WoW/OS X is much slower than WoW/Windows on MacBook Pro - why?

24

Comments

  • Reply 21 of 62
    kim kap solkim kap sol Posts: 2,987member
    It's not 'it could be a lot of things' in this particular case.



    One of the main problem with the test, and Gabe says so, is that both ran with max settings so there's no reason to believe that 'Full Screen Glow' was turned off during the Mac test.



    Blizzard says (and you can check it out in the link I provided) that full screen glow kills performance on Macs. I'm not saying WoW for OS X performance will suddenly be on par with WoW for Windows performance but I'm saying it won't be anywhere near twice as slow.



    Can someone with an iMac CD and Boot Camp do a test without full screen glow under OS X and Windows so this thread can die?



    I know I've said all this and it'll again fall onto deaf ears and the thread will still be overflowing with "it could be a number of things". "of course, Mac graphics drivers are not as optimized as Windows graphics drivers", once I stop posting.



    I'm not saying the full screen glow is Blizzard's fault. I don't know who's fault it is...it could be Apple's drivers, Blizzard's code, or both...but in general, OpenGL performance on Mac isn't 'much slower' or 'twice slower'.



    There are many instances where OS X beats Windows in OpenGL performance...look no further than Cinebench. There's also a Bare Feats test that shows Quake 3 for Mac beating Quake 3 for Windows by almost 50% (although this one's a bit hard to believe.)
     0Likes 0Dislikes 0Informatives
  • Reply 22 of 62
    gene cleangene clean Posts: 3,481member
    Quote:

    Originally posted by tonton

    Well, until Vista comes out, at least OS X file searches will always be 1000 times faster than Win.



    Google Desktop Search.
     0Likes 0Dislikes 0Informatives
  • Reply 23 of 62
    chuckerchucker Posts: 5,089member
    Quote:

    Originally posted by Gene Clean

    Google Desktop Search.



    Third party app. Doesn't count.
     0Likes 0Dislikes 0Informatives
  • Reply 24 of 62
    Is it possible that OS X, being a relatively young operating system, just does some things well and some things poorly, vis-a-vis Windows?



    There's no question that OS X is more secure, user friendly, and, most importantly, more modern (ex Unix shell) than Windows.



    However, is it possible that OS X has weaknesses, such as its ability to run games?



    This is a sincere question.



    Quote:

    Originally posted by kim kap sol

    It's not 'it could be a lot of things' in this particular case.



    One of the main problem with the test, and Gabe says so, is that both ran with max settings so there's no reason to believe that 'Full Screen Glow' was turned off during the Mac test.



    Blizzard says (and you can check it out in the link I provided) that full screen glow kills performance on Macs. I'm not saying WoW for OS X performance will suddenly be on par with WoW for Windows performance but I'm saying it won't be anywhere near twice as slow.



    Can someone with an iMac CD and Boot Camp do a test without full screen glow under OS X and Windows so this thread can die?



    I know I've said all this and it'll again fall onto deaf ears and the thread will still be overflowing with "it could be a number of things". "of course, Mac graphics drivers are not as optimized as Windows graphics drivers", once I stop posting.



    I'm not saying the full screen glow is Blizzard's fault. I don't know who's fault it is...it could be Apple's drivers, Blizzard's code, or both...but in general, OpenGL performance on Mac isn't 'much slower' or 'twice slower'.



    There are many instances where OS X beats Windows in OpenGL performance...look no further than Cinebench. There's also a Bare Feats test that shows Quake 3 for Mac beating Quake 3 for Windows by almost 50% (although this one's a bit hard to believe.)




     0Likes 0Dislikes 0Informatives
  • Reply 25 of 62
    chuckerchucker Posts: 5,089member
    It's possible that OS X's OpenGL implementation is sub-par. A lot of results appear to suggest this.



    It's also possible that OS X's graphics drivers are sub-par.



    Finally, it's possible that OS X's architecture generally delivers lower per-application performance and focuses more on overall performance, or that its algorithm for perceived performance vs. real performance works very much differently.



    However, this thread is too highly speculative to be of any value.
     0Likes 0Dislikes 0Informatives
  • Reply 26 of 62
    gene cleangene clean Posts: 3,481member
    Quote:

    Originally posted by Chucker

    Third party app. Doesn't count.



    Why not? It's searching files just fine.
     0Likes 0Dislikes 0Informatives
  • Reply 27 of 62
    chuckerchucker Posts: 5,089member
    Quote:

    Originally posted by Gene Clean

    Why not? It's searching files just fine.



    Don't be silly. The point was to compare Windows XP to Mac OS X Tiger. Tiger contains fast file searching (in fact, so did Panther). XP does not. Third-party software, or even first-party add-on software has nothing to do with it, regardless of its price.



    For the many features Mac OS X still lacks or has broken support for, you can't state either "yeah, but there's some thing you can download and then it's much better".
     0Likes 0Dislikes 0Informatives
  • Reply 28 of 62
    placeboplacebo Posts: 5,767member
    Google Desktop, and MSN Desktop Search which is MICROSOFT PROVIDED AND FREE AND THUS SHOULD BE COUNTED, are both cample replacement equivalents to Spotlight.
     0Likes 0Dislikes 0Informatives
  • Reply 29 of 62
    kim kap solkim kap sol Posts: 2,987member
    Quote:

    Originally posted by Placebo

    Google Desktop, and MSN Desktop Search which is MICROSOFT PROVIDED AND FREE AND THUS SHOULD BE COUNTED, are both cample replacement equivalents to Spotlight.



    So where do we draw the line then...if a line must be drawn. If no line can be drawn, I can name quite a few apps that run much faster under OS X than under Windows.



    Please, tell us what should count and what shouldn't...
     0Likes 0Dislikes 0Informatives
  • Reply 30 of 62
    kickahakickaha Posts: 8,760member
    I think he did. If it's provided for free by the OS vendor, include it.



    Personally, I'd rather see the line at what ships in the box, just because many people aren't going to be savvy enough to go grab the tool online. (Yes, believe it or not, there are still tons of people out there in that category.) But the above distinction is valid, and I can't *really* argue with it.



    Given the above line, I'd include MSN Desktop Search, but not Google's version. That's 3rd party, and I can't see a valid rationale for including those. Otherwise it becomes an intractable fanboi argument. "Oh yeah? Well I know about application RandomObscureApp that shows a 2% improvement over your piece of crap, so MY OS IS BETTER!" Feh. Mindless drivel, that.



    (Although I'm still trying to figure out what a 'cample' replacement is... )
     0Likes 0Dislikes 0Informatives
  • Reply 31 of 62
    mr. hmr. h Posts: 4,870member
    Quote:

    Originally posted by kim kap sol

    So where do we draw the line then...



    It depends upon whether you are trying to compare operating systems or platforms. Whilst the first is an interesting exercise, the later makes more sense as that is what you use, the platform as a whole, not just the OS in isolation.
     0Likes 0Dislikes 0Informatives
  • Reply 32 of 62
    kickahakickaha Posts: 8,760member
    Except that the latter is almost impossible to create meaningful comparisons for that are relevant to all but niches areas. Outside of specialized realms, it's hard to find two people who use the *exact same apps* in general. The best you can really do is say "App A runs better on OS X, but App B runs better on OS Y." and then let people mix and match the apps they use to try and decide which OS to select.



    Unfortunately, very precious few people want to do that, because it's like, y'know, hard and stuph, and instead they'd like to state unequivocally that one OS is better than another based on individual data points.



    Besides, look at what we're discussing - not just the actual OS, but the applications that come with it. Is that not a platform? You pay for a box o' software, which now includes the OS and some apps. That's the default install, which you are right, is a platform, but it makes a natural line in the sand for comparisons, IMO.



    Once you bring 3rd party apps into it, the permutations become essentially boundless, and any given comparison loses whatever statistical significance it may have had.
     0Likes 0Dislikes 0Informatives
  • Reply 33 of 62
    mr. hmr. h Posts: 4,870member
    Indeed, the user needs to think of what they want to achieve, and then decide which platform lets them do that best.



    However, there are a lot of users out there that fit into certain user-segments, such as image editing, video editing, publishing, engineering, consumer etc., and it is possible to compare Windows and Macintosh as platforms in each of these areas.
     0Likes 0Dislikes 0Informatives
  • Reply 34 of 62
    kickahakickaha Posts: 8,760member
    Quote:

    Originally posted by Mr. H

    Indeed, the user needs to think of what they want to achieve, and then decide which platform lets them do that best.



    However, there are a lot of users out there that fit into certain user-segments, such as image editing, video editing, publishing, engineering, consumer etc., and it is possible to compare Windows and Macintosh as platforms in each of these areas.




    Er, only vaguely and roughly, IMO. Heck, define 'consumer', and a list of the apps they'll be using. \



    I mean, yeah, you can fuzz the definitions out, but the fuzzier you make them, and the more generalized, the less pertinent and relevant they are.
     0Likes 0Dislikes 0Informatives
  • Reply 35 of 62
    mr. hmr. h Posts: 4,870member
    Quote:

    Originally posted by Kickaha

    Er, only vaguely and roughly, IMO. Heck, define 'consumer', and a list of the apps they'll be using. \



    I mean, yeah, you can fuzz the definitions out, but the fuzzier you make them, and the more generalized, the less pertinent and relevant they are.




    I agree. I just think they are likely to be more relevant to someone than a straight comparison of OSes.
     0Likes 0Dislikes 0Informatives
  • Reply 36 of 62
    zengazenga Posts: 267member
    World of Warcraft is the only game I play..

    Read below to see where I play it..

    I don't care if it's slower than WINSHIT..

    I use MAC OS.





    "for the alliance"
     0Likes 0Dislikes 0Informatives
  • Reply 37 of 62
    kickahakickaha Posts: 8,760member
    Quote:

    Originally posted by Mr. H

    I agree. I just think they are likely to be more relevant to someone than a straight comparison of OSes.



    Agreed - which is why I always try and talk to someone about the specific apps and tasks they need, when helping them decide on a system to buy. The more specific you can get, the more useful to that particular individual. Aggregate data in this case only works for aggregate groups of people, not individuals.



    Funny how we went from a specific app to generalizing about the quality of OS, and them from comparing OSs to specific apps, eh?
     0Likes 0Dislikes 0Informatives
  • Reply 38 of 62
    kickahakickaha Posts: 8,760member
    Just so you don't feel left out of your own thread...



    Quote:

    Originally posted by JavaCowboy

    Is it possible that OS X, being a relatively young operating system, just does some things well and some things poorly, vis-a-vis Windows?



    Well... *yeah*. No OS is perfect. Every OS has flaws. I can't imagine anyone arguing otherwise with a straight face.
     0Likes 0Dislikes 0Informatives
  • Reply 39 of 62
    Quote:

    Originally posted by Kickaha

    Just so you don't feel left out of your own thread...







    Well... *yeah*. No OS is perfect. Every OS has flaws. I can't imagine anyone arguing otherwise with a straight face.




    Yes, but the mere suggestion that OS X has inferior gaming performance seems to provoke a vitriolic reaction among some of the Mac zealots.



    I'd be very interested in seeing some objective, detailed benchmarks for some popular Windows games versus their OS X universal binary (or Intel only binaries, if applicable) equivalents such as:



    1) WoW (DUH!)

    2) Doom 3

    3) Quake 4

    4) Unreal Tournament



    Granted, this is probably not a sufficiently representative sample, but these are the only major games I know of that are available as universal binaries.



    My gut feeling is that OS X performance is noticeably inferior to Windows, but I have no evidence to back this up.



    However, once Leopard comes out, this could change dramtically. There are rumours that Apple wants to position Intel OS X and their implementation of OpenGL as a competitive gaming platform. There's also a Cringely rumour that Apple is considering replacing the Mach kernel with a monolithic kernel, which could also help overall performance.
     0Likes 0Dislikes 0Informatives
  • Reply 40 of 62
    kickahakickaha Posts: 8,760member
    Quote:

    Originally posted by JavaCowboy

    Yes, but the mere suggestion that OS X has inferior gaming performance seems to provoke a vitriolic reaction among some of the Mac zealots.



    Only the clueless ones.



    Notice that the only venom you dredged up here was because you were claiming, indirectly, that it reflected on the OS as a *whole*. That's just ludicrous.



    But saying that gaming on the Mac is behind the optimization curve will generally get you a big 'ol "Well duh..." from most folks here. No news there.



    Quote:

    I'd be very interested in seeing some objective, detailed benchmarks for some popular Windows games versus their OS X universal binary (or Intel only binaries, if applicable) equivalents such as:



    1) WoW (DUH!)

    2) Doom 3

    3) Quake 4

    4) Unreal Tournament



    Granted, this is probably not a sufficiently representative sample, but these are the only major games I know of that are available as universal binaries.



    My gut feeling is that OS X performance is noticeably inferior to Windows, but I have no evidence to back this up.



    WoW seems to be an aberration at this point, from what I've seen. Sorry, no citations off the top of my head, but I do know that ArsTech did a quick rundown using Q4, I believe, and the Mac version held up pretty well. For some reason 85-90%fps is sticking in my head, but I'm likely wrong.



    In other words, yes, it's slower... but not by much. Definitely room for improvement.



    Quote:

    However, once Leopard comes out, this could change dramtically. There are rumours that Apple wants to position Intel OS X and their implementation of OpenGL as a competitive gaming platform. There's also a Cringely rumour that Apple is considering replacing the Mach kernel with a monolithic kernel, which could also help overall performance.



    Ayup. Those blasted funnels, while clever and elegant back in the mid80s, have rather outstayed their welcome on most modern chips.
     0Likes 0Dislikes 0Informatives
Sign In or Register to comment.