision 2 board in-house for testing currently and will have results up in the near future as this board utilizes the same BIOS but has an optimized electrical layout for improved performance.
| Gigabyte GA-965P-DS3 Overclocking Testbed |
|
| Processor: | Intel Core 2 Duo E6300 Dual Core, 1.86GHz, 2MB Unified Cache 1066FSB, 7x Multiplier |
| CPU Voltage: | 1.5000V (default 1.3250V) |
| Cooling: | Scythe Infinity Air Cooling |
| Power Supply: | OCZ GameXStream 700W |
| Memory: | Geil PC2-6400 800MHz Plus (2x1GB- GX22GB6400PDC) (Micron Memory Chips) |
| Video Cards: | 1 x MSI X1950XTX |
| Hard Drive: | Seagate 320GB 7200RPM SATA2 16MB Buffer |
| Case: | Cooler Master CM Stacker 830 |
| Maximum CPU OC: (Standard Ratio) |
505x7 (5-5-5-15, 1:1, 2.3V), CPU 1.5000V, C2 Stepping 3535MHz (+89%) |
| . | |
 |
| Click to enlarge |
Our combination of components resulted in a maximum stable overclock of 7x505 FSB with the F9 BIOS. This is an improvement of 50FSB over the F6 BIOS and 30FSB over the F8 BIOS release. We were also able to boot and enter Windows XP at 7x505 with our memory set at 4-4-4-12, but we had to change our memory settings to 5-5-5-15 for the system to complete the entire benchmark suite (including eight hours of dual Prime95). The performance penalty from running the relaxed memory latencies was minimal in our benchmark testing but still resulted in a one to two percent decrease in several of the benchmarks. Overall, the improvements in this BIOS release are somewhat amazing considering our previous test results. However, these results mean that the DS3 just now equals other boards in the same price range. We still value stability and overall system performance over high FSB rates that will only be used by a minority of users but this board now offers the best of both worlds to a certain degree.
 |
| Click to enlarge |
We optimized the MCH (+.2V) , FSB (+.2V), and Memory (+.5V) voltages and were able to run the majority of our benchmark test suite at the 7X520 FSB setting but at very relaxed 5-6-5-18 settings. In order to complete the dual Prime95 testing we had to further relax our memory timings to 5-7-6-20, which resulted in the majority of our benchmark scores being worse than our 7x505 results. Our ASUS P5B-E 1.02G motherboard will run our entire benchmark test suite at 4-4-4-12 memory timings at the same 7x520FSB,so we feel Gigabyte still has some memory and MCH tuning to complete in the next BIOS release. Considering the fact that Gigabyte has already stated the motherboard needs a resistor change to exceed the 520FSB level, we believe the BIOS engineers have maxed out the board from an overclocking viewpoint and should concentrate on memory performance now. We have the revision 2 board in-house for testing currently and will have results up in the near future as this board utilizes the same BIOS but has an optimized electrical layout for improved performance.
Facebook
Twitter
RSS
14 Comments
View All Comments
imaheadcase - Thursday, January 4, 2007 - link
Can you tell us what kinda of temps you were getting on the overclocks you did? Those are equally worth mentioning, since some won't go over a certain temp vs others don't mind.Beenthere - Thursday, January 4, 2007 - link
SOS, DD...reaching BIOS F9 and still having not sorted out the performance and stability of a mobo is simply unacceptable. This is exactly what is wrong with the current PC mobo industry. Rush half-baked crap out the door to rave review, suck up the money and move on to the next trick-of-the-week, POS to-die-for mobo.
As long as people buy these defective products, that is exactly what manufacturers will ship. There is zero incentive to fix a piss poor product if naive, gullible consumers will buy garbage and pay premium prices for the abuse.
Genx87 - Monday, January 8, 2007 - link
Not sure how a board that cant run out of its specs can be considered defective. But my DS3 has been pretty stable. But oh yeah, it is running at stock speeds as well. Must be defective because it hasnt crashed from overheating :Dimaheadcase - Thursday, January 4, 2007 - link
The DS3 bios update have been pretty trivial to the normal user. Its not a "defective product" like you claim, I for one am tired of people thinking a motherboard "sucks" or is terrible because they can't overclock.The DS3 is the perfect motherboard as it is.