Announcement

Collapse
No announcement yet.

Gotta be a record for SuperPi 32M

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Snafu
    started a topic Gotta be a record for SuperPi 32M

    Gotta be a record for SuperPi 32M

    It's gotta be record for the slowest time posted to date. I ran SuperPi on an unclocked PIII-1GHz system.

    The result was 4h-9m-9s! Can you beat that!

  • Guest's Avatar
    Guest replied
    Sorry for bringing this back up from death, but...

    This thread made me wonder, if I can slow-down SuperPi significantly, to produce long computing times on relatively modern hardware, like AthlonXP CPU's. Most these old board can disable L1 and L2 caches, some even the SSE instructions support and even that disabling all but L1 does not affect the performance much, disabling L1 is "super-killer" of performance.

    So that makes me wonder, how slow it will go on the SuperPi 32M test.

    For the reference, all these machines can do SuperPi 32M w/o fail or crash or any other problems, all the computers have replaced their original bad caps to good ones and all these computers are 100% stable and I can run on th SuperPi 32M tests repeatedly w/o any trouble... BUT!

    Here we come. I got starting scores 12h 28min per loop, or even 17h 59min per loop:





    The later would produce (18hx24) a 432h long benchmark (18 days), and that is 1100MHz AXP using 256MB SDRAM (PCchips M810LR mobo).

    However, and that is where things get really bad, it always crash. Sooner or later it crash on any machine with disabled L1 cache in bios I run it. A good example is P4 3.4GHz CPU that fail w/o caches too:



    I would once again stress, that the machines are completely stable, there is no way that the error is hardware related. Not when it happen like 10x times.

    ...

    So I stopped trying the cache and I added CPU stressing (for example latest CPU-Z v1.73) that run at high priority, while SuperPi do run at low priority, causing it to slow-down to point that the whole calculation of 32M will took about 18 to 24h, depending on settings of the priority and stuff.

    Now imagine my surprise, when this also started and continued crashing. Numerous attepmts did not yield one SINGLE result. I'm "a bit" surprised and I cannot explain this behaviour. Normally systems w/o cache works just deadly slow (Aquamark3 is running on Duron 750 at 30x7.5 without caches since 13.8. and it is hardly in the half at the time of the posting - 27.8.), but works. No crashes or other problems (unless I click too much, one have to be carefull, even mouse have like 10sec lag when moved in Aquamark 3 ) happen, so my question is deadly simple - anyone have any idea, why this is happening?

    Ano no, please don't say stability. These boards ARE reliable and stable. They can run SuperPi 32M each day releatedly, but not w/o caches or when slowed down, witch is weird at least.

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Will SuperPI fit on the hard drive or will it have to run off the 5 1/2 floppy?

    Leave a comment:


  • Snafu
    replied
    I got a good one. My dad just handed me a Zenith Slimline 286 lappy with 640K of mem. I checked it out and got it to boot into windows 3.0

    Now I need to see if I can get superpi to run on it

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Maybe, just maybe I can beat that......... Unleashing the P negative 4 A.D.



    Attached Files

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Top of posted link right side....

    Thread: Slowest SuperPi scores ever
    http://www.xtremesystems.org/forums/....php?p=1526742

    Leave a comment:


  • Snafu
    replied
    now that is incredible

    Can you post teh link to the full thread? I think I want to take a gander at this.

    I will have to see if I got anything that can compete. Wonder what OS he was running, win32 perhaps?

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Just goes to show how far computers really have come
    This thread cracks me up, good job guys. lol

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Warning......thread Necro !!!

    This is not 32M........but so far....this guy wins hands down:

    http://www.xtremesystems.org/forums/...2&postcount=22

    46 hours + for 1M !!!

    Leave a comment:


  • Snafu
    replied


    I love seeing the times on these old machines. It is cool that folks even have some around and working

    Leave a comment:


  • Guest's Avatar
    Guest replied
    What??! You only beat me by like 27 hours, I still have the er... "Bronze Medal" woot!!

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Originally posted by nutcase
    Ok.. I give up before I even start
    A very wise decision!... Because HERE is the result of System 3: the winner, eh loser, eh what ever... scored 38h 44m 23s!

    Look and weep:
    Attached Files

    Leave a comment:


  • nutcase
    replied
    Ok.. I give up before I even start

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Geez you already better than tripled my score and just for an extra kick you are going to beat your own??? And I used to think you were civilized!!!

    I guess all is fair in the battle of "overclocking" !!! Hehehe......

    Alright Snafu, come and get 'im!!! :deviltail

    Leave a comment:


  • Guest's Avatar
    Guest replied
    Originally posted by Just Learnin'
    Aww Freddy stop making me sweat please... I knew it was too good to last!!

    To the mighty 486's!! Wield the death blow and let me return to the lesser ranks as a lowly PWN'ED subject once more!!!

    Alas, I may still breath easy as long as that lowly knave Snafu doth not hold the record!!


    Ok, stop sweating: the time has come!... My System 2 has finished the Super PI (v1.1e) 32M Test in: ... HEAR THE DRUMS ... ... 35h 35m!!!

    And to make it even worse, System 3 is about to finish its run and even though it is the most modern 486 EVER, it is a little slower than System 2: it is PCI (instead of the good old VLB ) based and it has an Enh. Am486DX5-160 in L1-WriteBack mode with 512 kB (instead of 1 MB) of WriteBack L2 Cache, 64 MB of 50 ns EDO (instead of FastPage) memory running at 40 MHz and a 120 GB IBM Deskstar 120GXP at UDMA-66 (instead of PIO Mode 5)!!!

    Leave a comment:

Working...
X