Apple's Press Release

[the future]

Apple is a small customer for both IBM and Intel. That Intel will deliberately change their CPU production for Apple? Not very likely. (...) IBM's future focus on multicore processors and the Cell processor will more than likely not be ready for desktop computing in the short to medium term. (...) The x86 platform for all its flaws WILL have much better performance for desktop computing in the short to medium term. Even Jobs (for all his x86 bashing and PPC evangelising) knew that the PPC path may eventually be limited. Why else would he have kept Mac OS X fully compiled for both PPC and x86? Personally, I think that his strategy is damned clever. Not only has he shown that migration to x86 from PPC is possible, but that it is ALREADY a reality.

It's true, Apple was just not important enough for IBM to concentrate on desktop (let alone laptop) performance right now. I mean, they sold their own PC business recently to concentrate on servers and consoles. It's a shame, though, as the PPC platform had/has so much potential technically. As long as Windows is dominant in the marketplace and clings to x86, so will Intel/AMD and now Apple as well.

[ThisistheEdge]

Did anyone see the keynote. It was (in typical Job's fashion) a good show. After announcing the switch he went to the system profiler on his presentation computer and it showed "Intel Pentium 4 at 3.6 GHz." Everything on Steve's Mac worked great. He went on to say that every release of OSX had been compiled for the Intel chip from the start "just in case".

[Case]

Even Jobs knew that the PPC path may eventually be limited. Why else would he have kept Mac OS X fully compiled for both PPC and x86?

Of course Jobs needed a backup plan. Depending on one supplier is too dangerous. But initial concerns would likely be supply quantity and negotiations, not the roadmap. The double platform development has been a public secret ever since Rhapsody DR1/Intel was shown. Was that 1998?

[miksu]

The boundless optimism and arrogance of the Mac fanboy...

Better be Mac fan than wearing stupid "Intel inside" shirt

Benchmarks are usually more or less optimized. It's well known, that intel for example has compiler that can't really generate usable code, but performs well for benchmark purposes. So that's the x86 side of honest benchmarking

Fact remains, there is still no guarante Intel can deliver nothing more than stable supply of chips. Roadmaps are these things "we believe and hope, that we can possibly do this thing two or three year later".

Don't want bother repeating more technical facts, everyone interested can read those in Arstechnica, so far best source on silicon information. In depth. Info on both G4 and Pentium M, as well as 970 and P4.

BTW, why Intel has officially abandoned it's Netburst-architecture (P4) as base for next-gen solution, in favor of Pentium M, which dates back to Pentium Pro? It wasn't able to deliver or scale.

www.arstechnica.com

[rumplestiltskin]

This tangential topic is beginning to sound like the cholos arguing about their Monte Carlos' hydraulic lift systems.

OSX on Intel. It's happening; get over it.
-or-
OSX on Intel. It's about time; rejoice.
-or-
OSX on Intel. Will it kick ass? Yes? Fine.

[willrob]

>>OSX on Intel. Will it kick ass? Yes? Fine. <<

Currently it's at 33% slower than a G5. But I assume Apple will optimize in order to make up for the lack of Altevec.

http://www.geekcoffee.net/archives/2005/06/apple_developer_1.html

Also, quoting from the Macintouch website:

Think Secret has some unofficial Xbench results from a 3.6GHz Pentium running Apple's Rosetta system at WWDC 2005:

Overall, the Intel Mac are scoring between 65 and 70 with Xbench, a far cry from the 200+ scores higher-end G5 systems reach. The CPU test is landing in the high teens compared with scores of 100 to 200 for G5 systems, but that appears to be primarily due to lackluster FPU scores. ... The Intel Mac performed substantially less well than the dual-2.5GHz G5 at Thread test, scoring an 82 compared to 225. In the Computation Thread test the Intel Mac scored a respectable 110 compared to 155 in the G5, but the G5 blew the doors of the Intel Mac in the Lock Contention test, scoring a 420 to the Intel Mac's 66.
The Memory Test tells a similar story: overall the Intel Mac scored a 214 to the G5's 378, but the Intel Mac actually exceeded the G5's Stream Memory Test: 351 to 319. The G5 trounces the Intel Mac at the system memory test, however, scoring a 464 while the Intel Mac musters a 154.
The Intel Mac scored a 125 on the Interface Test, compared to a 380 for the G5. The Intel Mac scored well in both the Quartz graphics and OpenGL graphics tests--almost matching or exceeding dual-2.5GHz G5 score--although it's unknown which video card is powering the system. There has been some speculation that Apple's emracement of Intel processors will also allow the company to take advantage of off-the-shelf PC video cards.

[rumplestiltskin]

...and this is why we are waiting until 2006 for the next gen Intel CPUs.

Better cut back on that caffeine, buddy.

[willrob]

Less caffinated evaluation copied from the xcelerateyourmac website:

Comments/Info on the Mac X86/Pentium 4 Development system - (revised per request)

" I'm going to keep this brief, so please write me with the questions you have and any tests you want run on one of the dev kits. I will have one of my own next week as well.

First, the thing is fast. Native apps readily beat a single 2.7 G5, and sometimes beat duals. Really.
(I asked about real-world apps - if any were already available in native code-Mike)
All the iLife apps other than iTunes, plus all the other apps that come with the OS are already universal binaries....

They are using a Pentium 4 660. This is a 3.6 GHz chip. It supports 64 bit extensions, but Apple does not support that *yet*. The 660 is a single core processor. However, the engineers said that this chip would not be used in a shipping product and that we need to look at Intel's roadmap for that time to see what Apple will ship.

It uses DDR-2 RAM at 533 MHz. SATA-2. It is using Intel GMA 900 integrated graphics and it supports Quartz Extreme. The Intel 900 doesn't compare favorably to any shipping card from ATi or nVidia. The Apple engineers says they dev kit will work with regular PC graphics cards, but that you need a driver. Apple does not write ANY graphics drivers. They just submit bug reports to ATi/nVidia. So, when we asked where to get drivers for better cards the engineers said "The ATI guys are here." He's right, they've been in the compatibility lab several times.

It has FireWire 400, but not 800. USB 2 as well. USB 2 booting is supported, FireWire booting is not. NetBoot works.

The machines do not have Open Firmware. They use a Phoenix BIOS. That;s right, a Mac with a BIOS.
(I asked if the Bios had any tweaks like Memory Timing which is common for many PC motherboards, although Intel OEM motherboards don't usually have any end user tweaks like that.-Mike)
They won't tell us how to get in the BIOS. I'm sure we can figure it out when out dev kits arrive.

They run Windows fine. All the chipset is standard Intel stuff, so you can download drivers and run XP on the box.

Rosetta is amazing. (see earlier post on limitations of the Rosetta emulator - it's a G3 emulator basically - will not run Altivec code, etc. and performance isn't going to be as good as native code, but most Mac apps will run on a G3.-Mike) The tests I've run, both app tests and benchmarks, peg it at between a dual 800 MHz G4 and and a dual 2 G5 depending on what you are doing.
(I mentioned to him the limitations of Rosetta (posted below)-Mike)
It's true Rosetta does not support Altivec, but most apps run on a G3, right? Rosetta tells PPC apps that it is a G3. Apps should fall back to their G3 code tree. Everyone I tested did.

The UI tests in Xbench exceed a dual 2.7 by a large margin. (other specific tests are much lower than a G5 per Xbench site results.-Mike)

I've been talking to and watching a lot of devs. There are a lot of apps from big names running in the Compatibility lab already. Some people face more pain, sure, but Jobs wasn't kidding when he said that this transition would be less painful than OS 9 to OS X or 68K to PPC.

Game devs seem optimistic. They see porting Windows/x86 to Mac/x86 as much easier. They look forward to the day they don't have to support PPC.
I was talking to a (game Developer) that said about 1/3 of the process is handling endian issues, the rest is Win32/DirectX. For the next 3-5 years, their job will be harder since they have to port to two processor architectures and most bugs *are* endian related and that they will have a hard time making the PPC versions run as well as the x86 versions.

This transition is not about current P4 vs G5. It is about the future directions of the processor families. Intel is committed to desktop/notebook and server in a big way. Freescale/IBM are chasing the embedded market and console market. Apple would have been in a lurch in 2 years.

Also, all the cell people and the AMD people need to be quiet. Apple evaluated both. AMD has the same, if not worse, supply problems as IBM. Their roadmap is fine, but the production capacity is not.

The tested Cell as well. That processor is NOT intended for PC applications. (it was designed for game systems, not as a general use CPU) The lack of out of order execution and ILP control logic creates very poor performance with existing software. Having developers rewrite for cell would have been MUCH more work than reworking for Intel. And that's what this is, you rework your codebase in ALL cases, not rewrite it. "

[DereX888]

Thank you, willrob.
Well, what did I say since begining?
It is all about MAC OS on x86, that includes having tweakable BIOS on your precious MACs, that includes drivers, and it is all about Pentium processors from Intel.
I don't care even if they date back to first ray-cathode lamps, they already outperform PPC architecture in many areas as is.
No "special" chips for Macs, just stock Intel x86/ia64 chips; most likely stock motherboards as well.

Im really glad Apple jumped on Intel's wagon, it would have been real waste to let Apple disappear from the market. I cant wait having some sort of OSX on my next PC

I think Jobs has made another 'revolution' in personal computing (perhaps even Revolution with capital R) with this switch. If everything will go well, MAC OS for Intel may be the one to overthrow Windows hegemony on PCs much sooner than anyone expected


:bow:

[rumplestiltskin]

While the roadmap to Intel from PPC doesn't have anything to do with OSX on "generic" Intel PCs, it does seem to be the logical conclusion. Steve said that there will not be a way of running OSX on an Intel box that doesn't have Apple's logo on it. However it's done (presumably with some proprietary chip on the mb), that's okay.

DereX888's wish to see "OSX on my next PC" will either have him buying a Mac or waiting until Steve finally drops the hammer on Microsoft and releases a "generic" OSX for Intel (and maybe AMD? Why not?).

There will always be differences in performance between OSX and Windoze even on the same box. IIRC, floating point and integer performance were not necessarily dependent on which processor but what calls the OS would make. (I may not have this 100% right but you all get my drift, eh?)

Always keep in mind what Steve's said: "The soul of a Mac is OSX." (And when the generic OSX is released, will that turn a Dell into a Mac? Hmmmmm.)

Frankly, I'm quite excited about the conversion. When my AppleCare expires in late 2007, it will be time for a bitchin' new Mac.

Hot dog!

[DereX888]

While the roadmap to Intel from PPC doesn't have anything to do with OSX on "generic" Intel PCs, it does seem to be the logical conclusion. Steve said that there will not be a way of running OSX on an Intel box that doesn't have Apple's logo on it. However it's done (presumably with some proprietary chip on the mb), that's okay.

DereX888's wish to see "OSX on my next PC" will either have him buying a Mac or waiting until Steve finally drops the hammer on Microsoft and releases a "generic" OSX for Intel (and maybe AMD? Why not?).

rumplestiltskin, youve got to be kidding me!
Once the "Mac OS for Intel" is out, with or without Apple's support it will be very quickly made to run on 'generic' x86 platform. Perhaps we'll have to buy some 'third-party' ROM replacement for generic mobo, or whatever, but Im sure it will be done, and I don't even doubt it for a second. Perhaps hardware (mobo?) manufacturers themselves will provide official or unofficial "MAC OS enabled BIOS" versions for their mobos, why not? When Intel locked multipliers on their chips almost all mobo manufacturers enabled FSB tweaks to beat Intel's inconvenience for customers... When PS2 originally wasn't able to read CD-R game backups - there were 'third-party' ROMs for PS2 enabling it... etc etc. Just release the genie out of the bottle, thats all what the world needs Once first MacIntel machines with their Mac OS for Intel show up on the market - Im sure within month or two there will be some solution (how to run Mac OS on non-Apple Intel machines) available
Personally I'd say it would be a giant mistake for Apple if it will "lock" its x86 OS for "MacIntel" machines only. Hopefully Microsoft-owned Apple stock is not big enough to prevent it from happening, because only this IMHO would be the only reason to halt unleashing Mac OS for x86 onto good-OS-hungry masses...

EDIT:
Just FYI - I would never buy any proprietary device if I have a choice. I will never buy a Mac just because of this, same as I have never bought Dell, HP or any PC like that. I always build the machines from the components I want exactly to my liking. I don't want Mac machine (be it PowerPC or Intel powered box). Apple will never make a penny on me same way as Dell or Sony has never made a penny on me. All I want is their great Mac OS to run on my own 'design' intel machine (well, the one I'll make in the future). I'll buy their MAC OS for Intel when its ready and if they decide to make and sell it for 'generic' x86 machine, or I'd simply download it if they won't sell it. Its up to them.

[vitualis]

Better be Mac fan than wearing stupid "Intel inside" shirt

Well, soon you'd be both simultaneously...

Benchmarks are usually more or less optimized. It's well known, that intel for example has compiler that can't really generate usable code, but performs well for benchmark purposes. So that's the x86 side of honest benchmarking

? Eh ?

The Intel compiler is used extensively to generate "usable" code. For example, VirtualDub compiled using the Intel compiler has better performance than using something else.

When I'm talking about the SAME benchmarks, I'm talking about the SAME program. For example, the same version of Photoshop for both Mac and PC. Or the same version of Mathematica. Or compiling the same source for PPC and x86 and then benchmarking.

Mac benchmarks like using ? Final Cut Pro (that nice NLE program) vs Premier on the PC are quite simply retarded. It would be like comparing MPEG encoding performance between a P4 and Athlon but using CCE on the P4 and TMPGEnc on the Athlon. It's a completely meaningless test.

Fact remains, there is still no guarante Intel can deliver nothing more than stable supply of chips. Roadmaps are these things "we believe and hope, that we can possibly do this thing two or three year later".

Of course, the fact that Intel already has working prototypes of these chips does count for anything.

BTW, why Intel has officially abandoned it's Netburst-architecture (P4) as base for next-gen solution, in favor of Pentium M, which dates back to Pentium Pro? It wasn't able to deliver or scale.

I'm sorry, but why is this a bad thing?. The last time I looked, Intel owns and makes both architectures and let us not forget they allow for binary compatibility. Yes, Netburst sucks. But the Pentium M rocks and there is no PPC equivalent to the Pentium M. The Pentium M development was not a "once off" either. Dual-core Pentium Ms (i.e., Yonah) are in development.

Currently it's at 33% slower than a G5. But I assume Apple will optimize in order to make up for the lack of Altevec.

Those benchmarks are meaningless because the benchmark program was run in Rosetta -- i.e., through an emulation layer. It would be like me running the same benchmark program on Windows and Linux, but on Linux, having to run it through Wine!

I think we should wait for "real" benchmarks before making judgements on how well Mac OS X will run on the x86 platform.

Also, we should remember that these are developer kits not the consumer product. There are no x86 Macs yet. Since these are Macs running on basically PC hardware, obviously they will have a BIOS, etc. There is no reason to expect that this will be the case in the consumer product.

I would be interested to see if Apple will really update Mac OS X to be 64 bit. Despite the hype Mac OS X is not really a 64-bit OS (not like WinXP 64-bit or 64-bit versions of Linux). It is as "64-bit" as Windows 95 was a "32-bit" OS.

Regards.

[rumplestiltskin]

DereX888,

You are playing with yourself and don't even realize it. If it all was as easy as you claim, why aren't you running OSX right now? The "hardware emulator" (for whatever chip will make a Mac "a Mac") has been tried before (Mac ROMs in a box that would plug into an AtariST). VirtualPC will do it right now for you (although you'll have to be running your Virtual PC on a Mac, eh?).

Tell you what - When you build your Mac using off-the-shelf, commercially available, non-proprietary components, come back and let us know about it.

Until then, stop your trolling and get a life.

[offline]

Being able to pick your own hardware
and still run OSX. A nice dream... but still
a dream!

[miksu]

Of course, the fact that Intel already has working prototypes of these chips does count for anything.

You can create working prototype pretty much on anything. Dual core 970 have existed as protype at least since last summer. To actually produce something working and at acceptable yields (=price) is another thing.

About the idea running on generic x86, say Asus board or Dull, Schiller has said in WWDC that it won't work anything except Apple hardware.

There is currently more info available on developer systems (seems to be generic micro-atx, maybe slightly tweaked), but propably customer products will be very different. Don't think they would be using Phoenix BIOS like things (since there is EFI or whatever) and legacy PC ports, among other things.

[cdssh]

I hope if Apple has to resort to Intel chips that is does make the OSX for Mac only. I have both Windows and Mac machines. I hate the illegal instruction errors,and I hope they don,t show up in the Mac OSX. Windows users have supported Bill Gates for years,why stop now! Windows is a very buggy operating system,and is not built to play nice in the graphics world. But I will say that they make a good gaming machine for what ever that is worth. Real world graphics work no very good. My wife works for the local newspaper,and they have left the windows world and went to Mac. 70% reduction in downtime.

[DereX888]

I hope if Apple has to resort to Intel chips that is does make the OSX for Mac only. I have both Windows and Mac machines. I hate the illegal instruction errors,and I hope they don,t show up in the Mac OSX. Windows users have supported Bill Gates for years,why stop now! Windows is a very buggy operating system,and is not built to play nice in the graphics world. But I will say that they make a good gaming machine for what ever that is worth. Real world graphics work no very good. My wife works for the local newspaper,and they have left the windows world and went to Mac. 70% reduction in downtime.

another Mac fanboy rant...
Grow up
No one has supported Gates, for years Windows was the only choice for average PC user until Linux appeared (and it is still for more advanced than average users unfortunately, otherwise Windows would be gone from PC desktops long time ago). Having Mac OS for PCs will sure dethrone Gates' almost-monopoly on a PC market.

No matter how buggy is Windows - it does the job too, try it out (and dont force it to be a Mac-clone, run Windows software on Windows and it'll run well for you; I have Win2000 machines that I haven't rebooted for close to a year; I see no difference in stability between well-configured Win2000/2003 PC and Macs at work, and actually Mac OSX was crashing more often on us than Win2K at first when we got it *nothing is bug-free* ).

[vitualis]

Agreed. A well configured Windows 2K or XP system (which means using it as a workstation and not installating dodgey hardware or software on it) is just as stable if not more so than Mac OS X. There is nothing special about Mac OS X in terms of its stability. It crashes just like everything else.

My experience with the first version of OS X and definitely OS 9 is that it is less stable than my computers running XP.

As for "real world graphics work" perhaps Adobe Photoshop and Premiere aren't "real" programs... since they have been optimised first for the x86 platform on Windows before on the Mac...

Regards.

[miksu]

Both systems propably can run close to eternity without crashing. Even old Os 9 is relatively stable (used it about 4 years, required maybe 10 resets).

But, both OSX & Win need to be restarted, if user wants to keep them up to date (security fixes etc.). Most updates require restart, so question about uptime is rather hypothethical.

[vitualis]

Agreed.

Assuming that you haven't trashed your OS with installing dodgey software and haven't been infected with spyware/trojans/etc (admittedly much easier to occur on Windows), both modern versions of Windows and Mac OS (and probably Linux) are stable enough that it would be pretty rare for the OS to go belly up on your in the middle of doing something innocuous.

Regards.

[i_wolf]

Those benchmarks are meaningless because the benchmark program was run in Rosetta -- i.e., through an emulation layer. It would be like me running the same benchmark program on Windows and Linux, but on Linux, having to run it through Wine!

Actually he made a somewhat valid point. What you are saying isn't necessarily true either. Yes Rosetta is a binary translator... however... most of the API calls made by Xbench (even when started from within Rosetta) can and are already serviced by existing OS X x86 API calls so these do not need to be emulated at all. Rosetta is able to intelligently call the native API's. Some of the OS X tests *should* run native under todays xbench.
I didn't manage to get to this years WWDC unfortunately. Have been there for a number of years though and a contact I have made who is there told me that if Rosetta encounters a PPC call to an OS X framework API it will use the native API instead of emulating. Apologies if I am explaining this badly.

[vitualis]

No. This is the PPC version of Xbench. They are emulated.

Your explanation makes sense and probably explains WHY it has such good performance (e.g., compare it to the open source PPC emulator). However, the small amounts of emulation required (i.e., the calls that can't just be "translated") will signficantly slow down the program.

The sentiment shouldn't be why the benchmark runs "so slow" but why it "runs so fast"! Frankly the performance from Rosetta is astonishing.

Regards.

[i_wolf]

Well some of it would be emulated. But anytime an OS X framework API call is made rosetta does not bother emulating, it merely calls the native OS X x86 call instead to maintain speed. From what I understand most of the tests in Xbench are based around making OS X calls.
I was reading about on the xbench forums and it appears that the majority of xbench calls are to inbuilt OS X calls.
So the benchmark while run under rosetta is somewhat indicative of the current performance of x86 under some scenarios.

One thing I would like to know myself about the new OS X x86 edition is whether or not Apple are still compiling OS X with GCC under x86. They may be utilising ICC since development kits are shipping with ICC for Apple.
Personally I believe that the G5 was an architecture that never really got to show its true potential, partly because of the way instructions are dispatched in groups and the very wide nature of its design. It requires a lot of effort to port an application from x86 to PPC 970. Secondly, somewhat like the Itanium , it is very sensitive to the choice of compiler. We conducted inhouse tests and often found that XLC showed a 70%+ speed improvement over GCC in floating point and also showed a good 30% improvement over integer. Again in a lot of cases these were worst case scenarios.
I think Apple doomed the PPC970 to a certain extent since the architecture is very dependent on the compiler for its performance (more so that x86 or prior PPC processors); shipping with GCC 3.3 was a joke.
I remember a while back Dr Craig Hunter from NASA tested a G5 dual 2GHz against a Xeon 3.2/3.4GHz (not sure which it was). His code was heavily G5 optimized. He had his results up on ars for a while. Even having hand tuned his code he noted that under GCC3.3 the G5 performed at around the equivalent of a 2.6GHz P4. He then recompiled with XLC and XLF and reported that he had gotten a near doubling of floating point results, 70% improvement on vector and 30% (similar to what we found) on integer.
I also think that its criminal that it took Apple and IBM so many years to finally implement autovectorisation in a GCC compiler (GCC4 under Tiger). I think the move to x86 may end up being a good thing since it will remove the onus on Apple to improve compilers and what not. Intel already has mature and high performance compilers readily available. Furthermore GCC under x86 is a damn sight better than GCC under PPC.
[/i]

[lordsmurf]

I use both systems all the time. I actually find OS X to be less stable than WinXP, gone are the days of OS8 and OS9 where everything worked (OS7 was hell, all those damned "bombs"). I prefer WinXP Pro and OS9 for normal use.

Both the G4 and G5 are a bit slower than Intel chips of the same timeframes. Not by much, but enough to notice when you work with video or large graphics.

Altivec is merely wasted on the "pretty" crap inside X that I tend to turn off on both OS X and WinXP systems. I need functional, not pretty. Beyond that, Altivec does nothing. Waste of code.

Vitualis and myself both seem to have similar experiences.

My wife works for the local newspaper,and they have left the windows world and went to Mac. 70% reduction in downtime.

Two words: user error. Either that or they (likely) were using old crap from a decade ago. Not uncommon in a small newspaper.

.
.
.

Apple on Intel is a move in the right direction. Eventually all computers will more or less have the same type of hardware and software. We've been moving that direction since the 1980s (started with the whole "stolen GUI" debaucle). Anybody that hasn't seen this coming is either blind or not paying attention.

[vitualis]

@ i_wolf: with regards to GCC, that's definitely one of the things Tom's Hardware Guide found when it was doing those recent (and rather illuminating) tests on the poor performance of Mac OS X servers.

GCC does a really poor job at making Altivec optimised code. I was frankly surprised to learn that Apple used GCC to compile its OS if this was the case.

Regards.

[the future]

I use both systems all the time. I actually find OS X to be less stable than WinXP

Wow, you must do some weird things if OS X is "unstable" for you. What do you mean exactly, the system actually crashes? Sorry for asking, but I'm on OS X since 10.1 and never ever had one single system crash.

My wife works for the local newspaper,and they have left the windows world and went to Mac. 70% reduction in downtime.

Two words: user error.

Isn't "user error" the standard Windows euphemism for "counter-intuitive/messed up GUI"?

[i_wolf]

I work for a software house where we work with many platforms from Windows XP to Solaris 9 and OS X. OS X without doubt in my experience (and I have a lot of experience with different operating systems) is extremely stable and certainly is more stable than Windows XP (assuming you repair permissions after updates). Primarly the main reason for this is the very nature that the same people who make the hardware also make the software so it tends to be well integrated together and I have yet to see a driver conflict under Panther or Tiger. On the other hand I have seen and continue to see blue screen of deaths under Windows XP with flaky drivers. A good example is the SIS drivers hosted by Biostar for their small form factor machines. Claimed to be compatible with Windows XP SP2, yet when the end user goes to their website and downloads/installs the latest SIS integrated video drivers... bang BSOD upon reboot. This is only one example amongst many I have seen. Incidentally these drivers are supposedly WHQL certified.
Of course OS X is lucky in that Apple doesn't have to worry about every permutation of hardware under the sun as Windows does, but OS X's weakness interms of the limited hardware configuration choice available to consumers certainly gives it an advantage in terms of robustness and stability IMHO. Different people's mileage can vary. Its Unix open source roots should give OS X an advantage in terms of stability though.
Actually the most stable platform I ever had the pleasure to work on was Irix 6.5

@virtualis : Yes I saw that review (think it was anandtech actually). Its amazing you get great performance if you move to yellow dog linux under PPC970. That Apache server performance should be a wake up call to Apple. Personally I disagree with anand's take that it is merely symtomatic of the microkernel. I believe its completely symptomatic of Apple's implementation of the micro kernel and not of the microkernel concept itself which I believe is sound and should not cause the problems seen under Apache. But thats for another debate

Kind Regards
i_wolf

[DereX888]

Lets not turn it to yet another Windows vs Mac thread.
I can run windows 2000 well and maintanance-free for years, and so are many other people. So is Mac OS9 and probably OSX.
I won't say anything about windows XP because in my opinion its a major piece of inflated beyond reasons nt kernel with bugs on top of bugs dating back to DOS probably ... but thats just my OT opinion...

The point is to have viable Windows competition for the Average Joe's PC, because without competition we have what we have - a stagnant personal computing market, where nothing is new, everything is old idea in slightly better form :/ Hardware-wise there havent been any major innovation since I dont know when, software-wise dominant OS is going to finally have replacement in the workable version of its Cairo concept dating before Windows 95 (thats Windows Longhorn if someone dont know what Im talkin about).
I was hoping Linux was *it*, but apparently it isn't and will never be - because it is "developed by geeks for geeks" as my friend put it once after trying it out, and his PC knowledge is slightly more than usual Average Joe's :/

So excuse me for 'trolling' on your holy Mac forum, but I am really glad that Apple moved to standard PC architecture, and forgive me if Im hoping for having Mac OS on intel machine, but IMO it is a major revolution for this little Macs market, and it may become major revolution (I hope so) on a 'standard PC' market as well - by giving a major push in personal computing progress.

[lordsmurf]

Try running OS X with video encoder software, pro stuff, like Bitvice or Megapeg.X

DVDSP and FCP can also have troubles.

We could blame the software, but the OS is the thing crashing.

OS X is no more or less stable than Windows XP, that's just a big myth. I'm only illustrating that while some people can show XP crashes a lot, my experience is total opposite and I'm not alone.

[miksu]

Try running OS X with video encoder software, pro stuff, like Bitvice or Megapeg.X

We could blame the software, but the OS is the thing crashing.

OS X is no more or less stable than Windows XP, that's just a big myth. I'm only illustrating that while some people can show XP crashes a lot, my experience is total opposite and I'm not alone.

I'm running those pro apps without problems, no crashing apps or OSX crashes. Have had less problems with W2k Pro (using it in internet gateway) than XP (using at work, for CAD). But that's my experience. Propably there is bunch of badly programmed apps that can cause lot of problems (for both platforms). Microstation 7 causes 'em for XP a lot.