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. Not entirely stupid. Rules are determined by the competition and you're not competing directly for hwbot rankings, you're competing in the "Team Cup". The "side effect" is that these scores also go into hwbot rankings. Yeah, you're sacrificing some MHz because of the screenshot, but you can always continue and make a highest possible validation (on the edge) just outside of the competition. Don't see how the screenshot rule is an obstacle for that. Just ease yourself and use one of the many lightweight screenshot captures, so you don't bother saving it with paint every time.
  2. Tried various lod, I think it was 4.8 for this run. 15 didin't give me anything more.
  3. Don't remember the readings on those caps, but measuring the vddq pins on the mem chips should read the same, no?
  4. I thought they are tied. vdd = vddq here, measured directly on the corresponding pins of the mem chips. Increasing vdd increases vddq as well and they are equal.
  5. It's the same then. Based on the pcb traces and datasheet of the controller - yes, should be the right one. Plus the mem voltage is increasing.
  6. No thoughts on the 97.44 driver? Old thread with screenshots: http://hwbot.org/news/1679_turrican_wins_oc_challenge_of_january_10_challenge_big_success Found this one as well, so I guess it's not allowed: http://hwbot.org/news/1849_update_rules_bugged_driver_rule_elaborated PS: Or allowed?
  7. Not sure how much would be the difference if we were using intel cpus. My cpu clock is more than enough, but still can't touch the top scores even with higher vga clocks. It might be my lack of skills, though...
  8. Yes, I know about memory, but this driver is the only one working on my system. I'm getting awful scores with other drivers, something like 100 points below that. Got something like 268 with 66.93 and every other driver I tried. Don't know if it's something AMD specific. Otherwise it's true that these cards are limited by the memory subsystem. I've managed to clock the core to 360MHz on air, but struggling with the memory. It even gets worse with more volts. I saw you have attached wire to fb1, so I'm going to look at the datasheet once again, maybe there's something I'm missing in my mod. And the performance is not good either. No way I can reach your score with the current situation.
  9. Great score. Id on't know how you guys pull out such good scores. My performance per clocks is awful
  10. Same here with my nLighted XP.
  11. yeah, 7.5GHz on air and 880G chipset. Another reliable guy.
  12. I have to bring out the old question about bugged driver 97.44. There are some submissions in the rankings with that driver and the question is if we are allowed to use it in the competition? It's either we all use it, or nobody uses it. Equal playing field and so on. If I have to vote, I'd say no, but don't want to be beaten only because someone used it and I didn't. Wanted to say it at least for those who are not aware of this. I know finding the right driver is part of the fun, but for gf4 it's obvious which one gives highest scores.
  13. Which bios version? I'm running 0905. This is by far the most stable for my board and I don't want to change to newest, cause I can't revert back.
  14. ryba RMA'ed his XP-M 2600+ and got a record-breaking one
  15. No difference, mate. Setting first core to x4 multi, fsb goes to 953, second core to 18GHz, memory at 3813
  16. Ok, uploaded some screenshot on imgur - http://imgur.com/a/KvRl0 Explaining one by one, although it's visible what i did. System info - Phenom II X2 555 unlocked to B55, running at 3.8, 2.8 cpu-nb and 1600 memory Crosshair IV Formula, Windows 7 x64 Ultimate 1. set multi on first core one notch up, resulting in 3.9GHz for the first core only, cpuz thinks I've reduced the fsb by 5-6MHz 2. multi set to lowest x4, cpuz thinks I've increased the fsb to 953Mhz, which leads to memory at 3813 (DDR3-7626) and uncore at 13.34GHz 3. fsb set to 205, nothing happens on cpu-z. TurboEVO crashed 4. multi on first core set to 1 notch down (0.5x), fsb goes 6MHz up according to cpuz 5. Correct 24/7 settings Opening cpuz after setting multi/fsb doesn't change a bit, still reads wrong clocks
  17. Unfortunately, it's bugged for me again. Will report back with more details.
  18. Very nice card. Can't beat that memory with my Gainward GS
  19. Can see some matches here with my list Almost all besides first two and I already had the 5900XT.
  20. He's not fair. I transferred my great core2 knowledge, so he had only one thing to do - push the button. So, basically all his scores are mine!
  21. Sweet app, thanks. Here's my feedback: - Sometimes there is no space on the screen or I just want to overlap different cpuz windows to save space or to arrange them better. It would be nice if I can set z-index for every cpuz/gpuz screen if I want to. Any other way to achieve this? - It would be also nice if I have a control for the delay. Sometimes opening several instances of cpuz/gpuz too fast is causing the apps to crash. - +1 for the customizeable path for screenshots. I don't like saving them on the desktop, but in a separate folder on a usb stick or different partition. But if the goal is "keep it simple", which I admire, then it's ok to skip this. PS: As I see the app opens cpuz windows based on the selected tab index, i.e. starts from the highest index (e.g. 4 for Memory) and opens this first, when in fact I need it opened last and overlap with the other window (index = 3 for Mainboard). Edit: Uhm, no...maybe the order is bottom to top, right to left on the screen? Open order would be so much easier.
  22. You can't beat me on DDR anyway, so...not a big deal
×
×
  • Create New...