Jump to content
HWBOT Community Forums

TASOS

Members
  • Posts

    1512
  • Joined

  • Last visited

  • Days Won

    35

Posts posted by TASOS

  1. 4 hours ago, Antinomy said:

    When you thought everything should be found after 20 years, lol.

    Another update to you!
    Change list:
    1) Added support for Mobile Pentium 4 Prescott CPUs. EIST is enabled so you can use any MSR utility of choice to change the multiplier (CrystalCPUID, CPUMSR, RMClock, etc.)
    2) Added tRAS values 9 and 10 to BIOS setup
    3) Fixed help message for Advanced memory settings section
    4) Set default system date to 01/01/2024

    http://valid.x86.fr/6pmnfz

    Should work with P4C800-E, P4P800 2.0 and P4P800 SE. Issues are reported with P4C800 (timings are not set). I don't have this board so can't check myself. P4P800 series (-SE too) work with memory in slots 1 and 3 only.

    benchBIOS v1.2.zip 434.21 kB · 1 download

    How do you edit the asus bios ? which tool ?
    How did you implement the mobile microcode ? you have the microcode as a file in hand ?

    edit
    Many years ago i was trying (with no luck) to extract the microcode from an Abit bios and implement it to Asus ... but never managed to do it.
    I always had some kind of checksum mismatch stopping me 😄

     

  2. On 4/11/2024 at 8:23 PM, kaliz said:

    when testing different openCL version for G7400 the difference in scores are also noticable. Seems like a version of 1.2 is faster then 2.1 and 3.0. Not sure which one. 

     

    Does anyone know which version is the fastest for Alder Lake non K ? thanks

    Not sure if an older version is faster ...

    People who know , will probably not tell.
    But now at least you have most of the available versions in hand , so you can test for yourself.

     

  3. This platform and it's bios modding , is still interesting in 2024.

    I want to thank all members , participants here for their shared knowledge and i hope they will continue to support this thread if they can.

    and again , THANK YOU all.

     

     

    • Like 3
  4. 4 hours ago, Mr. Fox said:

    No, I am sorry I only have those two versions saved. Intel has a bad practice of taking down old versions of drivers when they are replaced by newer versions and I was lucky to have saved those versions because they are difficult to find. The do the same thing with XTU and there are some older CPUs that only work correctly with older versions of XTU.

    Yeap , that's true and unpleasant.

     

    2 minutes ago, Mr.Scott said:

    I will check.  I may have those. 

    Thanks Scotty

  5. 1 hour ago, TAGG said:

    Thats well over a grand in GPUs...

    Make it actually afordable to go out and buy cards to have fun, one ATI one Nvidia and NV17/RV280 then it's like cheapaz but with GPUs we'd never get for cheapaz cause too old :D Maybe consider making Nvidia one 01 and ATI one 03 running single test of 03 is no fun.

    To be honest , i took as granted that almost all of us , have atleast one of each NV core.

     

    But still , cant be that expensive ... (i see most price range is from 15 to 70 euro and the geforce256 is the most expensive at 90++)

    https://www.ebay.at/sch/i.html?item=395320642025&rt=nc&_ssn=henengelber0&_pgn=1

    https://www.ebay.at/sch/i.html?_nkw=geforce+4+ti&_sop=2

     

  6. My proposal is classic.
    Plain and simple stages , no unicorns needed for someone that wants to have fun.

     

    3DMark03
    Nvidia GeForce (AGP) history revival

    Stage 1 : 1x NV10 score

    Stage 2 : 1x NV15 score

    Stage 3 : 1x NV20 score

    Stage 4 : 1x NV25 score

    Bonus Stage : 1x NV28 score (GF4 Ti 4200 or GF4 Ti 4800SE only)

     

    PiFast
    Fastest SDRAM

    Stage 1 : 1x socket370 score

    Stage 2 : 1x socket462 score

    Bonus Stage : 1x socket478 score (Willamette core only)

     

     

     

    • Like 1
  7. 10 hours ago, mcmanus9442 said:

    Latency is pretty much similar, but A die can also run MUCH tighter Trfc as opposed to M die. 

    What about tFaw and other performance secondary and tertiary timings ?

    I am also trying to figure out ,

    Why do people who bench aircooled tend to prefer the later versions of M-die sticks ?
    Lower voltages (and temps) for the same speed and primary timings compared to A-die ?

    Does an average good IMC have the same behaviour with a 32GB A-die kit and a 48GB M-die kit ?

     

    p.s.
    Sorry for asking more general questions in a review thread.

    • Like 1
  8. On 3/19/2024 at 7:36 PM, Mr. Fox said:

    ... the second variation of M-die used on the 24GB modules overclocks higher than the A-die and requires less voltage than the A-die.

    I think , this is the key point.

    A friend from the States was telling me the same thing about the Patriot extreme 5 kit , where the 8200 48gb kit (M-die) is performing better than the 8200 32gb kit (A-die)

    But what about latencies ?

    • Like 1
×
×
  • Create New...