I.M.O.G.

Full XP SP3 here as well. (trying to avoid weirdness) I can do a mini test if ATI has the problem as well as NVIDIA - I never paid attention when it does it before. Seems like everything, but then I've mostly only been benching NVIDIA stuff. Next up is a 4870, so we'll see if it sucks or not. @Crusty: I have deleted directcpl.dll before, and it did fix some problem, but I forget what.

Nice one!

I always run in high. Didn't try realtime. I would be happily surprised if that does it.

Thanks for the input. We need to find something on this. It's mainly just a hassle to do a fresh OS install each time we want to run it - once ran a few times the wrapper is murdering the scores like genieben mentioned. Tried again today on the 9800 GTX... Win7 producing the same lower scores I was getting on XP SP3. Will have to get a fresh OS install going to get "real" aquamark scores unless someone else has another solution.

Thanks for the confirmation. I think I've done a fresh install before to get around it. This time it appeared to be happening since the first run, although I did run AM3 a few times while pulling temps down and before having clocks fully up, so it could have gotten the good runs out before I had my clocks up. Anyone with an approach that doesn't involve an OS reinstall?

Nothing?

bump I ended up running it under win7, and the wrapper was scoring about 420K there, however I'm still missing 10-20K because I can't get the wrapper producing proper scores on XP SP3.

Scoring 434K with aquamark ran directly... 365K through wrapper. This is an example of a recent run, similar card, also XP, but didn't score low through wrapper... I just don't remember what the difference was: http://hwbot.org/submission/2296818_i.m.o.g._aquamark_geforce_8800_gt_512_mb_412959_marks

Is there a known fix for low result on XP with aquamark wrapper, and higher result running aquamark directly? On 8800 series card, I know I've gotten good scores in the past on XP with these cards, but I don't remember if I was still getting higher without wrapper. On a pretest run, the difference was about 50K. Results reproducible, and I think I've seen this discussed elsewhere in this forum before. FIX

Nice score!

It's probably leaving secondary or tertiary subtimings looser than the 890FX/Giga boards you are comparing to. Similar to how some of the z77 gigabyte boards set sub-timings too tight and have been the cause of some memory compatibility issues people have talked about. Anyways though, I always used the CIVE over the CVF when possible (non-bulldozer), because it yielded me the best results... But if you wanted to make sure you weren't spending money on something just for the sake of convenience, I'd look closer at the 2nd/3rd subtimings.

Nice result. I have a CVF and a CIVE without any plans of being run in the future. The CVF we could make a deal on, though the CIVE has sentimental value, but I could be talked into letting it go if that's what you really want.

Looks really good. I went through superpi, pifast, ucbench, and pcm05 specifically and they look good.

And now it says #1 UCBench World Record... Yet its ranked 72nd for hex cores. lol

Nice sub buddy!

Just a heads up, we're smashing hardware next week: http://www.overclockers.com/forums/showthread.php?t=713061

Thanks for sharing! pwned by swear filter, bunnyes! lol

The rule of benchmarks prevails... In motherboard reviews where benchmarks are heavily utilized by readers, it gives Gigabyte a perceived "efficiency" advantage because they tighten some things up more than other boards. In most reviews, the reviewer and reader pays no attention to the resulting sub-timings - they assume RAM timings are detected by SPD and set identically. So in a motherboard review where a board is put head to head against a different board, the variance between products is typically 1-3% with all other factors identical (within margin of error)... But a little tweaking like this can give a perceived edge to one board in the benchmark results. Gigabyte isn't the first to do it. With enough memory module testing while in engineering/bios development, most popular memory will just work and no one pays any mind... The only reason anyone is talking about manually changing these settings on Gigabyte is because some DIMMs weren't tested enough so the auto-settings don't work on some DIMMs at top-end frequencies, which lead to some people figuring out what was up. Finally, your point doesn't entirely make sense to me. The sticks do just work without fiddling with settings that 99.99% of people don't understand. It's only when pushing aggressive memory frequencies that are pointless for anything other than benchmarking itself that fiddling is required - and fiddling is exactly what we're benchmarking for.

Awesome submission Bart!

Nice work Evan!

Looks great! It still says superpi, but I imagine thats part of the WIP. The presentation is much more clear with how its laid out and what is specifically stated.

Gone completely from my profile now, like I never did it. And turrican had just reviewed and approved it earlier today. Lol

If you are submitting anything other than CPU Frequency, mem frequency, reference clock results... you can just use an old working version of CPUz for the screenshots. CPUz being broken in new versions really only impacts CPUz validations, which are only required for those benches. Just thought I'd mention this, as it seems maybe some people are worried about CPUz functionality for subs other than those.

Awesome score, impressive.

Not all... I have 2 currently, just sold one. 3218b957 was benchable at 6.3GHz, 6.4GHz with extra volts. 3218b958 was benchable at 6.5GHz. I didn't go past 1.92V though, and I try to stick to 1.85ish to keep 'em from dieing.