Mixing CPU boards in Everest machines - SGI

This is a discussion on Mixing CPU boards in Everest machines - SGI ; Here's my understanding of what you can get away with in Everest-class machines: 1. All combinations of R4400 boards work as long as the fastest boards are in the lowest-numbered slots. 2. Mixed-speed R8k boards don't work. 3. Can't mix ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: Mixing CPU boards in Everest machines

  1. Mixing CPU boards in Everest machines

    Here's my understanding of what you can get away with in Everest-class
    machines:

    1. All combinations of R4400 boards work as long as the fastest boards
    are in the lowest-numbered slots.

    2. Mixed-speed R8k boards don't work.

    3. Can't mix R4k4/R8k/R10k boards in the same system.

    Is that correct? What else should I know about mixing-and-matching?
    The goal is to guide hardware scrounging to beef up a Challenge XL.

    --
    David Evans dfevans@bbcr.uwaterloo.ca
    Ph.D. Candidate, Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
    University of Waterloo "Default is the value selected by the composer
    Ontario, Canada overridden by your command." - Roland TR-707 Manual

  2. Re: Mixing CPU boards in Everest machines

    David Evans wrote:
    > Here's my understanding of what you can get away with in Everest-class
    > machines:
    >
    > 1. All combinations of R4400 boards work as long as the fastest boards
    > are in the lowest-numbered slots.


    Doesn't matter which slots.

    >
    > 2. Mixed-speed R8k boards don't work.


    I've not tried this but I don't think there will be a problem.

    >
    > 3. Can't mix R4k4/R8k/R10k boards in the same system.


    Yep.

    >
    > Is that correct? What else should I know about mixing-and-matching?
    > The goal is to guide hardware scrounging to beef up a Challenge XL.
    >


    Don't forget memory interleaving to speed up access to memory on the MC3s.

    Cheers,
    Simon


  3. Re: Mixing CPU boards in Everest machines

    In article <40492248@derwent.nt.tas.gov.au>,
    Simon wrote:
    >David Evans wrote:
    >> Here's my understanding of what you can get away with in Everest-class
    >> machines:
    >>
    >> 1. All combinations of R4400 boards work as long as the fastest boards
    >> are in the lowest-numbered slots.

    >
    >Doesn't matter which slots.
    >


    That's good to know.

    >> 2. Mixed-speed R8k boards don't work.

    >
    >I've not tried this but I don't think there will be a problem.
    >


    I'm not sure where I heard this so I may be imagening it.

    >>
    >> Is that correct? What else should I know about mixing-and-matching?
    >> The goal is to guide hardware scrounging to beef up a Challenge XL.
    >>

    >
    >Don't forget memory interleaving to speed up access to memory on the MC3s.
    >


    Right, I knew about that one.

    Thanks a lot.

    --
    David Evans dfevans@bbcr.uwaterloo.ca
    Ph.D. Candidate, Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
    University of Waterloo "Default is the value selected by the composer
    Ontario, Canada overridden by your command." - Roland TR-707 Manual

  4. Re: Mixing CPU boards in Everest machines

    On 2004-03-06 07:13:38 +1100, dfevans@bcr10.uwaterloo.ca (David Evans) said:

    > Here's my understanding of what you can get away with in Everest-class
    > machines:
    >
    > 1. All combinations of R4400 boards work as long as the fastest boards
    > are in the lowest-numbered slots.


    Yes, but watch out; if you mix 150 and 200/250 MHz boards, your
    processor diagnostics on boot will take ages. It's apparently because
    the 150s are 3 volt and the 200/250s are 5 volt. At least, that's what
    I've heard.

    best always,

    Dave @ ATMOS

    --
    Dafydd Williams
    Project Lead
    ATMOS Software Productions Pty Ltd

    web: http://berserkir.ambrosia.net/~pipeline/viker/
    email: pipeline@atmos.com.au


+ Reply to Thread