I.nfraR.ed

Yes, I agree, but that's kind of different issue - modifying the c4d file, which is not secured in any way.

Yes, but you're actually showing a "saved/previous" score and not the one you get just after the bench finishes. It could be easily avoided if the screenshot requires the score you get on the top left corner. However, the bench needs to be investigated for other loopholes.

Yes, the conclusion is - use it without LOD and enable Catalyst A.I.

Yes, your chapter scores are completely in line with other scores. There's no way your GFX score is so low. On the other hand all other scores I checked have the CPU score in the 33K - 47K range. I think the wrapper failed to get the scores from AM right.

Why your cpu and gpu points are switched? Even if we take the cpu points as GFX and vise versa, 103K GFX ain't too high? Wrapper seems bugged, that's why I'm not benching aquamark until better one appears .

Maybe some cheap gpu pot for old VGA cards like nvidia 4, 5, 6 series, ati old 7, 8, 9 etc. Chipset cooler is also welcomed. Yeah, I know new platforms don't need it, but I'm benching a lot of old stuff and a proper pot would help me. Other variant is to make one "myself", in fact I have it like an idea, but had no time so far. However not really sure sure how many people would buy these...

Ok, fair enough. I hate to report people and do it really rare. However this was one of the cases that I couldn't close my eyes and say "let it go". I don't have anything personal against you, I do even like the scores. Btw, thanks for changing the background

It's about time you move to ProOC.

I usually go up to 1.7V on air/water if the card can take it (however needs cold) This one has some potential 980MHz-1GHz air, but there's some more work on it. 1220MHz on the ram is good for Qimonda without voltmod, most of them reach 1100-1150 stock. Best is Samsung 0.8ns, easily 1300+ with a little more voltage.

Tried 9.4 as well, but score is about the same. CPU is stable and when I lower vga mem or gpu fps go lower accordingly, so it's not unstable clocks either. VGPU is 1.55V; mem is Qimonda, but not voltmodded yet (I'm out of suitable trimmers); LOD = 4. OS is XP SP2, gonna try SP3 and Seven tomorrow. VGA on air, cpu on water. Score seems low for the clocks, more like I'm running 860-870MHz on the gpu...

Glad it helped, but I'm getting lower and lower fps after every reboot . Testing 4850 on air + XP + 9.12

Yeah, most of the top scores are with XP + 9.12. Still haven't HD48xx that seriously, but I have them on the long list, too. It could be the mem clocked too high or gpu/mem too cold (?) as Karl supposed.

What do you set LOD with? For 06 disable Catalyst A.I. and try to set LOD again. You can try also Vista/7 with 9.12 driver and 9.4 for XP.

Just finished 03, 05 and 06, but unfortunately 797MHz GPU is my max on air. Max mem clocks dropped a little. Next is 01, but need to install new OS.

You need to input the real frequency, not the effective one. With this in mind you can understand how incredible are the speeds achieved by the guys at the top.

He noted it, but it's arguable if higher LOD gives you higher result. Based on my testing during the CountryCup, LOD 4(5) was best for the low end cards in Vantage. Unfortunately I don't have any high end models, it might be different there.

1,589MHz 9-9-9-24 on s.1366 with air cooling...yeah, right

You both don't play fair, my card can only run 756MHz gpu Well, without hard voltmod yet, but will do it soon.

Excellent work. Your gpu is good, too

Qimonda, yes. MVDDC and MVDDQ around 1.68V. VGPU ~1.5V.

We did the same mistake for the CountryCup - 116MHz on Maximus V Formula, several screenshot, but it turned out the validation was made with cpu-z 1.60, sigh...

Message says 1.61.3+. Guess you've tried to upload a validation file done with version < 1.61

Absolutely no problems with Maximus V Formula and MVG before it. I had it also working for some hours in this condition. It had been flawless for now. I only use art eraser, neoprene and paper towels. Some of my boards are insulated with liquid tape on the back, but not this one.

Normally a trimmer between phase and fb pins, instead of fb - gnd.

Yes, it is. You need ~300-400MHz to compensate the bad performance.