Results 1 to 6 of 6

Thread: P-1 coordination thread

  1. #1
    Senior Member engracio's Avatar
    Join Date
    Jun 2004
    Location
    Illinois
    Posts
    237

    Discussions

    Hey guys,


    Read OmbooHankvald's FAQ. Looks simple enough even for a noob. I'd like to try factoring for a while. What is a good range to try for an amd duallie and do I create a separate instance for each cpu?

    In sieving 1T is a good range per cpu, what is it in factoring. Thanks.


    e
    Last edited by Joe O; 09-29-2005 at 09:23 PM.

  2. #2
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    About 10000 for starter should take you a couple weeks.

    Use B1=60k B2=700k for starters and see what happens, you can run stage1 on the first cpu then transfer all of the intermediates to the second folder for stage2. This way your able to use more memory.

    If you'd like I can setup a worktodo.ini for you.

    J

  3. #3
    Senior Member
    Join Date
    Jun 2005
    Location
    London, UK
    Posts
    271
    9750000 9780000 Greenbank 2 [abandoned] passed by
    10100000 10101000 Greenbank 0 [complete]

    Half the latter range was checked with B1=200,000 B2=3,000,000, the other half with B1=B2=800,000. Still nothing :-(

  4. #4
    Senior Member engracio's Avatar
    Join Date
    Jun 2004
    Location
    Illinois
    Posts
    237
    Okay reserved a range, created wtd file, started p95 running and been crucnhing since this morning. Done about dozen candidates out of 488 in the range.

    (4847*2^10220391+1 completed P-1, B1=70000, Wc1: 0869A9AF) Here is the latest entry.




    Few questions: I inserted the answer I found on the forum in parenthesis.

    The B1=70000 is that about right for now. When do I get to do B2. ( I increased the default memory settings on p95.24.14 from 8mb to 256mb since i have 512 on this box. B1=105000, B2=1365000. Will wait untill all 488 pfactor are done, then we'll see.)

    As I get a factor, do I immediately try to submit it or must I wait for the range to complete, i.e. there are more to do after B1. (Yes, asap)

    Garo suggested to use the p95 route instead of the sbfactor, any major difference. (According to the poll, with the very few people factoring most use p95.)

    Currently running this range on a single AMD XP3000+, using p95v2414 are there any better version or is p95 common enought to use this version for both Intel or AMD. (The famous v2.46, all links are broken, on main dowmload page the latest is v24.14)

    Since I mostly have duallie boxes, I must run two instance of p95 with different wtd files, Correct? (Yes)

    I am sure there are more questions will pop up. (Good luck)


    e
    Last edited by engracio; 10-01-2005 at 11:13 AM.

  5. #5
    engracio, your post is a little bit, err, not straightforward. I'm not sure if you have questions, but I'll try to write some answers.
    Stage1 needs less memory than stage2. Knowing this, p95 computes the bounds optimal for factor throughput, given the max. mem available(and some other stuff).
    This is if you put Pfactor=... in the worktodo.ini.
    Of course, if you have 2 computers, on with less mem than the other, it would make sense to run stage1 on the one with less and stage2 on the other one.
    Running tests with your own chosen bounds (or only one stage) means putting Pminus1=... to the worktodo.ini.
    You have a dual processor, right? It would be the first idea to set up two instances, one with less mem, running stage1, and one with more, runniing stage 2...you get the point.
    The problem is that you have only one memory. I tried it out, at least with hyperthreading, the bottleneck memory access slowed them down to the speed of one instance.
    As p95 does not have genuine support for multiprocessor systems, it makes sense to optimize on your own, running one times p95, and one times something else which doesn't need so much access to mem, like - the sieving client. (That's what I remember.) Or perhaps even PRP. This, you hve to try out.
    It's a little bit like if you mix 1liter alcohol and 1 liter water, it gives 1.8 liter of the mix.
    More information you will find on www.mersenneforum.org
    Good luck, H.

  6. #6
    Senior Member engracio's Avatar
    Join Date
    Jun 2004
    Location
    Illinois
    Posts
    237
    hhh,

    Thanks for the reply, I think in the last 36hrs I read enough info to get going with factoring. I also realize the boxes are slightly above the threshold of stability. Got sumout error on one instance per box. Stabilized once I knock the fsb down a notch.

    Not experienced enough to know how fast the client should be but running two sub project should be a good idea until I figure out the best route for me. Should have enough time to finish my reserved range before prp catch up.


    e

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •