Why is windows faster on my macbook than OSX?

2»

Comments

  • Reply 21 of 40
    irelandireland Posts: 17,802member
    Quote:

    Originally posted by Ichiban_jay

    .



    That is a very good point! I never thought of that.
     0Likes 0Dislikes 0Informatives
  • Reply 22 of 40
    kickahakickaha Posts: 8,760member
    I'm really confused by your experience, or your expectations, JeffDM. I have 512MB on my PowerBook G4, and frequently hit 5-6BG of VM *without* leaking apps. Maybe twice a day I get a 3-5sec pause as the VM catches up with swapping out/in a couple of memory hogs (Lotus Notes, I'm looking at you, you bastard), but it's certainly 'usable'.



    Do you perhaps not realize that the Virtual Memory column includes mapped files? ie, if three apps point to a shared library of 300MB, then you'll see 900MB of VM being 'used', but they're actually just pointing at the file on disk and treating it as if it *were* in RAM, and pulling it in only as necessary. This inflates the VM values tremendously for most people's actual use. A better way of measuring your *actual* VM use is to take a peek in /var/vm and see how much is being used in there among the swap files.
     0Likes 0Dislikes 0Informatives
  • Reply 23 of 40
    vineavinea Posts: 5,585member
    Quote:

    Originally posted by skatman

    What are you coding that 512 MB is not enough? Perhaps you should make the code more efficient?! I've coded a lot of things in my life ranging from word processors to large scale molecular dynamics simulations for clusters and there aren't a lot of things that 512 MB in not enough for while you're actually coding...



    3D terrain visualization using managed direct X and the .NET framework. Not to mention when I boot my machine it shows 400MB used because of all the cruft required by IT (virus checkers, altris, timbuktu, spyware scenners, 3rd party firewall, backup software, etc). I turn a lot of that off but boot time is silly. Even with everything off XP takes around 200-300MB.



    In any case, if you're using emacs or something that's fine. But Visual Studio takes around 190MB IIRC. 1GB is too little RAM for what I do. I end up paging all the time especially if I leave outlook running. 2GB would be much nicer as the app takes around 200MB+ in debug not including the web server required for the web service stack.



    These days very few things take under 3MB of ram to run. Look in task manager sometime. 512MB for a developer is penny wise and pound stupid as bricks.



    Vinea
     0Likes 0Dislikes 0Informatives
  • Reply 24 of 40
    kickahakickaha Posts: 8,760member
    Jeez.



    Okay, I admit, I'm not running a GUI IDE, and instead using TextMate or SubEthaEdit, but criminy... that's just amazing. I routinely work on datasets over a GB in size, and I quickly found that the algorithms were much more important than anything else to creating an effective system in 512MB of RAM.



    (My first cut at my research tool took over 6hrs to finish on a small set of test data... the final version took under 2sec. Algorithmic complexity is important.)
     0Likes 0Dislikes 0Informatives
  • Reply 25 of 40
    vineavinea Posts: 5,585member
    Quote:

    Originally posted by Kickaha

    Jeez.



    Okay, I admit, I'm not running a GUI IDE, and instead using TextMate or SubEthaEdit, but criminy... that's just amazing. I routinely work on datasets over a GB in size, and I quickly found that the algorithms were much more important than anything else to creating an effective system in 512MB of RAM.



    (My first cut at my research tool took over 6hrs to finish on a small set of test data... the final version took under 2sec. Algorithmic complexity is important.)




    Its not algorithmic complexity. Its just the nature of modern development with frameworks and IDEs. Hello World is probably a MB in C#.



    I used to do real time work in VxWorks where you have to worry about resources to a great degree. Here its just I develop on my laptop instead of my desktop because I can take it with me and it was ordered as a regular laptop meant to run Office, not as a dev laptop.



    Vinea
     0Likes 0Dislikes 0Informatives
  • Reply 26 of 40
    placeboplacebo Posts: 5,767member
    OS X always feels slower than Windows because there's more eye candy and overall a larger RAM dependency. Live with it.
     0Likes 0Dislikes 0Informatives
  • Reply 27 of 40
    kim kap solkim kap sol Posts: 2,987member
    Quote:

    Originally posted by Ichiban_jay

    .



    I'm glad you pointed that out.
     0Likes 0Dislikes 0Informatives
  • Reply 28 of 40
    jeffdmjeffdm Posts: 12,953member
    Quote:

    Originally posted by Kickaha

    I'm really confused by your experience, or your expectations, JeffDM. I have 512MB on my PowerBook G4, and frequently hit 5-6BG of VM *without* leaking apps. Maybe twice a day I get a 3-5sec pause as the VM catches up with swapping out/in a couple of memory hogs (Lotus Notes, I'm looking at you, you bastard), but it's certainly 'usable'.



    I wasn't saying anything about the VM crap ( I should have left that column out), but I'm still saying that for me, OS X need 1.5GB of real memory to be efficient so I don't have to wait when I switch tasks.
     0Likes 0Dislikes 0Informatives
  • Reply 29 of 40
    kickahakickaha Posts: 8,760member
    I guess we just have vastly different workflows or expectations. I have one app that has a problem with pauses, and that's Lotus Notes. It doesn't matter if I'm using 1GB of VM, or 11GB (like now), it pauses. Safari will pause for a moment or two if I have some insane number of windows open, or if it's been a long while since a cache purge. Other than those? Teh Snappie(tm).



    Not bad for 512MB. I'm curious what apps you use that cause these problems? Photoshop, by any chance?
     0Likes 0Dislikes 0Informatives
  • Reply 30 of 40
    http://www.p2pnet.net/story/8723



    Seems that mach has alot to do with OSX being slow. I was reading that the other day give it a look over. There are some debunks to it but I dont have a Intel Mac to be able to test it on I would love to hear the results of the debunk in real world uses.
     0Likes 0Dislikes 0Informatives
  • Reply 31 of 40
    kickahakickaha Posts: 8,760member
    Good god, is Sekhon still getting press over this?



    http://www.roughlydrafted.com/0506.linuxmyth1.1.html



    http://www.roughlydrafted.com/0506.linuxmyth2.1.html



    Short answer: Sekhon screwed up his code. He didn't optimize it for MacOS X, like he had for Linux. In addition, his OS X code was doing a large number of memory allocation/deallocations for no reason. His results are therefore, bogus.
     0Likes 0Dislikes 0Informatives
  • Reply 32 of 40
    Quote:

    Originally posted by Kickaha



    Short answer: Sekhon screwed up his code. He didn't optimize it for MacOS X, like he had for Linux. In addition, his OS X code was doing a large number of memory allocation/deallocations for no reason. His results are therefore, bogus.




    Well I mean I have to say when I am running things in OS X the exact same programs I run in linux they are quite a bit slower and I get beachball. Was just wondering Of course this is PPC.
     0Likes 0Dislikes 0Informatives
  • Reply 33 of 40
    archstudentarchstudent Posts: 262member
    Quote:

    Originally posted by SlicerDicer

    Well I mean I have to say when I am running things in OS X the exact same programs I run in linux they are quite a bit slower and I get beachball. Was just wondering Of course this is PPC.



    is it possible we are talking about completely different hardware? Yes I think it is.. Be precise and give details of your OS versions, which software you are using, what hardware you are using etc otherwise we cannot draw any useful logical conclusion from what you have said.
     0Likes 0Dislikes 0Informatives
  • Reply 34 of 40
    Well when I run say for instance 1080i HDTV on my mac it pukes but when I run it under linux its silky smooth. These are just raw mpeg2 files.



    This is using Mplayer of course on both systems. None of the things like quicktime even will play it at all.
     0Likes 0Dislikes 0Informatives
  • Reply 35 of 40
    kickahakickaha Posts: 8,760member
    Mplayer using which libraries on each system?



    Are these libraries optimized to the same degree on each system?



    All sorts of variables come into play...
     0Likes 0Dislikes 0Informatives
  • Reply 36 of 40
    archstudentarchstudent Posts: 262member
    Quote:

    Originally posted by SlicerDicer

    Well when I run say for instance 1080i HDTV on my mac it pukes but when I run it under linux its silky smooth. These are just raw mpeg2 files.



    This is using Mplayer of course on both systems. None of the things like quicktime even will play it at all.




    Yea, but you are clearly talking different hardware here.
     0Likes 0Dislikes 0Informatives
  • Reply 37 of 40
    Quote:

    Originally posted by Archstudent

    Yea, but you are clearly talking different hardware here.



    Clearly differnet hardware yes but for a system that is suposto be the _ultimate_ multimedia system that runs HDTV I kind of find it lacking. And the same issues happened with a friends Intel Mac.



    Regardless I think if there was a truely good solution you would not be talking how hardware is differnet but would give me a direct answer on what to use.



    Beachball hell happens all the time with these videos I happen to believe that its the mach portions of the kernel. I use Linux for all but one system that is a Mac. Unless you have a solution then please dont waste my time I asked for real life solutions that could be duplicated to show that OSX on the whole is faster but its clearly held back by mach.
     0Likes 0Dislikes 0Informatives
  • Reply 38 of 40
    lundylundy Posts: 4,466member
    If you are beachballing, and you know enough about what that means to say it's the Mach kernel that's at fault, then you should easily be able to generate a sampling using Apple's performance tool made for that purpose and show us that the kernel is causing the beachball. It's straightforward.
     0Likes 0Dislikes 0Informatives
  • Reply 39 of 40
    archstudentarchstudent Posts: 262member
    Quote:

    Originally posted by SlicerDicer

    Clearly differnet hardware yes but for a system that is suposto be the _ultimate_ multimedia system that runs HDTV I kind of find it lacking. And the same issues happened with a friends Intel Mac.



    Regardless I think if there was a truely good solution you would not be talking how hardware is differnet but would give me a direct answer on what to use.



    Beachball hell happens all the time with these videos I happen to believe that its the mach portions of the kernel. I use Linux for all but one system that is a Mac. Unless you have a solution then please dont waste my time I asked for real life solutions that could be duplicated to show that OSX on the whole is faster but its clearly held back by mach.




    How can I give you a direct answer when I don't even know what hardware you are talking about? Why do you even expect me to have any direct answers anyway? As Lundy said, if you know so much about mach kernals they why can't you specify some technical information.



    HOWEVER, I would recommend straight away using VLC rather than mplayer.. Also, some macs have a power saving cpu setting.. make sure your performance is set to maximum if your mac has such a setting.
     0Likes 0Dislikes 0Informatives
  • Reply 40 of 40
    greggwsmithgreggwsmith Posts: 523member
    Ok, I think i have the solution. I just ordered 2 1gig sticks of RAM and a 100Gig 7200rpm hd.
     0Likes 0Dislikes 0Informatives
Sign In or Register to comment.