Slashdot Mirror
Apple Switching to Intel
Steve Jobs announced at the WWDC keynote today that Apple is switching to Intel processors. MacNN has live coverage. The bottom line is that Mac OS X for the last five years has been running on Intel, the switch is expected to be complete in two years, and Rosetta will allow PPC apps to run on Intel-based Macs, transparently. If you're using Xcode, it is small changes and a recompile; otherwise, you might be seeing a lot of work ahead of you. You will be able to order the 10.4.1 preview for Intel today.
Late Friday afternoon, C|Net News published an extremely valuable trade secret about Apple and Intel, days before Apple was scheduled to announce it ( Apple to Ditch IBM, Switch to Intel Chips ). So, where's the friggin' lawsuit against C|Net to find out who leaked? Where is the judge who is going to claim that what C|Net published was "stolen property"?
6 /05/apple_intel_wheres_the_lawsuit_against_cnet.ph p
From: http://www.corante.com/importance/archives/2005/0
My prediction of when you'll be able to run Mac OS X on an x86 machine is still: never. Apple isn't a software company. They're a hardware company. Just because they're changing their processor does not mean you're going to be able to run it on your hardware.
Cyde Weys Musings - Scrutinizing the inscrutable
The powerbooks weren't cutting it and there was NO WAY to get a G5 in there.
Maybe I can get back to a 4-5hr runtime like the first generation Tibook had..
..don't panic
Keep in mind. Mac OS X is a unix OS, with lots of unixy underpinnings.
You loose *some* compatability with existing Mac apps.
More likely than not, all Linux apps will be recompilable for Mac. No sweat.
This means OpenOffice.org 2.0 will work *now*.
This means no more second-class Mac versions of popular OS apps.
Virtual PC will run *much* faster. No more cpu emulation is needed.
Vmware will run on a mac.
Plus, all the big name apps will run just as fast. Adobe, Macromedia (same company now). Not to mention the Apple Pro apps, Video stuff, etc. That stuff will be perfect.
WINE will run on a Mac. This is *HUGE*. Imagine running any Windows software, at native speeds, with OpenGL support, on Mac OS X.
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Sticking with the Mac would be annoying and difficult because of compatibility headaches, so you're switching to Linux?
Never. Apple will simply use a custom chipset in their hardware, and OS X will only run on that chipset. The chipset will be incompatible with Windows. Absolutely nothing will change with regards to compatibility between Macs and Wintels. Of course, something COULD change at any moment, and that's what's so beautiful about this plan. After Apple has successfully migrated the OS X developer community to MacIntels, it would be an easy step to open the floodgates and unleash OS X for ALL Wintel systems. My guess is that Apple isn't doing this until Microsoft is less of a threat (perhaps with a democratic administration in to pursue unfair business practices by Microsoft), but it's basically an "in case of unbridled euphoria, break glass" option.
"Talking nonsense is man's only privilege that distinguishes him from all other organisms." - Fyodor Dostoevsky -Chines
Now that Apple has announced that it is moved to Intel, who is going to buy a G5 now? I am sure as hell not. Apple just killed the sales of its hardware for the rest of the year. Also does this mean I will be able to buy a Dell PowerEdge 2850 running Mac OSX Server?
Strange women lying in ponds distributing swords is no basis for a system of government.
All of this would assume that they wanted the information kept secret. I have little doubt that if news.com was publishing this information, Apple didn't have that big of an interest in keeping it secret. With individual product releases, they are quite a bit more protective because they want to control how the products are treated in the media.
A good example of how this can work, if information came out on the shuffle well in advance of release, you'd see lots of reviews picking it apart for it's lack of a display, etc. So, before it ever hit the streets there would be a certain image of the device that could hurt their sales. But when Apple released it, they managed to spin the lack of display as a sort of feature. That the shuffle is about random playing, not picking songs out of a large library.
As far as this change goes, it doesn't really need to be handled in any particular way. They needed to keep it officially secret as a publicly traded company, but practically speaking I don't think they really cared. Ultimately the people most effected by it, ISV's, seem to have had some awareness ahead of time under NDA's (at least the bigger ones).
The end users of macs, for the most part, won't even understand what this means, or care. As long as the next mac they buy runs the software they have now and works as well as what they have now, they won't care.
This sig has been temporarily disconnected or is no longer in service
And this transition is different. There isn't a viable benefit to the customers.
No, this is bullshit. There's an extremely viable benefit to consumers: Apple will still be relevant in three years.
Why do you think Apple is doing this? It's not for shits and giggles. Those mobile G5s everyone's been waiting for, the one's that were going to save Apple's portable line from irrelevancy? It should be pretty obvious at this point that IBM has told Apple they aren't coming. Freescale dropped the ball, the G4 line is miles behind the times and Freescale lacks the ability to bring it up to date.
"Consumers don't benefit"? Bullshit. Consumers benefit because this is the only way Apple can keep their portables competitive. Laptops are the fastest growing segment of the market place, and Apple finally hitting 2Ghz with a G4 and its you've-got-to-be-shitting-me slow bus sometime next year wasn't going to cut it. Laptop sales fall, software makers lose interest, Apple fails, Apple's customers lose.
I'd rather they bet it all on a transition to keep the company relevant, rather than keep Freescale's incompetency and IBM's disinterest in laptop-suitable engineering as an anchor to hold them back in the market place until sheer inevitability kills the platform.
Just for the OS? I'm wondering.
Yes.
Peace be with you,
-jimbo
XML Tools for Mac OS X
Farewell?
(official "apple is dead" #94,549,238,192,204,223)
Apple has shown time and time again their resiliency to major hardware and software migrations. Once people get over the shock and awe of this announcement, people will start to realize it was a natural progression. We will be moving from a "niche" OS using a "niche" CPU to a "niche" OS using the "industry standard" CPU.
If next year, IBM sold off their PPC manufacturing, Apple would/could be dead in the water. Now that they are with Intel, they can just glide along with the industry.
And WINE/VirtualPC running so well may be the biggest disaster for MacOS -- why should Microsoft continue to support MSOffice/Mac when you can just run the Windows version in WINE? Why should Adobe build Acrobat for MacOS, when the Windows version (runs just as fast in WINE!) has more features and costs less??
Good Windows emulation is probably what killed OS/2, it can kill OS X too...
Think about it. We don't have a G5 Powerbook because we hear about the massive heat issues. Hell, just recently, I am having to take back my recently aquired G4 Powerbook because they are catching on bloody fire.
Secondly, I understand that Adobe is not making Photoshop and their other products for the Mac *first*. They are going to the PC, and then the Mac.
I mean, this quote says it all:
"I stood up here two years ago and promised you 3.0 GHz. I think a lot of you would like a G5 in your PowerBook, and we haven't been able to deliver that to you," said Jobs. "But as we look ahead, and though we've got great products now, and great PowerPC products still to come, we can envision great products we want to build, and we can't envision how to build them with the current PowerPC roadmap,"
So they go Intel. Who cares? Most of us are using Linux on x86, and we couldn't care less. The only thing that alarmed me was that they didn't choose AMD64, but thats just me. Hopefully, this will influence developers to port their stuff over to OS X now (which would benifit Linux indirectly imo). So hopefully we'll get a ton more games (yay!... games are a wasteland on the Mac) and apps because of this switch.
Things are abotu to get interesting now. Its like Jobs saying, "OK, Gates... lets fight in your ring."
++Om
The transition was so difficult for the audio and video industry, that for many people it STILL hasn't happened. You can find workhorse macs running OS9 in nearly every recording studio and post production house in LA.
Apple has obviously got an x86 gcc for Tiger and has already begun the process of porting the frameworks, most of which will probably not require massive porting effort. Frameworks like vecLib will probably require some more work to use SSE instead of Altivec though.
Even the concerns about things like endianness are not really a problem so long as the code was written the right way in the first place.
Dude, do you even know what "recompiling" means? Your entire post is nonsense.
/unix software on OS X is that GUI apps don't share quite the same API, which means they have to be run in an X server app, which is sort of kind of like wine, only 100% compatible and 100% ugly.
Linux apps are ALREADY recompilable and compatible for mac. All of them, just about. There were only problems when OS X beta first hit, and that was mostly because people had been writing their Makefiles poorly.
Modern computer software is almost never CPU-tied. The only problem is you have to recompile to run on different CPUs, which means you have to have source code. Linux apps, conveniently, you usually do, meaning transitioning between CPU archs as a linux user is effortless in a way it will not be for OS X users. The only problem with linux
This means no more second-class Mac versions of popular OS apps.
I assure you, no. The reasons inkscape is broken on my mac have nothing whatsoever to do with processors. I don't know what the holdup on openoffice 2.0 is, but I think it's less to do with chips and more to do with APIs. If there's some incompatibility between OO2 and Apple X11 I'm sure it would be fixed by now if someone felt like using a word processor inside the X11 battlemech were worth it.
What you're saying is kind of like "no more second-class windows versions of popular OS apps" because Cygwin exists there.
WINE will run on a Mac. This is *HUGE*. Imagine running any Windows software, at native speeds, with OpenGL support, on Mac OS X.
That does have interesting implications. But it's going to require a LOT of work to make that work, above and beyond what Wine's already doing. Wine will have to be practically rewritten for cocoa. Otherwise we'll be running the partially-incompatible wine translation layer inside the compatible-but-awkward X11 translation layer. Eww. I don't really expect wine for os x to get to the point your average person can run it for a long time, and I don't expect it to really work ever unless Apple themselves decide to put some work into it.
And Wine doesn't mean much to me personally. Again, great for Apple, great for switchers, not so much for anyone who's already invested in the mac. Windows apps are half the reason windows isn't worth using. The only thing it's really got worth keeping are games, and well, not only are those what Wine is worst at, that's what that little multicolored box plugged into my TV is for.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
Indeed, this will be a huge blow to Apple marketing. The PowerPC chips have always had some interesting feature, or excelled at some particular benchmark. Maybe they were faster, maybe they were slower. That's not the point. It was always possible to benchmark some obscure x86 worst case scenario to "prove" that they were selling the fastest computer in the world, ever.
Now, they will have negligible margins on Dell in the benchmarks. If they go a sane route and stay with OpenBoot or similar, they will still need video cards that don't depend on ugly PC BIOS, so they are still unlikely to be kings of 3D.
I understand the technical issues, but I would be surprised if IBM wasn't able to clean up their act with all the PPC chips they will be moving for embedded systems (game consoles as well as misc. other)
Color me worried.
You so need to sit down and take a chill-pill. Being x86 will not make it easier to make viruses. That whole aspect will depend on the OS, and it is still OSX. And your friends computer is suddenly not going to stop working. The transition is not happening for a wee while yet, and so Apple will still support his system. They're even going to allow the production of dual platform binaries. You're just getting worked up over nothing. I just think you and your friend are zealots - mac on x86 may be good. They might even have the rights to licence altivec over to intel processors. Just chill...
Hey, PowerPC was (and still is, really) a great platform concept.
But people don't buy computers for the concept. The x86 world beat out the PPC world when it comes to consumer chips by simply doing a better job of implementation. While IBM was promising 3 GHz performance that they couldn't deliver, Intel was cranking out a new chip which offers more performance per Watt on laptops than the "insanely great" G4.
x86 didn't look like it had a hell of a lot of potential three years ago, but AMD and Intel kept pounding. A good old "three yards and a cloud of dust" attack won the game.
Information wants to be anthropomorphized.
C|Net's article created fantastic media buzz for Apple. I'm betting that ten times as many people followed today's keynote address than otherwise would have. This allowed Steve to explain the transition in the best possible light, to a huge audience. And I do think he did a great job of putting a positive spin on this, with the CEO of Intel and the cofounder of Wolfram Research as eloquent guest speakers.
That that is is that that that that is not is not.
I freely and publically admit to being one of the people who said that no way would Apple be that stupid. I was wrong- Apple was that stupid.
When Apple does not die as a result of this, I trust that at some point you'll be as open and honest as you are now and admit that in retrospect it was you who were that stupid, and not Apple.
I know this because Tyler knows this.
Possibly not - the new version of XCode builds universal binaries for both Intel and PPC. So, what's the problem again?
The problem is when some "smart" developer decides to save space on his binary by simply not compiling in PowerPC support because "his userbase doesn't have that significant of a percentage of PowerPC users anymore". That's fine and dandy to the majority of x86 Mac users, but what about those left with a perfectly good aging PowerPC system?
They're suddenly unsupported and that's a horrible worthless feeling with nobody to blame it on except Apple for making, at worst, an arbitrary platform shift. At best, it's a failure of engineering which isn't terribly reassuring either.
I just wasted your mod points! HA!
Why? The Apple fans that buy Macs because they have OMG PPC970 will be chased away, sure. But not the ones that buy Macs because they are Macs. As long as it runs OSX and Photoshop, looks pretty sitting on their desk, and Steve Jobs said "Hey, you know, this is pretty good!" they are sold. The fact that they will most likely cost significantly less will be an added bonus for them, and likely attract even more customers than the switch chased away. People will likely not buy Dells, only to load them with OSX, because people generally use their computer the way it came until it dies. If someone wants OSX, they will buy an Apple, just like they do now.
And nothing has been said yet on if you WOULD be able to load it on any Dell or Gateway system. It could very well need some proprietary Mac hardware to run on. The CPU may be the core of the computer, but there are other things, too, such as the chipset and BIOS that could be Apple-exclusive.
I fail to see how this can have a SIGNIFICANT impact to Apple's install-base in the short term, and only see good things in the long term.
/usr/games/fortune
Regarding OO.org, theres plenty of architecture specific code in OO.org that had to be re-written for OS X. That's why the 1.0 port took so long. I'm not talking about NeoOffice/J, by the way, I'm talking about the X11 port. That's why the Mac X11 OO.org port alpha is 6 months overdue *so far*.While running under X11 is less than ideal, it'll still work nicely.
NeoOffice/J hasn't even started working on OO.org 2.0.
I understand the problems associated with an aqua port. Even without aqua, there are quite a few apps which make poor assumptions about the architecture they are running on, and quite a few libraries which use code that won't compile on a mac. I'm talking about just running stuff normalish linux apps on X11 on your Mac.
Not everything is a portable as you make it out to be. Plenty of programmers make poor assumptions when writing their software, including the sun guys who wrote the original star office codebase.
Oh, and Fullscreen opengl works great on the Mac's X11 implementation right now. I doubt that we'll see that go away on Mac OS X x86.
Why *shouldn't* wine work? We don't know the specifics of the OS yet, but Wine works on Freebsd. Transgaming believes that Cedega can be shoehorned onto Freebsd.
And cedega, if you haven't tried it, is fantastic for running Windows Games on Linux. Not 100%, mind you, but it handles a lot of games extremely well. In some cases, with better-than-windows performance.
Freebsd->Darwin isn't really that big of a jump, if you are talking about x86. Running Half-Life 2, even under X11, even under Cedega, could be quite a big selling point.
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Don't leave out Gamecube, they use PPC as well.
/* oops I accidentally made a comment, sorry */
I don't think so. I think they'll be pointing at IBM and saying, yeah, it was a really good platform up till now, but those guys in the suits dropped the ball on us, are too stupid to get the G5 right (a well-publicized problem), and Intel took the lead with the new Pentium portables. Fuck this -- we have always gone with the best chip out there, starting with the 6502, and we always will. Heck, with all of the Intel ads out there, your average consumer probably saw the PowerPC as more of a problem. Like, why aren't these guys using "the Centrino" like everybody else?
In fact, after a bit of quick footwork, this will be a beautiful position for Apple to be in. Look, they can say, this is what you can do with a Pentium -- if you have OS X. Look, kids, same hardware has your Windows box, but not one single virus, no crashes, no maleware...
Having Intel and Apple dovetail their marketing efforts -- scary, actually. But not bad.
This is a huge blow to PowerPC's credibility, though.
Not according to the NYT.
By contrast, the chips I.B.M. makes for Apple represent less than 2 percent of chip production at its largest factory in East Fishkill, N.Y. And while the microelectronics business as a whole is strategically important for I.B.M., it is a small part of the revenue of a company that increasingly focuses on services and software. A. M. Sacconaghi, an analyst for Sanford C. Bernstein & Company, estimates that the company's technology group - mostly microelectronics - will account for less than 3 percent of I.B.M.'s revenues and 2 percent of its pretax income this year.
For years, according to industry analysts, the work for Apple has been barely a break-even business for I.B.M. When the two companies were negotiating a new contract recently, Mr. Jobs pushed for price discounts that I.B.M. refused to offer. For I.B.M., "the economics just didn't work," said one industry executive who was briefed on the negotiations. "And Apple is not so important a customer that you would take the financial hit to hold onto the relationship."
I'm more interested in this quote:
However, [Apple Senior Vice President Phil] Schiller said the company does not plan to let people run Mac OS X on other computer makers' hardware. "We will not allow running Mac OS X on anything other than an Apple Mac."
Too bad. I'd like to run OS X w/out having to pay an Apple hardware premium.
FreeSpeech.org
It also about supply, IBM was bad enough making enough processors for Apple. AMD is a lot smaller than either Intel or IBM... They would be betting their future on a company on the ability for AMD to fill their demand. I don't think it was a risk they were willing to take.
nah, you'll buy a mac for the same price, and it'll look the same and work the same (only faster). I don't see how a switch from powerpc to x86 is an ideological shift or anything.
Apple has been slowly transitioning from proprietary hardware for a very long time. 20 years ago the system was all SCSI/68000/3.5" floppies (when PCs were IDE/x86/5.25"). That stuff cost too much money though (economics of scale), so they switched. The only thing left was the CPU, and its been killing them.
As long as the machines are still built by apple exclusively, this'll be more-or-less transparent to the mac user.
Jeremy
Your G4 processor was obsolete when you bought it.
It's not like your PPC is going to stop working next year. It's not like Apple is going to abandon PPC users. I'm sure that eventually, like the 68000 series, the PPCs will stop getting updates. I'm sure that date is a lot farther in the future than the usable lifetime of a G4 mini.
Personally, I'm still going get a G4 mini. I'm sure they will be faster, maybe cheaper in the future. Such is all technology.
It's funny how that happens... something could be said about evolution vs. revolution. It reminds me of all the times I've done a "clean rewrite" only to end up with something that just wasn't any better. Sometimes even worse.
A lot of people won't give up, though: in the face of enormous evidence they'll still assure themselves that because something is "new" and "clean" it's somehow better.
So if x86, with all it's hacks and kludges, is still faster and more efficient than these so called "clean" designs, what the heck is the point of having a clean design?
Cheers.
Chances are the Intel macs will have Open Firmware instead of a PC-style BIOS. There's a lot more to the PC architecture than the processor, and the OS legitimately relies on all of it. You'll be able to use the CPU from your commodity PC hardware, but you won't get it to read your hard drive unless you have a motherboard with Apple's firmware, and it won't run at all unless there's physical RAM where the BIOS is on a PC.
Sure we have lived with in-order cores before (out-of-order was introduced to the PC with the Pentium Pro), but it is troublesome, we are back to the compiler having to do the heavy lifting trying to put together the ideal instruction stream. It is actually a lot worse than it seems when I compare it with the original pentium, with its shallow pipeline and the relativly speaking lower memory latency of those days you could get away with a lot more without trashing performance.
Even if Apple through some magic manages to generate decent code for the in-order primary core (and it is not unlikely that they'd have to dump GCC since lots of hard-to-merge work would have to be done, and then they would lose the advantage of having a freely redistributable compiler) they will still be stumped on the vector units. Sure some of the heavier apps manage to make good use of Altivec, but that is a lot easier than trying to keep 8-16 vector units filled at the same time. Basicly only scientific and various extremely expensive pro applications would ever manage to invest the effort needed to actually manage to tap much of the power of the vector units (part because vectorization is hard, but also importantly because there are so many units to fill).
This all adds up to the Cell (and IBM's new in-order cores without the vector units) being quite unsuitable for any market where the applications are not written very specifically targeting the chip. It works for consoles since development is hardware-specific there, but putting out a computer with the Cell and expecting it to work out on peoples desktops is not in any way a good idea.
I don't think dual booting will happen. Too much work, and makes Mac users reboot (and we love doing that!)
I think the more likely scenario is a version of Virtual PC that doesn't suck. Runs the windows code semi-natively...
I love my Mac for the usability of its user interface (both CLI and GUI) and for the fact that it looks so damn good. It depresses me when I have to fire up my ugly old PC when I actually want my code to finish in a reasonable time.
Doesn't it make you feel good to know that our freedoms are protected by politicans, lawyers and journalists.
Let's ask someone who understands deeply the full and total differences between AltiVec and SSE2.
Like, me. I wrote the AltiVec emulation in PearPC. Thus, I have quite a bit of authority on the differences between the two.
AltiVec has a more fleshed out assortment of instructions. SSE2, and SSE both are missing a number of instructions. Most of these don't get used often, so you're not losing much in the way of speed, but AltiVec has a more complete implementation.
EXAMPLE:
PAVGB
PAVGH
but no PAVGW
PMINUB and PMINSW, but no PMINSB, PMINUH, PMINSH, PMAXUW
PSLLW and PSLLD, but no PSLLH, or PSSLB (same for all packed shifts)
Then, I'll point out a number of points upon the design straight from the Pentium 4 optimization guide.
Don't use SSE when 64-bits is all you're working on. This makes obvious sense for floating point code (denormals take a long time to calculate and can stall results for the stuff you want), but this is saying use MMX when only using 64-bits of data. Because, and I kid you not. They say that the 128-bit SSE is wider, and thus performs slower. (Why should it when it's PARALLEL execution.)
Also, SSE3 is breaking parallel operations by providing horizontal instructions. Why even vectorize these, they're going to run as slow as scalar operations. Ok, so you get out of passing it back out to memory, but come on, the idea of a vectorization unit is to perform parallel vector math. But I understand the strong desire to make things work fast rather than proper, and avoiding those few clock-cycles means that they're willing to stall a vector unit on a scalar operation.
Um... what do we have left. AH yes. The problem of XORPS vs PXOR. They both do the same thing right? They XOR the value of one 128-bit register against another 128-bit register. But there's a fundamental point here. If you use XORPS on an XMM register, which is integer, then you're going to get slow down. If you use PXOR on an XMM register, which is floating point, then you're going to get slow down. Now this really isn't a problem when you can track this information and such. But really. Shouldn't these both be equated to the same microcode, and handled by say, a logic vector unit that handles permutes (sorry, shuffles) and logic? WOULDN'T THAT MAKE SENSE. Not apparently to the SSE designers.
Now, SSE2 yes had double-percision floating point in 128-bit vector registers, which gets you a whole incredible 2 elements per vector. Wow, that's definitely worth the overhead of using vector registers, and insuring alignment, etc. Plus, the G5 can issue two identical FPU instructions at one time, and since all PowerPC math is done in double-precision (or better internally to an instruction) you get two double-precision operations per cycle. Wow, I can see a true benefit for hacking in double precision support into AltiVec.
Now, if you want to debate any of these points, I'll gladly point you to the proper resource to prove my point, as I use them constantly in my work on emulating AltiVec with SSE.
(BTW: emulating SSE with AltiVec would be almost painfully simple compared to AltiVec in SSE. It's almost entirely a proper superset of SSE.)
Oh, last, let's not forget about those wonderful instructions that Apple must have told someone to put in there, because they're used for Anti-aliasing fonts, and icons, and are just used all over the place in OSX: vmhraddshs, etc. Which will likely never have a single instruction equivalent in SSE.
I am unamerican, and proud of it!