Jump to content
HWBOT Community Forums

TAGG

Members
  • Posts

    701
  • Joined

  • Last visited

  • Days Won

    5

Everything posted by TAGG

  1. Looking for a stick of sdram with quimonda chips, doesn't need to be your top stick but 100% working and tested is a must as i'm trying to figgure out where my issues with these are...
  2. Nah not nearly as bad as my mobile prescott at 5,5GHz
  3. I spot some alpo caps board fuly recapped or just vcore?
  4. TAGG

    kingston hypers

    adapters gone, up for some cheap hypers
  5. You mean the NB single stage? that's the orange thing in the background
  6. TAGG

    kingston hypers

    hypers down to 10€ a stick
  7. TAGG

    kingston hypers

    MOFC sold, up for some adapters and hypers
  8. If i would suspect the extra performance comming from the .dll swap messing with timers or the ammount of work being done that'd be accurate, but all it does is provide a way for the cpu to more efficiently use it's instructions to do the work it's given by wprime, i actually even used a stopwatch to make sure it actually is faster and not just appearing faster like back in the day with the SR-2 It's kinda the OS equivalent to the AGESA patch thigie with FM1 and gpupi, providing a faster (allthough inoficcial) way to execute certain code ... (or as mentioned many times before using intel open cl with avx2 for ryzen)
  9. I'll just leave that here, you guys do whatever you want with the knowledge: Ideal OS: Win 2K datacenter edition https://drive.google.com/file/d/1Gx9JkXp0XvNL7D4Y-JZfiFB17XcsGKQo/view The tweak: for 775 get newest vb runtime here, on other platforms diffrent ons may be faster: https://www.dll-files.com/msvbvm60.dll.html I allso attached awardfabrik tweaker for realtime effi, now you can easily beat my scores, but be reminded that i can allso rebench wPrime_Tweaker_AF.exe
  10. I did the whole just pressing bench in spi 32M for years, never stood a chance, spent 20+ hours last week actually figguring out the effi there and now got it to a point where i stand a chance, not complaining just saying that there is abolutely no chance in many benches allready unless you spend hours on working out tweaks... I think that's exactly why @TerraRaptorshared it, so those guys can just replicate it and rebench, there is aboslutely no right to only have scores beat by higher clocked CPUs, if that were the case we might as well get rid off all benches but CPU-Z valid.
  11. tiny diffrence there is that the .dll i changed is not part of the benchmark, but of the OS, if you change some runtimes in the OS that makes 3d effi better why not? I mean we run intel open cl runtime on ryzen 3 for gpupi, i run vb runtime from win7 on non win7 OS diffrence here is just that one is runtime from microsoft and other one is 3rd party runtime they are both not used as intended, but undoubtably work.
  12. It's reproduceable (and very consistent), allso i didn't invent this tweak, it has allready been public on the bot for a while now
  13. No idea about real temp, evap is -95ish, NB is on small SS i build for that purpose (running -30ish)
  14. Don't feel like there's any point for ln2 on these, it's pure fsb limit anyways
  15. TAGG

    kingston hypers

    7700K sold, added some reject hypers
  16. That is abolute no from me, i mean i don't care if you leave 05s globals, but it's still 3D bench since you need modern GPU to compete... Moving it to 2D just screws things up same reason for no globals for TS Physics, you allso need DX12 GPU there, right?
  17. I don't like seeing a PCMark with globals, would assume that leads to the usual "Tweaks" sooner or later, other than that any reason for not giving any single core geekbench globals?
  18. TAGG

    kingston hypers

    L3014, DFIs SOLD
  19. TAGG

    kingston hypers

    One more pricedrop on the 7700K added Xeon L3014
  20. TAGG

    kingston hypers

    Pricedrops on what's left
  21. Here you go https://discord.gg/nwutMx
×
×
  • Create New...