Slashdot Mirror


Early Tiger Benchmarks Show Slight Speed-Ups

GatorMarc writes "Geek Patrol has published early speed benchmark tests on Tiger. Despite the fact that Tiger is still in development, the results are promising. Could we see a similar performance improvement as we did upgrading from Jaguar to Panther?"

39 of 111 comments (clear)

  1. Debug symbols by Anonymous Coward · · Score: 3, Insightful

    Was it built with debugging symbols on?

  2. Move along by isd_glory · · Score: 5, Informative

    From the article: "Looking at the results, it appears that Tiger is as fast (if not faster) than Panther in all areas except for UI performance." His numbers show Tiger barely edging out Panther, using Xbench (a purely synthetic benchmark) as the test basis.

    I've tried Tiger out on my G4 powerbook, and have actually noticed a *decrease* in Xbench ratings, despite an overall "snappier" feel. Maybe the increase isn't really going to happen for those without 64 bit machines. Then again, its a pre-relase, so there is plenty of room for change.

    1. Re:Move along by foidulus · · Score: 5, Insightful

      Don't most of the speed increases come closer towards the end of the development cycle? I know that is usually the case for games.

    2. Re:Move along by radicalskeptic · · Score: 5, Insightful

      Yeah. And Tiger is to be released in the first half of 2005. Which gives them between 5-11 months to make these changes.

      I think it's just too early to tell how fast the final release is going to be, since there's probably 3/4 of a year more development to be done.

      --
      WARNING: If accidentally read, induce vomiting.
    3. Re:Move along by isd_glory · · Score: 4, Informative

      Apple implements Xfree86 4.3.0 for both Tiger and Panther. As far as I can tell, they seem to be running comparably.

    4. Re:Move along by RevAaron · · Score: 3, Insightful

      Quite. Some even consider premature/early optimization to be a bit of a curse, or at least not a very good idea. In software development in general, definately not just limited to games.

      --

      Working toward a usable PDA environment in the spirit of Newton OS: Dynapad
    5. Re:Move along by Ianoo · · Score: 5, Informative

      IIRC, Apple's X implementation is a fork of XFree with some custom enhancements to run in rootless mode under Aqua and integrate with the native window manager and hardware drawing acceleration. So it's really "neither". If you were so inclined, presumably you could load up Darwin/XNU in console mode and compile and run either XFree or X.org fullscreen, since the code bases are almost identical at this point.

  3. Is this version of Tiger by foidulus · · Score: 5, Interesting

    built with the new version of gcc that Apple is releasing with Tiger. The compiled code(on both G4 and G5, moreso on the G5 which they used) is supposed to be much more efficient for certain operations than the previous version of GCC. Wonder if they used this?
    more info here

  4. Give me a break... by Chief+Typist · · Score: 5, Insightful

    This release was obviously pulled together for the conference -- a Herculean effort by the engineers at Apple to show what will be available in a year for now. A wonderful release for us third-party developers!

    No one in their right mind is going to think that this release is fit for benchmarking. There may be some gains that are side effects of internal changes (new versions of gcc, etc.), but anyone with a clue will realize that minimal optimization has been done.

    When they say DEVELOPER PREVIEW they mean it...

    -ch

    1. Re:Give me a break... by (mandos) · · Score: 5, Insightful

      So, if we're already seeing gains in the "unoptimized" version, then we should see some great speedups in the final version a year from now. Seems to me that these benchmarks do serve a purpose by letting us know that we are getting "Bigger, Better, and Faster" rather then the Windows version of that, which is typically slower.

      Mike Scanlon

    2. Re:Give me a break... by Orick · · Score: 5, Interesting

      Exactly. With Windows, a new OS release seems to mean you need to upgrade your computer. The history for Macs seems to be more like a new OS release allowing you to use your older computer a little longer.

      --
      Kirby Reviews

    3. Re:Give me a break... by johnpaul191 · · Score: 5, Interesting

      it is funny..... someone was playing with my G4 400 last week and commented how it seemed a lot faster than their much newer machine.... turns out they were still running 10.2.x and i was running 10.3.x. i have to say i'm happy that 10.3 keeps the machine usable that much longer, i hope 10.4 is at least as usable.

    4. Re:Give me a break... by Ineffable+27 · · Score: 4, Insightful

      What you're describing is definitely true for 10.0 --> 10.1 --> 10.2 --> 10.3; I suspect that much of that has to do with how new OS X is/was, and how many improvements they needed to make to the 'not ready for prime time' releases 10.0 and 10.1. (Don't be surprised if 'snappiness' of performance starts to 'plateau' with subsequent OS X releases.)

      But I don't think your generalization holds prior to OS X. Many users who moved from OS 8 --> OS 9 found they had to upgrade their RAM to achieve the same performance and responsiveness (depending on their machine). I suspect the same was true during the transition from System 7 --> OS 8.

      Panther is, aggravatingly, still not nearly as responsive as OS 9 on the same hardware. They still have some way to go. I hope Tiger finally closes this gap.

      --
      "He'd be a broader guy if he had dropped acid once." - Steve Jobs on Bill Gates
    5. Re:Give me a break... by shawnce · · Score: 2, Informative

      Panther is, aggravatingly, still not nearly as responsive as OS 9 on the same hardware. They still have some way to go. I hope Tiger finally closes this gap.

      Do you ever run more then one application at a time?

      Mac OS X is far more responsive and efficient then Mac OS 9 under any amount of load.

    6. Re:Give me a break... by frankie · · Score: 4, Insightful
      No, you should give us a break. You're being willfully obtuse.
      • When you upgrade from Win2K to WinXP (and presumably to Longhorn or whatever), bloat increases and speed decreases.
      • When you upgrade from 10.0 to Puma to Jaguar to Panther (and presumably to 10.4), features increase AND speed increases.
      That's the comparison worth mentioning. And yes, I agree that Steve is rude for not providing version discounts (except for 10.1 which was free, thanks Steve). Nevertheless, each upgrade has been worth my money.
    7. Re:Give me a break... by Steveftoth · · Score: 4, Interesting

      The speed up from release to release is mostly due to the fact that Apple engineers were thinking ahead when they designed the software. Not because it is less mature.

      Basically the hardware is finally catching up to the design of the software.

      I'm talking mostly about the gui system here, the command line/ unix stuff is very well optimized for what it does but the GUI has yet to be enhanced.

      Aqua, Quartz, Cocoa and the drawing APIs for OSX were designed such that they can be accelerated easier by future graphics cards easier. Remember that when OSX first came out the graphics cards that were in most macs was the Rage Pro or below. Now that much more powerful graphics cards are coming out it's possible to use that power for GUI, but an API was needed to access it. Aqua, Quartz, allow some access and the new CoreVideo allows even more access to the underlying graphics hardware. Windows API doesn't allow access to the graphics hardware in the windowing envoriment, any advanced features you have to use DirectX/OpenGL and those APIs are not designed with a multi-program, windowing enviroment in mind (well OpenGL can handle it better then DirectX ever could on SGI but not so on Windows X86 machines, DirectX is designed for games).

  5. Missing the big picture by Cycline3 · · Score: 5, Insightful
    I think everyone has missed the bigger picture here. As OS X evolves - it gets better AND faster. I run both PCs and Macs at home and the office and I can say that with every new version of Linux and Windows, the new features have cost in terms of performance. Only OS X has repeatedly delivered speed and features. Say what you will, but Apple is just doing something right with OS X. I can't imagine being stuck in a Windows (or Linux) only world after using OS X for an extended time. Kudos to Apple. I for one am eagerly awaiting Tiger's official release.

    -Sean

    1. Re:Missing the big picture by bhima · · Score: 2, Interesting

      I was thinking this is because OS X is increasing it's 64 'bitness' or are you using it on a g4?

      --
      Nothing in the world is more dangerous than sincere ignorance and conscientious stupidity.
  6. Perceived speed vs throughput by caseih · · Score: 5, Informative

    These two benchmarks seem to be continuously confused by slashdotters over the years. Of course it is debatable which is really more important. I think OS X has definitely concentrated on perceived speed, which is good because that is what the user "feels" and sees as he interacts with the computer. This does in no way mean the whole OS is faster; it just feels faster.

    OS X has definitely not improved dramatically in throughput and raw horsepower over the last few releases. In fact I'm sure it has decreased slightly. Sacrificing a little of that throughput for smoother rendering yields a significant percieved speedup that the users really like. I would say that every release of OS X has gotten a little heavier and is a little bit slower. A sacrifice I'm willing to make for my pretty Panther desktop, though.

    Windows has gotten slower on both counts over the years.

    Linux's throughput has actually increased fairly dramatically in the last year or two. Unfortunately as the weight of the desktop comes to bear, and due to current weaknesses in X11 and the toolkits (most notably the lack synchronized redraw which make resizing appear really slow), the perceived speed of linux has seemed to decrease with recent distros. The 2.6 kernel provided some speedup in this area (the interactive scheduler), but there is still much work to be done.

    The experimental X server from www.freedesktop.org implements a lot of features that will lead to a perceived speedup. For example the damage and composite extension reduce redraws when windows are uncovered. Work is also being done to allow windows to resize smoothly (synchronizing the widget drawing and compressing events). Even with the vesa driver and no acceleration, it feels faster than normal accelerated X.org. Again perceived speed vs throughput. Give it a try. It's cool.

    Fortunately I think Linux will deliver on both benchmarks. Expect exciting things over the next year from linux desktops.

    1. Re:Perceived speed vs throughput by zhiwenchong · · Score: 5, Interesting

      Perceived speed is actually quite important; from the perspective of a personal computer user, it is quite possibly more important than actual throughput. It influences a user's beliefs about his own productivity on the machine, which in turn affects actual productivity. I don't know about you, but on a machine where widgets just zip, I work faster... slow GUIs have the effect of insidiously slowing me down.

      Somehow the responsiveness of the Panther UI leaves something to be desired. (I'm running OS X 10.3.4 on a 1GHz G4/768MB RAM)

      It is a well known fact that the UI in Cocoa apps are a little sluggish, and in some cases more sluggish than in Windows apps. High-end G5 Mac users probably don't notice it, but it is actually quite obvious on midrange Macs. One gets used to it, of course, and very soon one ceases to be bothered by it--but one cannot help but feel it when one uses a Windows machine at work.

      John Siracusa at ArsTechnica actually did an informal test (on scrolling and such, somewhere in this review) and recorded the results in a Quicktime movie file. He compared the speed of Cocoa controls to Win32 controls.

      This is also evident if you compare Cocoa to X11. Even the UI in X11 apps running under OS X is more responsive than Cocoa, especially in scrolling. I'm not entirely sure why this is but some people attribute it to overheads in Cocoa.

      In my own experience, I have come to the (unscientific) conclusion that for now, Windows apps do seem more responsive than Mac OS X apps on comparable machines (1GHz G4/768MB RAM vis-a-vis a 1.8 GHz Pentium 4/256MB RAM). That doesn't mean the Win32 apps are actually faster, mind you, just that their UIs feel that way. This would one area where Apple could make improvements on, and I think it will.

      P.S. Having said all that, Mac OS X has many productivity enhancing features that compensate for its UI sluggishness: robust and fast multitasking, extensive keyboard control, aesthetically pleasant UI, automation capabailities, exceptionally well-designed tools, creatively crafted free apps (like Butler and Tofu) from the community etc. Yes, I cannot deny that I am actually more productive and creative in je ne sais quoi ways on my Mac than on Windows.... for the most part, it's just more pleasant to work on a Mac.

      You're generally right about "perceived speed". This article explains how:
      10 Things Apple did to make Mac OS X faster.

    2. Re:Perceived speed vs throughput by John+Newman · · Score: 5, Interesting
      It is a well known fact that the UI in Cocoa apps are a little sluggish
      At least some of that sluggishness is by design. I used to be rather frustrated at the glacially slow appearance and disappearance of sheets in OSX. It was most apparent when opening or saving files, and when using a certain spreadsheet that relies on sheets in all of its wizards. Just that one element made the whole UI feel slow, especially compared to Windows. Until I discovered that the speed of sheets is set by a system variable, which I immediately cut by three-quarters. Presto! Lightning-quick sheets, and a much-happier me.

      So I think much of the perceived slowness of OSX's UI has nothing at all to do with how fast the machine is actually capable of performing the pretty functions, even on rather humble hardware (like my oldish iBook).

      That and the "action-on-release" instead of "action-on-click" thing. It may make the UI feel less "snappy", since all actions are delayed by however long it takes you to take your finger off the mouse button, but it certainly makes for a more pleasant experience.
    3. Re:Perceived speed vs throughput by j1walker · · Score: 5, Informative

      There's an article about this on macosxhints.com here:

      http://www.macosxhints.com/article.php?story=20040 51208143172&query=sheet+speed

      I'm off to try this myself now...

  7. Tests by whfsdude · · Score: 3, Insightful

    Keep in mind these tests were done on the G5's. Tiger is the only version of Mac OS X to have 64bit support. One has to wonder if it is really faster on non-64 bit operating systems.

    1. Re:Tests by Lars+T. · · Score: 4, Informative
      64-bits doesn't make normal operations faster. 64 bits makes big number crunching faster, and allows you to access more memory, and that's it.

      To be more precise: it makes 64 bit integer number crunching faster (as opposed to floating point).

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

  8. OS X, keeps getting better by The+Lost+Supertone · · Score: 4, Interesting

    I always find it funny, Microsoft can be the most successful software vendor on the planet, but a company with what maybe 5% of their value (if that) can come up and best them time and time again. With that much cash shouldn't they be able to buy all the developers possible?

    1. Re:OS X, keeps getting better by tobes · · Score: 3, Interesting

      It's because Apple is developing for their user base of esthetically correct, ergonomically aware individuals. MS is developing for the AOL/QVC/Clear Channel/Office Depot crowd.

    2. Re:OS X, keeps getting better by jeffehobbs · · Score: 3, Funny


      (Even though I'm just proving your point for you)

      It's because Apple is developing for their user base of esthetically correct,

      That's aesthetically correct, you insensitive clod.

      ~jeff

    3. Re:OS X, keeps getting better by hunterx11 · · Score: 3, Funny

      That's æsthetically correct, you insensitive clod.

      --
      English is easier said than done.
    4. Re:OS X, keeps getting better by Anonymous Coward · · Score: 2, Funny


      That's option-' for those keeping score at home - point goes to hunterx11 and the ball remains in play

    5. Re:OS X, keeps getting better by selfish · · Score: 2, Informative

      Ligatures rule!

      --
      This is not an official Fugazi sig.
  9. All about "Debugging Code" vs "Debugging Symbols" by mbessey · · Score: 4, Informative

    "What is it about debugging code? It's almost mythical the level of speed upds people think they'll gain once all this debug code has gone."

    People are ignorant...

    Okay, for you non-developer folks playing along at home, here's what you need to know:

    1. "Debugging Symbols" are extra information stored (typically) in the program's executable file, that make it easier to run that code under a source-level debugger and see the right names for variables, functions, and other program entities.

    In general, any software that's released to customers (yes, INCLUDING beta versions) will have the debugging symbols "stripped" from the programs, because they're not useful for the customer, and also because many companies fear (for no apparent good reason) that they represent a leak of confidential information. There is a slight performance penalty on some platforms for running a binary with symbols, but it's only going to effect load time, and only by a tiny fraction.

    2. "Debugging Code" is not as well defined of a term, but sometimes, early development builds of software will include extra checks on the integrity of key data structures, or extra error-recovery code. Again, in general, these sorts of builds would never be distributed to a customer.

    3. So why do programs generally improve in performance towards the end of the development cycle? It's for the rather self-evident reason that the software has to be working correctly before it's worth the effort to try to make it run faster. In fact, optimizing performance before the feature set has been frozen is one of the classic blunders of software development!

    -Mark

  10. a graphics card question for those in-the-know by bodrell · · Score: 3, Interesting
    The new Core Image features seem really cool. However, the ATI Rage M6 (?) card in my Powerbook G4 is not listed under supported graphics cards. They do say "Core Image automatically scales as appropriate for systems with older graphics cards, for compatibility with any Tiger-compatible Mac," but what does that really mean? Which of the Core Image features am I not going to be able to use? Will any of those unuseable features be necessary for rudimentary GUI operations?

    Put another way, what do the newer cards have that I don't? Core Image looks to me like the best new feature in Tiger, if I can actually use it. Quartz Extreme gave me a pretty good speed increase when it came out, and I would expect CoreImage to allow most graphics programs to work snappier, but that's all irrelevant if my card isn't up to snuff.

    --
    Si la vida me da palo, yo la voy a soportar Si la vida me da palo, yo la voy a espabilar
    1. Re:a graphics card question for those in-the-know by Jeremy+Erwin · · Score: 4, Informative

      The ATI Rage M6 chipset is underpowered, and is unlikely to be supported by Quartz Extreme, much less Core Image. Apparently, it's a Rage 128 derivative. found in the 550 and 667 MHz Powerbook G4s. Quartz Extreme relies on hardware support for GL_EXT_texture_rectangle, which Apple notes is only supported by "Radeon, Radeon Mobility, Radeon 7500 Mobility, Radeon 8500, Radeon 9000, Radeon 9200, Radeon 9600, Radeon 9700, Radeon 9800,GeForce 2MX, GeForce 4MX, GeForce 3, GeForce 4Ti, GeForce FX." Presumably, you can verify this for yourself by running a GLInfo program.

      Apparently, it's less flexible than the proposed GL_ARB_texture_non_power_of_two.

      Core Image is much more sophisticated, and levergaes the vertex and pixel fragment capabilities of an current generation OpenGL card.

    2. Re:a graphics card question for those in-the-know by Chucker23N · · Score: 4, Informative

      "Radeon Mobility" -> RageM6.

      So yes, it supports Quartz Extreme indeed. But as those models only came with 16 Megs of VideoRAM, IIRC, support is limited.

      As to Core Image, it is not designed to improve performance, but to add, for example, visual feedback through effects. For example, when opening a Dashboard widget on a fully Core Image-supported computer, you get a water ripple effect. If your graphics hardware is word, the effect simply gets discarded - which doesn't remove any functionality. You can already see similar things in Panther, like the fast user switching cube effect, which doesn't appear on slower machines.

  11. OpenGL? No problem. by revscat · · Score: 2, Insightful

    The most common lament I hear on /. (about Mac)is that there's no port of a specific game. Make a Mac-only game with the same 12-year old boy appeal as Halo, and you might sell yourself some Macs.

    Wait.. are you saying that more games should be ported to the Mac, or that there should be games developed *solely* for the Mac? If the first, that's already taken care of: anything written in OpenGL works on the Mac. For example: Halo.

    If the latter... that's kinda silly. No one would want to limit their market appeal. As great as I think Macs are, I would never expect a software company to limit their product to a single platform, no matter which platform we're talking about.

  12. Apple's Strategy On Games by The+Lost+Supertone · · Score: 2, Interesting

    It's clear Apple is out to lunch on games. However this isn't that bad. What they need to do is cause a paradigm shift on games! Back to the console using the computer. What they REALLY should start doing is, since both the PS3 and GameCube are going to be PPC based (I doubt M$ would play along) find out some way to work with the GameCube and PS3 for a sweet gaming experience on the Mac. If the DS has some standard wireless tech it should be able to work with any Mac. Heck think about it, we already have some great emulators available. How about the iGames ROM Store? Have everything from classic SNES games, to N64, to GameCube, to whatever revolution will be called. Get SEGA on board, who knows maybe even Sony could be convinced to play along with PS1 tittles. Apple could move on games, but to compete with Microsoft, they would need to do it in an entirely different way.

  13. My goodness... by FredFnord · · Score: 2, Insightful

    You've run XP acceptably on 6 year old machines? Well, you are probably the only one in the world. Many PC-compatible 6 year old machines are limited in RAM to between 128 megs or so, which is not enough for XP. Basically, our three-year-old laptops (Dell business machiens, limited to 256 megs of RAM) are bad enough; I can't imagine what you'd do with a 400 mHz machine with 128 megs of RAM running XP. That's assuming it's still working; 2/3 of our three-year old (Dell) laptops have failed more than once in their third year of operation. Once the extended warrantee is up, they're getting pitched. On the other hand, all our Macs are doing just fine, except the laptop that got abused by an airline baggage handler. (Which I won't claim is anything other than luck... if you've got ten heavily used computers of any kind in a company and none of them have gone in for service in over three years, you're just plain lucky.)

    As for apple locking things out of the BIOS, well, you're right that there are some six-year-old Macs that won't run Mac OS X without your using some little tricks to get them to work. However, MOST six-year-old Macs (the first generation of iMacs, the PowerBooks, the Blue & White PowerMac which was introduced in 1998) work fine with it, and all of them will take at least 512 mb of RAM, which is plenty. It's hard for me to blame Apple for not supporting the vintage 1997 (that would be seven years old, though some were sold in 1998) beige G3s, with their onboard SCSI, their ADB-connected keyboards and mice, and the (pathetic) Rage II+ graphics chips that many of them had. If you want them to work, you can get them to work, Apple just makes it clear that they're not supporting them.

    Whereas if you buy a retail version of XP and install it on a 7-year-old PC and you call up Microsoft, they'll be happy to spend as many hours as you want on the phone with you to get it up and running. On your dime. Just remember that half your hardware is probably not supported by XP/2000 drivers, and that the tech support phone call is liable to cost you as much as one of those ultra-cheap PCs.

    -fred

    --
    Sign #11 of Slashdot overdose: You see the phrase 'moderate Republican' and you wonder if that would be a +1 or a -1.
  14. Re:gracias by Jeremy+Erwin · · Score: 2, Informative

    Just to clarify, Core Image requires least a Radeon 9500 or GeForceFX 5200 to use all the features, although the library does scale. These cards support floating point vertex and fragment shaders.

  15. Don't forget compiler optimizations by hargettp · · Score: 3, Insightful

    Although you do make good points about debug symbols mostly taking a space instead of slowing things down, one can't forget the compiler optimizations that are often enabled in release builds but not enabled in debug builds. Without those optimizations (particularly for C++ apps, as many commercial software products are still C++), some operations may be orders of magnitude slower.

    Hence, not suprising that debug builds are often perceived as slower.