sdurazzo Posted November 22, 2015 Posted November 22, 2015 I have watched Rbuass's video (Drop the bomb), and regardless he did not accuse any overclocker, I would like to ask respectful about scores bellow. Can someone explain how it is possible? Fire Strike Extreme 1st (KP) 14333 http://www.3dmark.com/fs/6340698 GPU 1856 / 2170 Mhz CPU 5710 Mhz 2nd Rbuass 13674 http://www.3dmark.com/3dm/9320519 GPU 1924 / 2168 CPU 5932 Rbuass is using a Rampage 5, which is a recognized great board, and has the CPU 222 MHz stronger (even higher VRAM and NB) with 70 Mhz more on the video card. How is possible to score almost 700 points higher? What kind of Tweak? On Fire Strike, is even hard to understand, since CPU power is more important, and VRAM is not too important. 1st (KP) 27551 http://www.3dmark.com/fs/6349265 GPU 1856 / 2170 Mhz CPU 5765 Mhz Graphics Score 31985 Physics Score 27603 Combined Score 13491 Graphics Test 1 156.23 fps Graphics Test 2 125.31 fps Physics Test 87.63 fps Combined Test 62.75 fps 2nd Rbuass 26844 http://www.3dmark.com/3dm/9319937 GPU 1920 / 2157 CPU 5932 Graphics Score 30655 Physics Score 28714 Combined Score 13225 Graphics Test 1 151.18 fps Graphics Test 2 119.17 fps Physics Test 91.16 fps Combined Test 61.51 fps Looking analyze without being biased Physics Power, that is very important, are ok, rbuass 91,16 > KP 87,63, but how is possible to have GT2 119.17 x GT2 125.31. What kind of magic can do it? Again, I do not want to accuse, but I want the right to suspect, and also to tell this scores need to be justified. Sure both contenders used the LOD tweak (old SI). Please note Mr. Kingpin's recent submissions of FS/FSX/FXU in Futuremark's Hall of Fame, without the "LOD" tricks are way out of line as well, which outscored normal top scores with the same unreasonable FPS. How could that be possible ? I think people have the rights to know http://www.3dmark.com/fs/6400442 http://www.3dmark.com/fs/6471755 http://www.3dmark.com/fs/6436818 Sorry Rbuass, I hope you do not mind for have brought the question under your scores, and sorry Mr. Kingpin, I am not accusing you, but you even need to understand that this difference is abnormal. Best regards to all.
Lays Posted November 22, 2015 Posted November 22, 2015 980 TI Kingpin has a fancy BIOS that makes it perform better clock for clock. Apparently it tightens the memory timings, and makes the GPU more efficient per clock. (Nearly as efficient as a Titan X at the SAME clocks) That is most likely the reason right there. 1
Gunslinger Posted November 22, 2015 Posted November 22, 2015 980 TI Kingpin has a fancy BIOS that makes it perform better clock for clock. Apparently it tightens the memory timings, and makes the GPU more efficient per clock. (Nearly as efficient as a Titan X at the SAME clocks) That is most likely the reason right there. ^^ This 100% ^^ 1
sdurazzo Posted November 22, 2015 Author Posted November 22, 2015 It is because is Kingpin, so do not need to explain??? Do not tell me that, because this difference can never be justified because of memory timing. We come here, ask a question with respect and want an answer to justify, not to come to talk to people that what they believe is true. Sorry, but for me, this proves nothing at all
Lays Posted November 22, 2015 Posted November 22, 2015 (edited) It is because is Kingpin, so do not need to explain???Do not tell me that, because this difference can never be justified because of memory timing. We come here, ask a question with respect and want an answer to justify, not to come to talk to people that what they believe is true. Sorry, but for me, this proves nothing at all There is plenty of proof here: https://www.kingpincooling.com/forum/showthread.php?t=3816 Also you're comparing the gpu clocks wrong, compare the gpu boost clocks as listed on their submissions. Firestrike normal for Rbauss was at 2172/2156, Vince was at 2172/2170 Vince having a higher memory clock paired with the more efficient bios is why he is getting a better score. Edited November 22, 2015 by Lays 1
sdurazzo Posted November 22, 2015 Author Posted November 22, 2015 I'm sorry but, politely. There is no answer at all. The topic that you sent, does not explain anything. It does not explain this absurd difference. Even if he had the fastest memory in the world (the Galax and MSI are also Samsung) would justify this difference. I'm not here to attack, but I am not to see friends protecting him because he is Kingpin. If it were 100 points, it would be. If they were 200, it will be too much; but are 700 and would like a real justification, not a justification that shows - "is because he is Kingpin".
Lays Posted November 22, 2015 Posted November 22, 2015 I'm sorry but, politely.There is no answer at all. The topic that you sent, does not explain anything. It does not explain this absurd difference. Even if he had the fastest memory in the world (the Galax and MSI are also Samsung) would justify this difference. I'm not here to attack, but I am not to see friends protecting him because he is Kingpin. If it were 100 points, it would be. If they were 200, it will be too much; but are 700 and would like a real justification, not a justification that shows - "is because he is Kingpin". Did you not read all that I wrote? The gpus are at the same clock speed, memory on the 980 ti kingpin was faster. Pair that with tighter memory timings and that is why the score is higher. The difference in gpu score is like 4%. Memory timings on cpu can net that type of Performance easily, why can't a gpu do the same? Firestrike is heavily reliant on gpu score and combined score, physics helps a lot too, but raw gpu power always gives higher score than physics.
sdurazzo Posted November 22, 2015 Author Posted November 22, 2015 I know quite well the benchmark, so there is no justification. Rbuass has the stronger GPU so, with stronger CPU it is much more likely and acceptable that he used "bclk tweak" than justify memory timings doing absurd scores. I will not discuss, I would wanted justifications and not lawyers. For me, there is none until to prove me. No sense to accept this difference without challenge. All the best to you guys. what I had to ask I did and no satisfactory answer I received
Lays Posted November 22, 2015 Posted November 22, 2015 I know quite well the benchmark, so there is no justification.Rbuass has the stronger GPU so, with stronger CPU it is much more likely and acceptable that he used "bclk tweak" than justify memory timings doing absurd scores. I will not discuss, I would wanted justifications and not lawyers. For me, there is none until to prove me. No sense to accept this difference without challenge. All the best to you guys. what I had to ask I did and no satisfactory answer I received Where are you getting this "rbauss had stronger gpu" from? Look at the dam submissions dude, Kingpins gpu had same gpu clock, and higher memory clock. http://hwbot.org/submission/3036711_rbuass_3dmark___fire_strike_geforce_gtx_980_ti_26844_marks http://hwbot.org/submission/3023964_kingpin_3dmark___fire_strike_geforce_gtx_980_ti_27551_marks Just because you don't like the answer you've been given doesn't mean it's incorrect.
steponz Posted November 22, 2015 Posted November 22, 2015 (edited) I like how this guy just joins and starts bashing kingpin. Join date nov. 2015 Only 5 posts at this time ever. Highly suspect of who this really might be. The English is also off.. Kingpins scores are legit plain and simple. Gpu score is strictly based upon gpu... cpu doesn't help gpu score at all... you can run 4ghz cpu and get the same gpu score. If you know the benchmark so well.. where's your submissions and why have I not heard of you? Edited November 22, 2015 by steponz
Mr.Scott Posted November 22, 2015 Posted November 22, 2015 part of the clone army You misspelled clown. 1
Crew Don_Dan Posted November 22, 2015 Crew Posted November 22, 2015 I think we can all agree that KP's scores are legit. I'll close this thread before it gets out of hand.
Recommended Posts