Page 3 of 3 FirstFirst 123
Results 21 to 27 of 27

Thread: Sun Studio 12 vs. GCC3 vs. GCC4 Benchmarks

  1. #21
    Join Date
    Jan 2008
    Posts
    772

    Default

    Quote Originally Posted by kebabbert View Post
    For instance, why focus on compile time? If the resulting code is twice as slow but compiles 10 secs faster, is it good?
    It's good for developers, particularly when trying to fix bugs in large projects. Besides, I didn't see the article "focus" on compile time at all; it was just one benchmark presented alongside a bunch of others.

  2. #22
    Join Date
    Jul 2008
    Location
    Greece
    Posts
    3,778

    Default

    I think it's a good benchmark. It does exactly what it says: compares GCC 3.4 and 4.0 against Sun Studio 12. Of course we don't use those ancient GCC versions anymore, but the benchmark never mentioned it tests anything newer than GCC 4.0.

    As for testing compilation speed, any compiler benchmark has to test it, regardless of whether people consider it useless or not. It's a must to test.

  3. #23
    Join Date
    Nov 2008
    Posts
    418

    Default

    Quote Originally Posted by Ex-Cyber View Post
    It's good for developers, particularly when trying to fix bugs in large projects. Besides, I didn't see the article "focus" on compile time at all; it was just one benchmark presented alongside a bunch of others.
    Yes, maybe you are right. There was no "focus" on compile time in the article. I want to take my first post back. Apologize for that.

    Anyway, I like this test better than the others. I would like GCC people to also give their input. Not only SUN people. Otherwise it isnt fair if only SUN people can give input.

  4. #24
    Join Date
    May 2008
    Posts
    98

    Default

    I am currently very short in time because of my "official" work, but I will definitely perform some Intel vs. gcc tests on some Linux apps when possible.

  5. #25
    Join Date
    Oct 2007
    Posts
    370

    Default

    Quote Originally Posted by Ex-Cyber View Post
    It's good for developers, particularly when trying to fix bugs in large projects. Besides, I didn't see the article "focus" on compile time at all; it was just one benchmark presented alongside a bunch of others.
    yeah... not really.

    build systems typically only rebuild changed parts, and as such the compiletime is very minimal whilst developing. Furthermore, with quadcore+ workstations, its practically irellevant.

  6. #26
    Join Date
    Jan 2008
    Posts
    772

    Default

    Quote Originally Posted by Redeeman View Post
    build systems typically only rebuild changed parts
    That's the idea. Sometimes it doesn't really work, or other circumstances minimize the benefit (some projects have single files that take a while to build, for example).

    Quote Originally Posted by Redeeman View Post
    Furthermore, with quadcore+ workstations, its practically irellevant.
    Lots of people don't have such workstations yet.

  7. #27
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    4,994

    Default

    Also, some projects do not use a build system (OpenLieroX, I'm looking at you!)

Posting Permissions

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