Massman Posted March 18, 2010 Posted March 18, 2010 Yes, because that's what I do normally, right? Not explain, not elaborate, not discuss ... just say fck all and move on. Your assumption is "it hasn't been done before" and the conclusion "it's a glitch", I've just shown that your assumption is incorrect. As to your GPU-Z problem, I've figured this out a month and a half ago: http://www.madshrimps.be/?action=getarticle&articID=988 GPU-Z might get the frequencies right when you use 0.3.9, although I had major issues with the Gigabyte H55 at home (back then I was debugging with W1zzard on this tool). GPU-Z is mainly used to identify the right videocard. Quote
rickss69 Posted March 18, 2010 Posted March 18, 2010 Yes, because that's what I do normally, right? Not explain, not elaborate, not discuss ... just say fck all and move on. Your assumption is "it hasn't been done before" and the conclusion "it's a glitch", I've just shown that your assumption is incorrect. As to your GPU-Z problem, I've figured this out a month and a half ago: http://www.madshrimps.be/?action=getarticle&articID=988 GPU-Z might get the frequencies right when you use 0.3.9, although I had major issues with the Gigabyte H55 at home (back then I was debugging with W1zzard on this tool). GPU-Z is mainly used to identify the right videocard. I don't believe that is so when 80% of the gpu-z screens do not record the frequencies. Your link does not address gpu-z at all and there still has been no explanation where the stated frequencies were taken from. Just trying to get to the bottom of it... Quote
Massman Posted March 18, 2010 Posted March 18, 2010 My article is build on the fact that no application is capable of reading the true frequency of the GMA HD ... Again, W1zzard of techpowerup claims it's possible to read out the frequencies. I can't say more than that, I don't make those apps. The states frequencies are probably following the formula from that article (which has been confirmed to be correct by Asus). You're trying to get to the bottom of what exactly? Quote
rickss69 Posted March 18, 2010 Posted March 18, 2010 For illustration: I can see what the numbers are here in this screen, be they correct or not. I just did not understand where other's were getting thier numbers from. The fact that some gpu-z tabs read while others do not led me to think that we were not getting accurate results from the benchmark. Quote
icebob Posted March 18, 2010 Posted March 18, 2010 The problem is 32 bits vs 64 bit os, 32bits show correct frequencies 64 don't, on both vista and win 7 Quote
chuchnit Posted March 18, 2010 Posted March 18, 2010 Sucks to find a bug in the middle of the month Any chances further contest will have specified driver/OS at the beginning? Quote
Massman Posted March 18, 2010 Posted March 18, 2010 It's amazing that this hasn't been caught before! I suppose setting rules to define driver and OS are almost inevitable ... it's a pity though that more rules are needed to avoid this kind of problem. Quote
chuchnit Posted March 18, 2010 Posted March 18, 2010 It sucks, but might be the only way to run these official competitions. Then it would rely on contestants to have their system tuned perfect to take the win. Quote
dasa Posted March 18, 2010 Posted March 18, 2010 (edited) If we are to discuss obvious glitch's then I must bring up another item. Both bennylodewijk's and dasa's submissions state core frequencies of almost 1300mhz...these chips simply won't run there. I do not know where they got those numbers from or if it was an honest mistake. Has anyone else been able to run Vantage at those frequencies? If not, I must conclude they were glitch runs. My view is that if gpu-z will not show the core/mem frequencies that is a glitch in itself. How about making that a requirement for submissions? yer i must have hit a wrong key when i was calculating that speed sorry it was running 790x205 so 1217mhz 1.68v Edited March 18, 2010 by dasa Quote
lodewijk1978 Posted March 18, 2010 Posted March 18, 2010 Take a look at ekkyjengkol's screenshot GPU-Z says 1300++ GPU clock... Quote
steponz Posted March 19, 2010 Posted March 19, 2010 The problem with this is in 64 bit windows GPUz will not read the proper speed. So then 32 bit windows would have to be a requirement. Quote
rickss69 Posted March 20, 2010 Posted March 20, 2010 What kind of card is that in the pci slot on some of the competitor's mb's? They all appear to be the same with some kind of post code led showing. Quote
lodewijk1978 Posted March 20, 2010 Posted March 20, 2010 What kind of card is that in the pci slot on some of the competitor's mb's? They all appear to be the same with some kind of post code led showing. It's a debug code PCI Card. Quote
rickss69 Posted March 21, 2010 Posted March 21, 2010 Any other benchmark would have been preferable to Vantage for this competition. Way too many bugs here to be realistic. When a score is put up that is far and away higher does that constitute a bugged driver...even when said driver operates correctly in every other benchmark? Notice the myriad of OS/driver combination's thus far. It is clear that this is a driver carnival for sure. Quote
Massman Posted March 21, 2010 Posted March 21, 2010 I'm sure that if this error would've been found earlier on, the benchmark would've been different. But then again, this has never been reported before ... Scoring 40% higher due to a severly impropper rendered benchmark is bugged for sure. Quote
Massman Posted March 21, 2010 Posted March 21, 2010 Rickss69: please don't submit any bugged runs. Quote
rickss69 Posted March 21, 2010 Posted March 21, 2010 Big bugs, small bugs...who ever knows for sure. Now you are starting to understand. Quote
Massman Posted March 21, 2010 Posted March 21, 2010 Are you trying to teach me a lesson of some sort? Quote
Topalof Posted March 26, 2010 Posted March 26, 2010 (edited) Personally, I tested the drivers available for vista/win7 32-bit (1968/2057) and win7 64-bit (1995/2057/2092), and no score bugged. All scores are quite similar. home, the display tweaks are important and provide pts;) i can run the IGP at ~1327/920. edit : 1337/930 edit : 134x gpu passed edit : 1358 gpu passed edit : 136x ok and 1371mhz not passed one cpu dead on two pcs edit : 1371mhz passed,but low score...unstable 661pts gpu lol Edited March 29, 2010 by Topalof Quote
Pt1t Posted March 31, 2010 Posted March 31, 2010 (edited) Hello, An hwbot crew member, Thor941, has blocked one of your submitted scores. It has been marked as 'incorrect submission'. This was the reason the user gave:1995 driver and Vista 32 are not allowed You can view your result here: http://hwbot.org/community/submission/976419_pt1t_3dmark_vantage___performance_gma_hd_clarkdale_976_marks You can find out more about the reporting user here: http://hwbot.org/community/user/thor941 WTF with that guy, score was previously checked OK by moderator and now some hours before the end of contest it s checked incorrect. Neither Vista 32 or 1995 affect score ... P976 for 5.3 and 1178Mhz IGP is normal score Edited March 31, 2010 by Pt1t Quote
Thor941 Posted March 31, 2010 Posted March 31, 2010 ... 1995 driver in combination with Vista 32-bit is no longer allowed in the competition. The scores have been removed. ... Quote
dasa Posted April 6, 2010 Posted April 6, 2010 So who was the random user selected? I would like to know this as well Quote
Massman Posted April 6, 2010 Posted April 6, 2010 http://hwbot.org/article/news/topalof_wins_gigabyte_h55h57_oc_challenge "SniperOZ is the lucky winner of GIGABYTE’s most high-end P55 mainboard, the GA-P55A-UD7, for his P731 entry." Quote
SniperOZ Posted May 12, 2010 Posted May 12, 2010 Woo Hoo!!!!!!........I just recivied a GA-P55A-UD7 in the mail today Thanks Gigabyte and Hwbot Cheers Sniper Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.