Results 1 to 2 of 2

Thread: speed factor

  1. #1
    Junior Member
    Join Date
    Oct 2010
    Posts
    26

    speed factor

    Hi,
    I'd like to ask a small question, is the speed up factor bounded above by the number of cores in a GPU?
    I mean if I implemented an algorithm sequentially and run it on CPU then implemented it in parallel and run it on GPU of 16 cores, is the maximum speedup factor that I could achieve 16x???
    or is there other factors that might increase this factor... could you please help me get more info about this?

  2. #2
    Senior Member
    Join Date
    May 2010
    Location
    Toronto, Canada
    Posts
    845

    Re: speed factor

    Code :
    I mean if I implemented an algorithm sequentially and run it on CPU then implemented it in parallel and run it on GPU of 16 cores, is the maximum speedup factor that I could achieve 16x???

    It's not that simple at all. Not all "cores" have the same performance, and given two "cores" made by different companies it's hard to define which one is "faster", because the definition of which is faster will depend on the workload you are trying to execute.

    This is a very complex matter.
    Disclaimer: Employee of Qualcomm Canada. Any opinions expressed here are personal and do not necessarily reflect the views of my employer. LinkedIn profile.

Similar Threads

  1. Curious speed up
    By tayloral in forum Interoperability issues
    Replies: 0
    Last Post: 05-23-2012, 05:13 AM
  2. -O0 and -O3 speed difference
    By deNorma in forum OpenCL
    Replies: 2
    Last Post: 01-15-2010, 02:43 PM

Posting Permissions

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