knopflerbruce

I was in hospital during the last few days of this competition, so I couldn't push my 939 score higher, but I think 58k is doable

I'm not going to make this decision alone, as this case isn't that clear. Strictly speaking, it's not allowed to pass the card around like that (- VGA may not be shared for 3D Benchmarks). However, in this competition sharing doesnt give much of an advantage as only the best score pr team counts, so basically all other submissions are worthless. The only way you can gain something here is to get that one score that matches the VGA to the best CPU and perhaps the bencher who can get the most efficient run - which IMO is not against the spirit of this month's competition at all, it's about team effort and not personal gain. If any punishment is to be given, I'd say a warning + removal of all except one of the 7950GT AGP submission is enough. It's pretty clear that this was done in good faith. Maybe we need to add a paragraph about this type of sharing later, to avoid more problems like this. Now, stop fighting and shake hands - both of you have valid arguments, the error here is on our side. I hope this made sense, it's almost 3 AM and I've had a damn headache all day... Gnite!

I think Vince will wet his pants when he sees this Nice score!

Im not sure how the rules apply in this case, as there is a huge difference between this competition and regular rankings: here only the best score counts, so the others who ran it gained NOTHING for their team, or themselves (only two members ran this model and submitted the scores for boints).

What BIOS for UD4-B3?

socket F!

I can delete it if you wish, then you resubmit:)

If you mean what fraction of saved screenshots that get published, I'd say ~50% here ^^

THis is most likely by mistake (if it was photoshopped it would've been different). I moved the result

I'm not sure if the validation thingy works correctly, but you don't need it for submissions - a screenshot is enough. No, this doesn't answer your question, but maybe it helps anyway?

Then only 4-way GPU and giga-servers get boints. How interesting is that? Not so much. The ranking isn't perfect (most of the gold cups with 10+ more submissions that get 2 boints pr run are alot more difficult to beat than the 50th place 20 boint scores in the popular categories is another example), but we can only try to improve it. Perhaps if 300 people benched those servers it would be different... Then you'd get alot of boints for those results.

That's pretty much the case, unfortunately. Pictures help a bit, but they're definately not enough to tell for sure if someone did or didn't share CPUs and/or GPUs. Let's hope rev. 4 will make things easier:) PS: this is a general statement, not accusing anyone who provided pics when asked for it.

I don't mind people benching together, but to avoid confusion it's a good idea to notify the crew if you're going to bench the same CPUs/GPUs (a russian team posted some pics in this thread a long time ago: http://hwbot.org/forum/showthread.php?t=1547 for example).

Well, all these similarities make the scores look fishy, no doubt about that. Some sort of investigation is needed, pics of CPUs and cards should help a bit I guess.

What happened that day in 2006??

Haha "we remove the stuff you dont need, and put all the $$$ we save in our own pocket":o Well, let's see if this is just launch hype price, or if it stays that high:)

Very nice score for a Canadian!

I'm wondering if I should try 478 this weekend... no idea how to OC this platform, though. Ln2 should help. Then I can try some more 939 action later, 57k shouldn't be too hard really.

I noticed But I'm not done:D 57k should be possible.

If it's no boints, then it's harmless - however, the tweak fucks up the benchmark, so it should not be allowed.

5B70 and 5B73 are also x1050 device IDs

I cant find any online reference that says 64bit or 128bit belongs to one or another. Plus, gigabyte only has ONE x1050 and it's a 64bit version:) The score seems too low for 128bit as well, but reasonable for 64bit based on clock speeds.

Half the setup is hidden under my desk, grainy as hell because of it:)

Isnt PCMark more about storage anyway? No need to separate GPUs? I'm thinking the following is a good set of benchmarks: 01/PCI 01/AGP 01/PCIE PCMark05/ANY superpi 32m wprime 1024m ...although I'm not sure if it's a good idea to have that many long benchmarks (3). Maybe wprime 32m instead?

Too bad I'm too lazy too often:p