Results 1 to 7 of 7

Thread: 7-zip benchmark. Bloomfield vs Core2Quad

  1. #1
    Member
    Join Date
    Sep 2007
    Location
    Denmark
    Posts
    870
    Thanks
    47
    Thanked 105 Times in 83 Posts

    7-zip benchmark. Bloomfield vs Core2Quad

    just to share some information.

    Strictly a secret between us

    7-zip benchmarks (8 threads with bloomfield and 4 threads with Q9300)

    Bloomfield 3.2GHz = 18171 MIPS
    Q9300 @ 3.Ghz = 10643

    Nice huh ?

  2. #2
    Programmer Bulat Ziganshin's Avatar
    Join Date
    Mar 2007
    Location
    Uzbekistan
    Posts
    4,497
    Thanks
    735
    Thanked 660 Times in 354 Posts
    this is funny, but not very informative. can you please show results with the same amount of threads (4, 2 and 1, if possible)?

  3. #3
    Member
    Join Date
    Sep 2007
    Location
    Denmark
    Posts
    870
    Thanks
    47
    Thanked 105 Times in 83 Posts
    I dont have the Bloomfield equipment anymore

    But some info.

    The bloomfield is 4 core with hyperthreading ( 8 virtuel cores)
    The Q9300 is "only" 4 cores. that why i teste with the differnet numbers of threads

    Howevwr i have some info with only 4 thrad in bloomfield i will psto them later.

    But 4 threads is slower then 8 threads
    4 threade with manually affinity to even cores is faster then just 4 threads where windows allocates them.

    So its damn importent with low numbers of threads to avoid hyper threading colletions

  4. #4
    Programmer Bulat Ziganshin's Avatar
    Join Date
    Mar 2007
    Location
    Uzbekistan
    Posts
    4,497
    Thanks
    735
    Thanked 660 Times in 354 Posts
    http://www.bit-tech.net/hardware/200...5-965-review/9

    without SMT, it's only 3.5% faster in the same environment. and since we know that 7zip doesn't actually use more than 2 threads, it's the real difference

    for rar, difference is 13% without SMT and 34% with SMT

  5. #5
    Member
    Join Date
    May 2008
    Location
    Antwerp , country:Belgium , W.Europe
    Posts
    487
    Thanks
    1
    Thanked 3 Times in 3 Posts
    Quote Originally Posted by SvenBent View Post
    I'm still busy writing my mini review on Bloomfield cpu. (its a killer cpu for compression)
    Any chance we can see your review on the net ?

  6. #6
    Programmer Bulat Ziganshin's Avatar
    Join Date
    Mar 2007
    Location
    Uzbekistan
    Posts
    4,497
    Thanks
    735
    Thanked 660 Times in 354 Posts
    he cited numbers from the page i mentioned

  7. #7
    Member
    Join Date
    Sep 2007
    Location
    Denmark
    Posts
    870
    Thanks
    47
    Thanked 105 Times in 83 Posts
    Quote Originally Posted by Bulat Ziganshin View Post
    he cited numbers from the page i mentioned
    Now i didn't
    these are my own results

    anyway some more info

    4threads (windows allocation)
    9038 MIPS


    4threades (affinity set to even numberede core)
    12718

    So by dividing the threads around to avoid Hyperthredaing collision
    you gain more than 33% boost in 7-zip benchmark performance


    --- edit ---
    BTW since the NDA has be liftet all my test are here (no gaming test)
    http://www.dailyrush.dk/features/reviews/1369/
    Last edited by SvenBent; 3rd November 2008 at 23:38.

Similar Threads

  1. QuickLZ ZIP - new zip/deflate library
    By Lasse Reinhold in forum Forum Archive
    Replies: 23
    Last Post: 1st October 2007, 22:08

Posting Permissions

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