Jump to content
HWBOT Community Forums

I.nfraR.ed

Members
  • Posts

    2464
  • Joined

  • Last visited

  • Days Won

    36

Everything posted by I.nfraR.ed

  1. And throttling. Need to cool it down like Hiwa, otherwise a mid/low-end MTK device @ 1.2GHz beats this...
  2. I've said it before, but I've tested about 6-8 (or maybe even less, don't remember really ) chips during last 2-3 years. I thought 7GHz might be possible, but the cpu is quite intolerant to voltage.
  3. kotori-san, it was not needed. Sorry if I made it sound that I don't trust you. It's a very good score, I could do about 740 with X2, but need to work more. Latency with Opteron is higher, so it's not an option and Expert board is more efficient than Ultra-D. CPU frequency does not help - it could only reduce latency by 0.1ns or so.
  4. No, latency is higher and score is lower. I think the key here is dual-core with 2x512MB L2, not 2x1024 like Opteron/FX
  5. No need for video. I have very different score with Opteron 180... I also do everything in bios and boot straight to os.
  6. Hmm, ok. So I will try with Expert, although my board can't clock BH5 very high for some reason. I get sometimes similar read and write, but I thought it is bugged
  7. I have ran it and the score is about the same. I think his score is bugged What are you getting with similar clocks? Bypass max is wrong here, but everything else is tight enough Edit: btw, expert and venus should be more efficient than ultra-d/sli-d, but I don't think the difference in read and write could be that huge.
  8. Can you reproduce this? Your latency is higher than mine, but read and write seem like mine (even higher) with v1.82, while mem copy is lower. I don't know what is real anymore
  9. NCP5835 datasheet
  10. No, I haven't forgot, but 2-2-2-5 is faster than 2-2-2-11 on my board/bios.
  11. Same for me. I don't even have 4770K and I'm almost always at least a generation behind. Good that I was lucky with 3770K, but it still requires a lot of LN2 to bench all these VGAs. But I promise you will have some competition for s.939
  12. These Matrix cards are not as rare as I though some time ago, however it's all about binning or luck. Higher chances with ref cards, I think. I had ref that did about 1010 - 1020 on default voltage and 1120 1.475V on water. Mem was good for 1330 1.35V. My Matrix can't touch this and last time I tried it on LN2 it was also not good, but I will revisit it some time.
  13. poparamiro, Asus non-ref like these or some other non-ref/ref card? Because I have Matrix as well and it is not very good. My ref cards are better
  14. I don't think it's necessary. Leave it this way. I'm fine with it as long as we all use the latest version for the competition. I don't even run this bench and I wouldn't if it wasn't the competition .
  15. I'm already close to the top scores, but only with the old version. It's impossible with newer versions.
  16. Most of the guys with old scores have absolutely no fault. I've updated my score and with old version I get 790 vs 710 with same settings and timings.
  17. Yes, there was a latency bug in some of the versions, probably 1.82. Didn't test, but it appeared when you boot at higher clocks and then clock down the fsb a notch or two within OS. Rules on hwbot say we should use the latest official version, which seems to be 2.01, but on the hwbot site 1.95 is linked. It seems these records will never be broken.
  18. How are we supposed to compete with older scores done with v1.82?
  19. Another warmup? I'm already overheating...
  20. Don't worry, I will try 6.6+ GHz this weekend
  21. Nice one! You beat me with 200MHz lower clock.
  22. The fact he moved to XOC doesn't solve the "problem" for me either. It doesn't matter which league you belong to. Trying to fool people is still not acceptable. I guess people are asking themselves if the cooling is chilled water indeed, then how that screenshot was produced? Because the frequency is really hard to believe. If the cooling is something else like LN2, then he was lying from the beginning. Either way, it is not good. A simple video showing gpu running even at lower fequency than 1100, plus cooling and temps would be enough. In my opinion, if I had such a great card, I'd want to show it in every possible way
  23. The thread seems accessible, thanks. So the question about vGPU still remains unanswered.
×
×
  • Create New...