AMD Test Results: Overclocked
We then looked at the highest overclocked speed that we could reach with our early 3200+ and OCZ 3700EB. While a stable 275x7 was possible, a higher processor speed was possible at a stable 268x8. As demonstrated in the review of the AOpen AK89 Max, the more recent 3200+ processors have much greater overhead and overclock much further.While this is a modest processor overclock, the memory is running at a significantly higher speed as a result of the lower multipliers. In fact, we reached DDR536 to DDR550 (268 to 275) with the OCZ 3700EB on the Chaintech nForce3-250 compared to the DDR524 maximum reached on the Intel 875 using the same memory. In fairness, we are running single-channel on the Athlon 64 compared to Dual-Channel on the P4. Dual-Channel on Athlon 64 arrives next month with the new Socket 939 for Athlon 64.
Using the 2142MHz as the standard overclocked speed, we compared performance at the same overclocked processor speed and 3 different memory speeds: DDR428 (214), DDR478, and DDR536. The CPU is running at the same speed in all 3 tests; only the memory speed is varied by using different ratios.
| Fixed CPU Overclock and Variable Memory Speed OCZ 3700EB (DDR466) - 1 x 512Mb Double-Bank |
|||
| Benchmark | 2142 MHz 214x10 4xHT 2.5-2-3 |
2142 MHz 238x9 3xHT 3-2-3 |
2142 MHz 268x8 3xHT 3-2-3 |
| Super PI 2M Places |
95 seconds | 93 seconds | 92 seconds |
| Quake 3 FPS |
424.7 | 433.4 | 446.7 |
| Sandra Memory Test Standard Buffered |
INT 3306 FLT 3304 |
INT 3656 FLT 3653 |
INT 4137 FLT 4137 |
| Sandra Memory Test Standard UNBuffered |
INT 1913 FLT 1939 |
INT 2110 FLT 2142 |
INT 2354 FLT 2404 |
| UT2003 FPS |
Flyby 274.2 Botmatch 108.9 |
Flyby 276.6 Botmatch 110.8 |
Flyby 279.6 Botmatch 112.7 |
| Aquamark 3 Standard Score Run |
43,952 | 44,297 | 44,575 |
Lower memory timings and HyperTransport was required at the higher memory speeds. Despite those performance handicaps, you can see that higher memory speed is providing the best performance, even in simple calculation benchmarks like SuperPI.
The impact of Memory Speed on performance at a fixed CPU speed is even clearer in the Performance Charts.
Facebook
Twitter
RSS
12 Comments
View All Comments
nycxandy - Monday, June 21, 2004 - link
"OCZ is breaking new ground with their latest Enhanced Bandwidth series, and 3700EB extends the EB performance envelope. While rated at DDR466, we were able to reach a stable DDR524 on our Intel test bed and an even more remarkable DDR550 on our AMD nForce3-250 test platform."Does the 3500EB reach a speed higher than 510 (on the Intel platform) when paired with a nF3 or K8T chipset?
rustybx - Tuesday, June 8, 2004 - link
OCZ 4200EL was ~8% better than OCZ 3700EB on the Intel platform. Should the same be true for AMD?4200EL and 3700EB are nearly identical in price. Which will be faster on an MSI K8N (nForce3-250GB)?
Pumpkinierre - Wednesday, May 19, 2004 - link
Good review and good to see the a64 well featured. Your o'clock/FSB results have two that are at the same HT and latency settings. So tentatively a comparison can be made. An increase of 30MHz (at the same CPU speed)on the A64 FSB shows a 262pt (12%) sandra fp (unbuffered) improvement. Unfortunately, this doesnt translate to the games (Q3 the best at 3%- 13.3fps) but your 1024x32bit setting may have been a bit harsh (I know...you get criticised if you run it at 640x480 but it is a memory test not a gpu). Intel shows better: For a 16MHz increase (DDR466 to DDR500) they get Sandra unbuf. fp increase of 258pt (8%) and a Q3 increase of 25.6 fps (7%) although this last result is also affected by cpu speed increase because the P4 is multiplier locked. You failed to include the UT and aquamark results for the Intel tests.So it is not as good as I hoped for the a64 but the memory bandwidth % increase is hopeful. If this doubles with dual bank memory and your tests are gpu restricted then this tweaking avenue may still be worth pursuing.
TrogdorJW - Wednesday, May 19, 2004 - link
I would really like to see what sort of performance you were able to get from the Athlon 64 when using two sticks of RAM. I know that it's not a dual-channel motherboard, but in my experience, it is much more difficult to get the RAM to run at DDR400 with two DIMMs in an Athlon 64 board. If that's the case, it is imperative that we hear about it. No sense in considering OCZ or any other high-end (expensive) RAM if we're still going to be forced into running DDR333 by the motherboard.The flip side is also true: if OCZ EB RAM runs flawlessly with current Athlon 64 boards at DDR400 and even overclocked settings, I would love to hear it. I am looking at building an A64 3000+ system for a friend in the near future, and I want to know the best memory to get for A64 compatibility. Any advice would be appreciated!
ska - Wednesday, May 19, 2004 - link
could you guys color differently the benchmarks of the curently reviewed piece of hardware? it's really great that there's so many benchmarks to compare from but after like 5 or 6 it's hard to figure out where the current motherboard/processor/RAM/video card stacks up to the rest of the benchmarks.bigtoe33 - Wednesday, May 19, 2004 - link
If you are looking for the beta bios have a look here, http://www.bleedinedge.com/forum/showthread.php?t=...Wesley Fink - Wednesday, May 19, 2004 - link
#5 -As specified in the Atlon 64 Test Configuration on p. 11, Chaintech supplied a Beta BIOS dated 5/07/04 with multipliers in the BIOS. We will provide more information on ratios in our review of the Chaintech which will appear next week. We are working on an nF3-250 roundup of Epox, Chaintech, Gigabyte, and MSI which should post next week.
Rich5 - Wednesday, May 19, 2004 - link
How were you able to adjust the multiplier on the Chaintech VNF3-250 motherboard? I know that it's not (yet) adjustible in the bios - did you use the Clock Gen utility from wcpuid.com? From what I could see at their website, it only appeared that they had versions for motherboards with the nforce 3 150 chipset (and nothing specifically for Chaintech boards).Wesley Fink - Wednesday, May 19, 2004 - link
#1 -When the dust settles we will upgrade our standard test video card, but it is too early to make that decision. We use a standard video card so results are comparable to past reviews. We do not change our test hardware just because soemthing new is out this week.
We have an X800 PRO in the motherboard test lab, but we are not ready to decide which card is our future standard yet.
Illissius - Wednesday, May 19, 2004 - link
I've seen such benchmarks, can't remember where though (possibly xbitlabs). IIRC it was in the context of a K8T800 Pro review. At any rate, performance is equivalent at 5x (1GHz) and 4x (800MHz), takes a (small) hit in some applications at 3x (600MHz), and only really starts becoming a major bottleneck at 2x (400MHz).