So how long will it take for Intel chips running Mac OS X to catch up to TODAY'S PowerPC G5 chips running Mac OS X?
May the computing Gods help us all. \
I will miss Altivec, but those benchmarks are irrelevant : for one Benchmark demonstrating that the G5 is faster, there is one who demonstrate the contrary.
You will note also that the future of Intel chips are not the P4 Chips, but the pentium 3 derived ones. It's also possible, that Intel will improve his SIMD unit with the help of Apple. Who knows ?
This feeling of betrayal is simply childish! This is not a religion, This is not a philosophy. This is a company making hardware and software. They do what they think is best for them, and by extention, us. The whole reason for them doing this was that they know that their customers, and their potential customers are not going to be happy down the line if IBM can't or is unwilling to do what they need.
I'm reading here that we lose the Cell, the Power 5 and 6 etc. That's nonsence. We don't have the Cell, the Power 5 and 6 etc.!
Moving to the Cell would be more disruptive than moving to x86. Apple was ready for that for years. The Cell is so different that it's being doubted that it would be effective as a general purpose cpu. Maybe later as a daughter card for rendering, yes. But this move does nothing to stop that should Apple decide to go thar way later.
As for the Power chips, these are VERY expensive chips, as are the IA-64 chips from Intel. Try $3,000 and up for the top of the line units. Slower ones are cheaper, but start at $1,000 or so.
What gain there?
Perhaps Apple might contemplate keeping XServes with the PPC if customers demand it, but there would have to be significant demand.
I understand this is probably a blanket reply, but you quoted me.
I never mentioned any feeling of betrayal. I don't think MOST people feel betrayed. But just because this move may show probable long term benefits, you've got to weather the short term hurt that IS coming to get there. And people have every right to be concerned about that.
I will miss Altivec, but those benchmarks are irrelevant : for one Benchmark demonstrating that the G5 is faster, there is one who demonstrate the contrary.
You will note also that the future of Intel chips are not the P4 Chips, but the pentium 3 derived ones. It's also possible, that Intel will improve his SIMD unit with the help of Apple. Who knows ?
I was going to mention the SIMD at some point.
What we have to understand here is that Intel is not the dog uninventive company that some think it is. As Ottelini (spelling?) shower, for those who might not have known, Intel inventor the DRAM, and the microprocessor itself.
They were one of the originators of Flash memory and are the biggest supplier of that as well.
Over the years Intel made several contributions to cpu design only to find MS in opposition to those innovations because they wanted to do it in software. Their software.
Intel came out with several SIMD extensions that were never used, and were therefore dropped. The ones in place now are the ones that MS supports, or work without software support.
If Intel was willing to take the risks over the years of coming out with innovations that they didn't know MS would support, they might be interested in doing that again. But this time with the support of Apple.
If this happens, and it's a greater possibility than some of what I've been reading here and elsewhere, then MS would feel compelled to support them as well, or lose performance to an x86 Mac OS.
Intel might be very happy at this turn of events because it will allow the to regain control of all of their chip development.
Who knows, this could be one of the reasons they've been wooing Apple for so long.
Moving to the Cell would be more disruptive than moving to x86. Apple was ready for that for years.
That's not true according to what I've read. The Cell's core processor is a PowerPC processor. Presumably that would make the "transition" non-existant.
Quote:
The Cell is so different that it's being doubted that it would be effective as a general purpose cpu. Maybe later as a daughter card for rendering, yes.
I think those "doubts" are probably more of a combination of a few possible actual concerns and a large pile of FUD, no doubt spread by the PR machines of Intel and AMD. There is a nice article about Cell at http://www.igeek.com/CellProcessor.pdf that addresses some of these concerns.
Quote:
But this move does nothing to stop that should Apple decide to go thar way later.
Agreed, which is why I'm no longer concerned about it. If the Cell materializes as a usable desktop processor Apple will have no problems utilizing it in future Macs, thanks to universal binaries.
I understand this is probably a blanket reply, but you quoted me.
I never mentioned any feeling of betrayal. I don't think MOST people feel betrayed. But just because this move may show probable long term benefits, you've got to weather the short term hurt that IS coming to get there. And people have every right to be concerned about that.
I'm sorry, but a major defect of these forums is not including previous posts in the quote. I'm on a number of other forums and they include past quotes. That makes the replies more intelligible.
Others have mentioned it though.
Sometimes when I'm writing I'll remember to copy and include them if necessary, sometimes not.
I didn't mean to change the direction of your post. I did emphasize the concept of going overboard in this. sometimes a little bit of negativity to those who understand much less than we do can hurt much more than we realize. You know the phase "A little knowledge is a dangerous thing."
My stock is a little off today. So far I'm holding because I've got a long way to go before it goes down enough to cause a loss (a LONG way). But I'd hate to think people are selling for unfounded reasons.
The same when thinking about buying an Apple product.
That's not true according to what I've read. The Cell's core processor is a PowerPC processor. Presumably that would make the "transition" non-existant.
I think those "doubts" are probably more of a combination of a few possible actual concerns and a large pile of FUD, no doubt spread by the PR machines of Intel and AMD. There is a nice article about Cell at http://www.igeek.com/CellProcessor.pdf that addresses some of these concerns.
Agreed, which is why I'm no longer concerned about it. If the Cell materializes as a usable desktop processor Apple will have no problems utilizing it in future Macs, thanks to universal binaries.
I don't know what you've been reading. but do you know anything about the way the PPC as embodied in Apple's machines works? do you know anything about the way ANY general purpose cpu today works?
Can you explain to me why the PPC in the Cell is easily adapted to Apple's OS and the way the g4 and g5 process instructions?
This has nothing to do with FUD. That's the excuse that people who don't understand something use to hold their beliefs tightly.
Read the articles on the Cell in Arstechnica.com. Read the articles on why this would be difficult for Apple to use.
I think Apple is a little tired of empty promises and expectations. It's time to just move on and make great products ( which have nothing to do with the type of processor)
Until yesterday we speculated if and how Intel will enter the game. Now we speculate if and how IBM will enter again.
That's the deal. Jobs declared the PowerPC dead. No backdoor open. But now who will trust him again?
I wouldn't sweat the trust thing. In 3 years, there will be only 2 corporations in the world that will be able to produce 45 nm desktop chips. One of them will be Intel. The other one, I don't know, perhaps UMD+TSMC or Japanese/Korean conglomerate or IBM/AMD/Philips conglomerate. I'll bet that Intel will get their first, a year before the other companies will.
IBM will be perfectly happy producing a 65 nm 3.2 GHz Cell and Xenon well into 2008, barely bothering to improve their performance.
Hard to believe, but it appears the x86 ISA will live on forever until quantum computing, molecular computing, or some or fancy thing that can change the landscape comes into being.
I wouldn't sweat the trust thing. In 3 years, there will be only 2 corporations in the world that will be able to produce 45 nm desktop chips. One of them will be Intel. The other one, I don't know, perhaps UMD+TSMC or Japanese/Korean conglomerate or IBM/AMD/Philips conglomerate. I'll bet that Intel will get their first, a year before the other companies will.
IBM will be perfectly happy producing a 65 nm 3.2 GHz Cell and Xenon well into 2008, barely bothering to improve their performance.
Hard to believe, but it appears the x86 ISA will live on forever until quantum computing, molecular computing, or some or fancy thing that can change the landscape comes into being.
I don't know what you've been reading. but do you know anything about the way the PPC as embodied in Apple's machines works?
Yes. And if the Cell has a core processor that understands the PowerPC ISA there is no reason why any current Mac software would not run on it without a recompile. I'm not talking about internal operating system or hardware changes. That kind of transition is solely the problem of Apple's engineers and is not the topic if this debate. The transition Steve Jobs talked about was the developer transition. If developers have to do no work, then there is no transition, regardless of what people at Apple HQ might have to do. A quote from what I've been reading, which you could have figured out yourself given that I posted the link:
Quote:
From the article: The Cell Processor uses a PowerPC as the core processor, making it compatible with many of the tools and users that are familiar with that chip.
Quote:
do you know anything about the way ANY general purpose cpu today works?
Yes, I do. What the hell is your point other than to be inflammatory?
Quote:
Can you explain to me why the PPC in the Cell is easily adapted to Apple's OS and the way the g4 and g5 process instructions?
I'm not talking about adaptability to Apple's OS, I'm talking about application-level compatibility. If current PowerPC applications run on a Cell based Mac, there is no transition. And this has nothing to do with any of the technological challenges of integrating a Cell processor into Apple's hardware designs. I was under the impression that the entire issue of transition had to do with developer/user transition, not with internal Apple engineering transitions.
Quote:
This has nothing to do with FUD. That's the excuse that people who don't understand something use to hold their beliefs tightly.
I'll simply dodge your inflammatory bait. I'll go back to what I initially responded to, which was your remark "Moving to the Cell would be more disruptive than moving to x86." Maybe you meant something else, but by "disruptive" I assumed you were addressing the disruption to developers and users. Frankly I don't care how disruptive something is to Apple hardware engineers. If they got it to work well enough to ship a Mac based on it then its "disruptiveness" would be marginal at best compared to the current x86 transition, again given that it would be driven by a PowerPC core. There would be no recompiling and there would be no emulation. Applications would just work.
Quote:
Read the articles on the Cell in Arstechnica.com. Read the articles on why this would be difficult for Apple to use.
Then come back here, and we'll talk.
The arrogance is remarkable. I'm aware of the problems Cell currently has. The possibility of moving to Cell would presuppose that Cell were ready for desktop use. If it were, that again would presuppose that the problems with memory controllers and the likes would have been solved, otherwise Apple wouldn't make that move in the first place. And it hasn't, which is why I never advocated that Apple move to Cell unless/until the technology is ready for prime-time. If it had been by now, then a move to Cell would _not_ be more disruptive, which is the entire point I was making. Your flaming was not necessary.
OK, I've had time to calm down a bit. This was quite a shock to my system. Who would have actually thought Apple would go to Intel(nothing against Intel by the way, I think they must have top engineers to have continually pushed the envelop of x86).
The only recurring thought I have is that something either horribly went wrong at IBM, either engineering or politics, or Intel truly showed Jobs something he couldn't turn down.
We'll know in the weeks/months ahead if engineering @ IBM went horribly wrong if there are upgrades to the Powermacs. If, as the chud tools indicate, a dual core 970 mp appears then Intel must truly have something great in the works. I believe IBM's engineering is more than capable, so I discount engineering, but can't rule out internal IBM politics. But, I've never been much a conspiracy buff and IBM has been touting PPC Everywhere. Just doesn't add up.
This is the biggest gamble Apple is making "EVER" and Steve Jobs knows it.
This is a toe to toe, nose to nose knock down battle between Apple and Microsoft. Apple won't be able to "niche player it" this time. The hardware will pretty much be the same, the OS's and software will decide. Apple has a huge task at hand. This is what makes me think Intel has something they showed Jobs that tipped him over the fence. It also makes me think that Apple may actually get some modifications on the chip, eventually, not available to Windows. I'm woefully ignorant of cpu architecture, but I just get a sense that Intel sees something in Mac OS X that will make it easier to make minor architectural changes to enhance speed and possibly reduce costs.
I give Jobs credit, this decision has taken a huge set of cahonies.
This is a toe to toe, nose to nose knock down battle between Apple and Microsoft. Apple won't be able to "niche player it" this time. The hardware will pretty much be the same, the OS's and software will decide.
Why would you think that way? Mac OS X will not run on a vanilla PC you know, no way in hell! What is the difference if they both have Intel inside or not. Why does this all of a sudden put Apple "toe-to-toe" with Microsoft. On the contrary, I think this will open up things like Office and Outlook so they can finally release a damn email client that fully supports Exchange.
Yes. And if the Cell has a core processor that understands the PowerPC ISA there is no reason why any current Mac software would not run on it without a recompile. I'm not talking about internal operating system or hardware changes. That kind of transition is solely the problem of Apple's engineers and is not the topic if this debate. The transition Steve Jobs talked about was the developer transition. If developers have to do no work, then there is no transition, regardless of what people at Apple HQ might have to do. A quote from what I've been reading, which you could have figured out yourself given that I posted the link:
Yes, I do. What the hell is your point other than to be inflammatory?
I'm not talking about adaptability to Apple's OS, I'm talking about application-level compatibility. If current PowerPC applications run on a Cell based Mac, there is no transition. And this has nothing to do with any of the technological challenges of integrating a Cell processor into Apple's hardware designs. I was under the impression that the entire issue of transition had to do with developer/user transition, not with internal Apple engineering transitions.
I'll simply dodge your inflammatory bait. I'll go back to what I initially responded to, which was your remark "Moving to the Cell would be more disruptive than moving to x86." Maybe you meant something else, but by "disruptive" I assumed you were addressing the disruption to developers and users. Frankly I don't care how disruptive something is to Apple hardware engineers. If they got it to work well enough to ship a Mac based on it then its "disruptiveness" would be marginal at best compared to the current x86 transition, again given that it would be driven by a PowerPC core. There would be no recompiling and there would be no emulation. Applications would just work.
The arrogance is remarkable. I'm aware of the problems Cell currently has. The possibility of moving to Cell would presuppose that Cell were ready for desktop use. If it were, that again would presuppose that the problems with memory controllers and the likes would have been solved, otherwise Apple wouldn't make that move in the first place. And it hasn't, which is why I never advocated that Apple move to Cell unless/until the technology is ready for prime-time. If it had been by now, then a move to Cell would _not_ be more disruptive, which is the entire point I was making. Your flaming was not necessary.
I'm sorry, but what you posted shows nothing. Application level? And what is that supposed to mean? So you are saying that the apps run but the OS doesn't?
If you did know, you would have said that the PPC in the Cell is a non-reentry model. It has two threads, and no reusable instructions.
It's considered to be extremely difficult to get the Mac OS functioning on it. That would include any programs that run as well of course.
I don't know anyone who is impressed with the Geek article.
COULD Apple get it to work? Possibly. To what effect?
A primitive PPC core, much simpler than any Apple has ever had, designed to be a controller for the spe's (though it does have a similarly primitive VMX unit), though it can do the cpu work required for the games designed just for it, plus the spe's themselves which are basically small vector computers.
None of this in any way resembles the G4 or the G5.
You think that with all of the work that Apple has put into x86 over the years, this would be a viable alternative?
No recompile and hand tweaking would get this to work for Apple. In later years, IF IBM were to rework the chip for Apple, possibly. But we all know that they wouldn't go to all that work.
I'm just tired of people looking at the (theoretically) high number of flops this is supposed to do, and cry out- "This is it!"
I have had a chance to calm down and I am now excited about the news. If the DEVELOPERS don't get pissy (big if in my opinion) and support the PPC for 3 years past 2007, all is good. Otherwise, Steve just pissed a lot of people off.
Steve's reputation will be made upon the Developers for the Mac platform. I guess most people are worried, including myself, that our machines we have now will run the latest software in 2008 & 9. That is what it comes down to, right? As long as the developers are in for supporting PPC a few years past 2007, I'm in.
I am ready for a 12" PowerBook with dual 2.0 Ghz processors that will last 4 - 5 hours on one charge and render video like my G5 iMac has never seen.
You haven't "shown" anything either. Quit this derogatory nonsense.
Quote:
Application level? And what is that supposed to mean? So you are saying that the apps run but the OS doesn't?
Given that the Cell requires quite the underlying architectural change in terms of hardware design as well as operating system magic to deal with parallelism, threading, etc, then yes, the OS would likely require significant work. However, you continue to fail to present any reason why a run-of-the-mill Mac OS X application could not run without a recompile if a standard PowerPC instruction set were available, as early Cell specs seem to indicate. If you wan't to inform me exactly why that wouldn't work, please do, otherwise cut out the baseless posturing.
Quote:
If you did know, you would have said that the PPC in the Cell is a non-reentry model. It has two threads, and no reusable instructions.
This is entirely meaningless. What does this have to do with running PPC binaries without recompiles? Sony would not be talking about building workstations based on Cell if it weren't a viable desktop architecture.
Neither microarchitectural details nor the performance characteristics of the POWERPC Processing Element were disclosed by IBM during ISSCC 2005. However, what is known is that the PPE processor core is a new core that is fully compliant with the POWERPC instruction set, the VMX instruction set extension inclusive. Additionally, the PPE core is described as a two issue, in-order, 64 bit processor that supports 2 way SMT. The L1 cache sizes of the PPE is reported to be 32KB each, and the unified L2 cache is 512 KB in size. Furthermore, the lineage of the PPE can be traced to a research project commissioned by IBM to examine high speed processor design with aggressive circuit implementations. The results of this research project were published by IBM first in the Journal of Solid State Circuits (JSSC) in 1998, then again in ISSCC 2000.
Quote:
It's considered to be extremely difficult to get the Mac OS functioning on it.
Which would be Apple's internal problem and not a transition issue.
Quote:
That would include any programs that run as well of course.
No, it wouldn't. That would only include programs that are specifically trying to take advantage of parallelism/threading. There's no reason why single-threaded or even traditionally multi-threaded applications couldn't run on the PPE.
Quote:
I don't know anyone who is impressed with the Geek article.
And what is that supposed to mean? I should discard the article because some anonymous guy on the AI forum said he doesn't "know anyone who is impressed" with it?
Quote:
COULD Apple get it to work? Possibly. To what effect?
A primitive PPC core, much simpler than any Apple has ever had, designed to be a controller for the spe's (though it does have a similarly primitive VMX unit), though it can do the cpu work required for the games designed just for it, plus the spe's themselves which are basically small vector computers.
None of this in any way resembles the G4 or the G5.
A traditional G5 processor could serve as the core as well.
Quote:
You think that with all of the work that Apple has put into x86 over the years, this would be a viable alternative?
Who said anything to that effect? How do you know what Apple's research efforts in other directions have been? Moreover, my entire comment was predicated on the assumption that if we were to be actually in the process of a Cell transition, these issues would have been resolved somehow. In which case, no, a Cell transition would not be "more disruptive." Or do you really think Apple would float some half-baked, slow alpha-level prototype crap into the market?
Quote:
No recompile and hand tweaking would get this to work for Apple. In later years, IF IBM were to rework the chip for Apple, possibly. But we all know that they wouldn't go to all that work.
You continue to fail to mention why current code wouldn't be able to run on this technology, if it were properly factored for desktop use.
Quote:
I'm just tired of people looking at the (theoretically) high number of flops this is supposed to do, and cry out- "This is it!"
It's not.
So that's what this entire rant and arrogant posturing has been about? About something I never even said? All I've said is this has potential, and that if the technology were currently in a state where it were actually a viable option for Macs that it would not be "more disruptive" than an x86 move. And I continue to maintain that. Adding VMX to the G4 didn't stop software from working, utilizing the GPU for image/video processing didn't stop software from working, and there's no reason to believe that a Cell processor with a PPC core couldn't run current software. Nothing would force a developer to refactor everything into cells, in fact that wouldn't even make sense for all software.
I have had a chance to calm down and I am now excited about the news. If the DEVELOPERS don't get pissy (big if in my opinion) and support the PPC for 3 years past 2007, all is good. Otherwise, Steve just pissed a lot of people off.
Steve's reputation will be made upon the Developers for the Mac platform. I guess most people are worried, including myself, that our machines we have now will run the latest software in 2008 & 9. That is what it comes down to, right? As long as the developers are in for supporting PPC a few years past 2007, I'm in.
I am ready for a 12" PowerBook with dual 2.0 Ghz processors that will last 4 - 5 hours on one charge and render video like my G5 iMac has never seen.
So, what has happened with IBM?... "INTEL inside" ??? Nooooooo!!!!
No more Altivec? No 64 bit? Inferior hardware? FAT? God knows what else...
Sorry Steve, but I believe that the trust has been broken for a lot of Mac users. We can now look forward to tons of bloatware and maybe even some tasty viruses too. Maybe I'm being paranoid but this decision just does not feel or sound right at all. Sounds like Apple have spat the dummy on this one.
I fear IBM making a major breakthrough and trouncing Intel in the future with some super PowerPC technology, then Apple will really be screwed. Dell anyone?
So, what has happened with IBM?... "INTEL inside" ??? Nooooooo!!!!
No more Altivec? No 64 bit? Inferior hardware? FAT? God knows what else...
Sorry Steve, but I believe that the trust has been broken for a lot of Mac users. We can now look forward to tons of bloatware and maybe even some tasty viruses too. Maybe I'm being paranoid but this decision just does not feel or sound right at all. Sounds like Apple have spat the dummy on this one.
I fear IBM making a major breakthrough and trouncing Intel in the future with some super PowerPC technology, then Apple will really be screwed. Dell anyone?
I, too, have my mac for the OS, not the processor.
On the other hand, I have my Apple stock because of the whole ball of wax. And a large chunk of that ball (harware sales) are totally unpredictable right now. I don't think anyone expects a sudden surge in sales, and most don't reasonably expect sales to remain at their current levels. Most likely, we're looking at pretty weak hardware sales for the foreseeable future.
Unless of course they slash the prices, but in that case increased sales doesn't mean shit.
So yeah, many people who are nervous don't necessarily pin those feeling on the fact that OS X is going to be on Intel. It's about the state of their stock until that transition is well underway...
2 thoughts this morning (it's 7am now where i am, i have to go teach a Flash class in a few hours zzzzz still waking up
1.
apple stock
i commend those that sold their stock, and also those that are hanging on. my guess (i've never owned any AAPL) is that i wouldn't get in now, if i had stock i'd bail at $35, and if i sold all my APPL i'd get back in at $25 or $30 if i wanna be in for a 1-2 year long run ~ this requires a level of faith though that many still do have....
2.
sales
yes, sales, revenues, profit, are generally expected to take a hit in the medium term. short term (within 1 month), most people outside of geekLand won't know what's going on, many are still pestering their parents/spouses/bosses/whatever to get a new iPod. medium term (2months - 1 year) is where many see sales suffering.
okay, 3 thoughts.
2.(b)
steve job's strategy to 'weather teh storm'
i am starting to believe that stevie J has some very sweet powerPC stuff up his sleeve for the next 12 months. i for one am not going to buy anything apple until the dust settles. but for the july-sep 2005 quarter, christmas quarter, then macworld sf2006, you can be sure he's going to bring out some big guns to hold revenues and profits at a reasonable level. if there's one thing about companies, even if they have a very strong and visionary long term view (as in this switch to intel), they are very very very reluctant to let their quarter-over-quarter profits slide in the hopes of 'future promises' of 'things getting better'. there'll be some sweet deals in the works for the rest of 2005 because
2.(b)(i)
any powerPC you buy now runs mac os X sweet
(ii)
said powerPC is robust, works, no gamble on intel
(iii)
large existing library of software that just works (well, at least on Panther)
2.(c) (i'm on a roll here )
my guesstimate is that apple execs will choose to show the market strong sales and revenues if at the cost of profit margins. they could even pull off something like offering 1GB standard in teh Powerbooks towards the christmas season. profits might take a slight hit because they can't have their traditionally killer margins, but their priority is
2.(c)(i)
showing continued growth in iPod, iTMS and Mac sales
(ii)
showing clear progress on Intel at macworld SF 2006
2.(d) (oh man i'm never going to shut up am i)
unfortunately for us wanting to buy a new mac now, you can forgettabout it for a few weeks unless you find a really good deal or you really need one (i'm going to hog my dad's iBook g4 as much as i can). however, there are two no-brainer sales drivers in the rest of the year where we can shoot to lock in some nice powerPC-mac deals
2.(d)(i)
back-to-school, july-sep quarter, paris: new iBooks, mac Mini, etc etc
standard warranty may be extended to 2 years, or some sort of much cheaper 'Applecare lite', or 'Applecare Xtra' which is a form of insurance covering spills, drops, theft (dell is doing it in my country for CHEAP right now!!) because people are going to start to get turned of 3 year applecare, because, that 3rd year is going to be 2007-2008 where one can easily be 'worried' (albeit with not much cause) that no one is going to bother to support your old powerPC Mac. powerPC in 2008 may seriously start to go out of fashion. 3 years is an eternity in the consumer tech world.
2.(f) shit i've gotta go get ready. peace. yes good things, but we need a mourning space. maybe i'll set up something on blogger.com R.I.P. PowerPC, we had a good run. like a phoenix it may rise again, but for now, let us quietly reflect on its ashes.
Comments
Originally posted by e1618978
I was being sarcastic. You seem to have not noticed.
Most definitely.
Originally posted by CosmoNut
From Apple's G5 performance page:
So how long will it take for Intel chips running Mac OS X to catch up to TODAY'S PowerPC G5 chips running Mac OS X?
May the computing Gods help us all.
I will miss Altivec, but those benchmarks are irrelevant : for one Benchmark demonstrating that the G5 is faster, there is one who demonstrate the contrary.
You will note also that the future of Intel chips are not the P4 Chips, but the pentium 3 derived ones. It's also possible, that Intel will improve his SIMD unit with the help of Apple. Who knows ?
Originally posted by melgross
This feeling of betrayal is simply childish! This is not a religion, This is not a philosophy. This is a company making hardware and software. They do what they think is best for them, and by extention, us. The whole reason for them doing this was that they know that their customers, and their potential customers are not going to be happy down the line if IBM can't or is unwilling to do what they need.
I'm reading here that we lose the Cell, the Power 5 and 6 etc. That's nonsence. We don't have the Cell, the Power 5 and 6 etc.!
Moving to the Cell would be more disruptive than moving to x86. Apple was ready for that for years. The Cell is so different that it's being doubted that it would be effective as a general purpose cpu. Maybe later as a daughter card for rendering, yes. But this move does nothing to stop that should Apple decide to go thar way later.
As for the Power chips, these are VERY expensive chips, as are the IA-64 chips from Intel. Try $3,000 and up for the top of the line units. Slower ones are cheaper, but start at $1,000 or so.
What gain there?
Perhaps Apple might contemplate keeping XServes with the PPC if customers demand it, but there would have to be significant demand.
I understand this is probably a blanket reply, but you quoted me.
I never mentioned any feeling of betrayal. I don't think MOST people feel betrayed. But just because this move may show probable long term benefits, you've got to weather the short term hurt that IS coming to get there. And people have every right to be concerned about that.
Originally posted by Powerdoc
I will miss Altivec, but those benchmarks are irrelevant : for one Benchmark demonstrating that the G5 is faster, there is one who demonstrate the contrary.
You will note also that the future of Intel chips are not the P4 Chips, but the pentium 3 derived ones. It's also possible, that Intel will improve his SIMD unit with the help of Apple. Who knows ?
I was going to mention the SIMD at some point.
What we have to understand here is that Intel is not the dog uninventive company that some think it is. As Ottelini (spelling?) shower, for those who might not have known, Intel inventor the DRAM, and the microprocessor itself.
They were one of the originators of Flash memory and are the biggest supplier of that as well.
Over the years Intel made several contributions to cpu design only to find MS in opposition to those innovations because they wanted to do it in software. Their software.
Intel came out with several SIMD extensions that were never used, and were therefore dropped. The ones in place now are the ones that MS supports, or work without software support.
If Intel was willing to take the risks over the years of coming out with innovations that they didn't know MS would support, they might be interested in doing that again. But this time with the support of Apple.
If this happens, and it's a greater possibility than some of what I've been reading here and elsewhere, then MS would feel compelled to support them as well, or lose performance to an x86 Mac OS.
Intel might be very happy at this turn of events because it will allow the to regain control of all of their chip development.
Who knows, this could be one of the reasons they've been wooing Apple for so long.
Originally posted by melgross
Moving to the Cell would be more disruptive than moving to x86. Apple was ready for that for years.
That's not true according to what I've read. The Cell's core processor is a PowerPC processor. Presumably that would make the "transition" non-existant.
The Cell is so different that it's being doubted that it would be effective as a general purpose cpu. Maybe later as a daughter card for rendering, yes.
I think those "doubts" are probably more of a combination of a few possible actual concerns and a large pile of FUD, no doubt spread by the PR machines of Intel and AMD. There is a nice article about Cell at http://www.igeek.com/CellProcessor.pdf that addresses some of these concerns.
But this move does nothing to stop that should Apple decide to go thar way later.
Agreed, which is why I'm no longer concerned about it. If the Cell materializes as a usable desktop processor Apple will have no problems utilizing it in future Macs, thanks to universal binaries.
Originally posted by rageous
I understand this is probably a blanket reply, but you quoted me.
I never mentioned any feeling of betrayal. I don't think MOST people feel betrayed. But just because this move may show probable long term benefits, you've got to weather the short term hurt that IS coming to get there. And people have every right to be concerned about that.
I'm sorry, but a major defect of these forums is not including previous posts in the quote. I'm on a number of other forums and they include past quotes. That makes the replies more intelligible.
Others have mentioned it though.
Sometimes when I'm writing I'll remember to copy and include them if necessary, sometimes not.
I didn't mean to change the direction of your post. I did emphasize the concept of going overboard in this. sometimes a little bit of negativity to those who understand much less than we do can hurt much more than we realize. You know the phase "A little knowledge is a dangerous thing."
My stock is a little off today. So far I'm holding because I've got a long way to go before it goes down enough to cause a loss (a LONG way
The same when thinking about buying an Apple product.
Originally posted by AppleRISC
That's not true according to what I've read. The Cell's core processor is a PowerPC processor. Presumably that would make the "transition" non-existant.
I think those "doubts" are probably more of a combination of a few possible actual concerns and a large pile of FUD, no doubt spread by the PR machines of Intel and AMD. There is a nice article about Cell at http://www.igeek.com/CellProcessor.pdf that addresses some of these concerns.
Agreed, which is why I'm no longer concerned about it. If the Cell materializes as a usable desktop processor Apple will have no problems utilizing it in future Macs, thanks to universal binaries.
I don't know what you've been reading. but do you know anything about the way the PPC as embodied in Apple's machines works? do you know anything about the way ANY general purpose cpu today works?
Can you explain to me why the PPC in the Cell is easily adapted to Apple's OS and the way the g4 and g5 process instructions?
This has nothing to do with FUD. That's the excuse that people who don't understand something use to hold their beliefs tightly.
Read the articles on the Cell in Arstechnica.com. Read the articles on why this would be difficult for Apple to use.
Then come back here, and we'll talk.
Originally posted by PB
Until yesterday we speculated if and how Intel will enter the game. Now we speculate if and how IBM will enter again.
That's the deal. Jobs declared the PowerPC dead. No backdoor open. But now who will trust him again?
I wouldn't sweat the trust thing. In 3 years, there will be only 2 corporations in the world that will be able to produce 45 nm desktop chips. One of them will be Intel. The other one, I don't know, perhaps UMD+TSMC or Japanese/Korean conglomerate or IBM/AMD/Philips conglomerate. I'll bet that Intel will get their first, a year before the other companies will.
IBM will be perfectly happy producing a 65 nm 3.2 GHz Cell and Xenon well into 2008, barely bothering to improve their performance.
Hard to believe, but it appears the x86 ISA will live on forever until quantum computing, molecular computing, or some or fancy thing that can change the landscape comes into being.
Originally posted by THT
I wouldn't sweat the trust thing. In 3 years, there will be only 2 corporations in the world that will be able to produce 45 nm desktop chips. One of them will be Intel. The other one, I don't know, perhaps UMD+TSMC or Japanese/Korean conglomerate or IBM/AMD/Philips conglomerate. I'll bet that Intel will get their first, a year before the other companies will.
IBM will be perfectly happy producing a 65 nm 3.2 GHz Cell and Xenon well into 2008, barely bothering to improve their performance.
Hard to believe, but it appears the x86 ISA will live on forever until quantum computing, molecular computing, or some or fancy thing that can change the landscape comes into being.
How true. It's hard to believe though, isn't it?
Originally posted by melgross
I don't know what you've been reading. but do you know anything about the way the PPC as embodied in Apple's machines works?
Yes. And if the Cell has a core processor that understands the PowerPC ISA there is no reason why any current Mac software would not run on it without a recompile. I'm not talking about internal operating system or hardware changes. That kind of transition is solely the problem of Apple's engineers and is not the topic if this debate. The transition Steve Jobs talked about was the developer transition. If developers have to do no work, then there is no transition, regardless of what people at Apple HQ might have to do. A quote from what I've been reading, which you could have figured out yourself given that I posted the link:
From the article: The Cell Processor uses a PowerPC as the core processor, making it compatible with many of the tools and users that are familiar with that chip.
do you know anything about the way ANY general purpose cpu today works?
Yes, I do. What the hell is your point other than to be inflammatory?
Can you explain to me why the PPC in the Cell is easily adapted to Apple's OS and the way the g4 and g5 process instructions?
I'm not talking about adaptability to Apple's OS, I'm talking about application-level compatibility. If current PowerPC applications run on a Cell based Mac, there is no transition. And this has nothing to do with any of the technological challenges of integrating a Cell processor into Apple's hardware designs. I was under the impression that the entire issue of transition had to do with developer/user transition, not with internal Apple engineering transitions.
This has nothing to do with FUD. That's the excuse that people who don't understand something use to hold their beliefs tightly.
I'll simply dodge your inflammatory bait. I'll go back to what I initially responded to, which was your remark "Moving to the Cell would be more disruptive than moving to x86." Maybe you meant something else, but by "disruptive" I assumed you were addressing the disruption to developers and users. Frankly I don't care how disruptive something is to Apple hardware engineers. If they got it to work well enough to ship a Mac based on it then its "disruptiveness" would be marginal at best compared to the current x86 transition, again given that it would be driven by a PowerPC core. There would be no recompiling and there would be no emulation. Applications would just work.
Read the articles on the Cell in Arstechnica.com. Read the articles on why this would be difficult for Apple to use.
Then come back here, and we'll talk.
The arrogance is remarkable. I'm aware of the problems Cell currently has. The possibility of moving to Cell would presuppose that Cell were ready for desktop use. If it were, that again would presuppose that the problems with memory controllers and the likes would have been solved, otherwise Apple wouldn't make that move in the first place. And it hasn't, which is why I never advocated that Apple move to Cell unless/until the technology is ready for prime-time. If it had been by now, then a move to Cell would _not_ be more disruptive, which is the entire point I was making. Your flaming was not necessary.
The only recurring thought I have is that something either horribly went wrong at IBM, either engineering or politics, or Intel truly showed Jobs something he couldn't turn down.
We'll know in the weeks/months ahead if engineering @ IBM went horribly wrong if there are upgrades to the Powermacs. If, as the chud tools indicate, a dual core 970 mp appears then Intel must truly have something great in the works. I believe IBM's engineering is more than capable, so I discount engineering, but can't rule out internal IBM politics. But, I've never been much a conspiracy buff and IBM has been touting PPC Everywhere. Just doesn't add up.
This is the biggest gamble Apple is making "EVER" and Steve Jobs knows it.
This is a toe to toe, nose to nose knock down battle between Apple and Microsoft. Apple won't be able to "niche player it" this time. The hardware will pretty much be the same, the OS's and software will decide. Apple has a huge task at hand. This is what makes me think Intel has something they showed Jobs that tipped him over the fence. It also makes me think that Apple may actually get some modifications on the chip, eventually, not available to Windows. I'm woefully ignorant of cpu architecture, but I just get a sense that Intel sees something in Mac OS X that will make it easier to make minor architectural changes to enhance speed and possibly reduce costs.
I give Jobs credit, this decision has taken a huge set of cahonies.
Originally posted by rickag
This is a toe to toe, nose to nose knock down battle between Apple and Microsoft. Apple won't be able to "niche player it" this time. The hardware will pretty much be the same, the OS's and software will decide.
Why would you think that way? Mac OS X will not run on a vanilla PC you know, no way in hell! What is the difference if they both have Intel inside or not. Why does this all of a sudden put Apple "toe-to-toe" with Microsoft. On the contrary, I think this will open up things like Office and Outlook so they can finally release a damn email client that fully supports Exchange.
I'll take that pill now.
Originally posted by AppleRISC
Yes. And if the Cell has a core processor that understands the PowerPC ISA there is no reason why any current Mac software would not run on it without a recompile. I'm not talking about internal operating system or hardware changes. That kind of transition is solely the problem of Apple's engineers and is not the topic if this debate. The transition Steve Jobs talked about was the developer transition. If developers have to do no work, then there is no transition, regardless of what people at Apple HQ might have to do. A quote from what I've been reading, which you could have figured out yourself given that I posted the link:
Yes, I do. What the hell is your point other than to be inflammatory?
I'm not talking about adaptability to Apple's OS, I'm talking about application-level compatibility. If current PowerPC applications run on a Cell based Mac, there is no transition. And this has nothing to do with any of the technological challenges of integrating a Cell processor into Apple's hardware designs. I was under the impression that the entire issue of transition had to do with developer/user transition, not with internal Apple engineering transitions.
I'll simply dodge your inflammatory bait. I'll go back to what I initially responded to, which was your remark "Moving to the Cell would be more disruptive than moving to x86." Maybe you meant something else, but by "disruptive" I assumed you were addressing the disruption to developers and users. Frankly I don't care how disruptive something is to Apple hardware engineers. If they got it to work well enough to ship a Mac based on it then its "disruptiveness" would be marginal at best compared to the current x86 transition, again given that it would be driven by a PowerPC core. There would be no recompiling and there would be no emulation. Applications would just work.
The arrogance is remarkable. I'm aware of the problems Cell currently has. The possibility of moving to Cell would presuppose that Cell were ready for desktop use. If it were, that again would presuppose that the problems with memory controllers and the likes would have been solved, otherwise Apple wouldn't make that move in the first place. And it hasn't, which is why I never advocated that Apple move to Cell unless/until the technology is ready for prime-time. If it had been by now, then a move to Cell would _not_ be more disruptive, which is the entire point I was making. Your flaming was not necessary.
I'm sorry, but what you posted shows nothing. Application level? And what is that supposed to mean? So you are saying that the apps run but the OS doesn't?
If you did know, you would have said that the PPC in the Cell is a non-reentry model. It has two threads, and no reusable instructions.
It's considered to be extremely difficult to get the Mac OS functioning on it. That would include any programs that run as well of course.
I don't know anyone who is impressed with the Geek article.
COULD Apple get it to work? Possibly. To what effect?
A primitive PPC core, much simpler than any Apple has ever had, designed to be a controller for the spe's (though it does have a similarly primitive VMX unit), though it can do the cpu work required for the games designed just for it, plus the spe's themselves which are basically small vector computers.
None of this in any way resembles the G4 or the G5.
You think that with all of the work that Apple has put into x86 over the years, this would be a viable alternative?
No recompile and hand tweaking would get this to work for Apple. In later years, IF IBM were to rework the chip for Apple, possibly. But we all know that they wouldn't go to all that work.
I'm just tired of people looking at the (theoretically) high number of flops this is supposed to do, and cry out- "This is it!"
It's not.
Steve's reputation will be made upon the Developers for the Mac platform. I guess most people are worried, including myself, that our machines we have now will run the latest software in 2008 & 9. That is what it comes down to, right? As long as the developers are in for supporting PPC a few years past 2007, I'm in.
I am ready for a 12" PowerBook with dual 2.0 Ghz processors that will last 4 - 5 hours on one charge and render video like my G5 iMac has never seen.
Originally posted by melgross
I'm sorry, but what you posted shows nothing.
You haven't "shown" anything either. Quit this derogatory nonsense.
Application level? And what is that supposed to mean? So you are saying that the apps run but the OS doesn't?
Given that the Cell requires quite the underlying architectural change in terms of hardware design as well as operating system magic to deal with parallelism, threading, etc, then yes, the OS would likely require significant work. However, you continue to fail to present any reason why a run-of-the-mill Mac OS X application could not run without a recompile if a standard PowerPC instruction set were available, as early Cell specs seem to indicate. If you wan't to inform me exactly why that wouldn't work, please do, otherwise cut out the baseless posturing.
If you did know, you would have said that the PPC in the Cell is a non-reentry model. It has two threads, and no reusable instructions.
This is entirely meaningless. What does this have to do with running PPC binaries without recompiles? Sony would not be talking about building workstations based on Cell if it weren't a viable desktop architecture.
[quote]http://www.realworldtech.com/page.cfm?ArticleID=RWT021005084318
Neither microarchitectural details nor the performance characteristics of the POWERPC Processing Element were disclosed by IBM during ISSCC 2005. However, what is known is that the PPE processor core is a new core that is fully compliant with the POWERPC instruction set, the VMX instruction set extension inclusive. Additionally, the PPE core is described as a two issue, in-order, 64 bit processor that supports 2 way SMT. The L1 cache sizes of the PPE is reported to be 32KB each, and the unified L2 cache is 512 KB in size. Furthermore, the lineage of the PPE can be traced to a research project commissioned by IBM to examine high speed processor design with aggressive circuit implementations. The results of this research project were published by IBM first in the Journal of Solid State Circuits (JSSC) in 1998, then again in ISSCC 2000.
It's considered to be extremely difficult to get the Mac OS functioning on it.
Which would be Apple's internal problem and not a transition issue.
That would include any programs that run as well of course.
No, it wouldn't. That would only include programs that are specifically trying to take advantage of parallelism/threading. There's no reason why single-threaded or even traditionally multi-threaded applications couldn't run on the PPE.
I don't know anyone who is impressed with the Geek article.
And what is that supposed to mean? I should discard the article because some anonymous guy on the AI forum said he doesn't "know anyone who is impressed" with it?
COULD Apple get it to work? Possibly. To what effect?
A primitive PPC core, much simpler than any Apple has ever had, designed to be a controller for the spe's (though it does have a similarly primitive VMX unit), though it can do the cpu work required for the games designed just for it, plus the spe's themselves which are basically small vector computers.
None of this in any way resembles the G4 or the G5.
A traditional G5 processor could serve as the core as well.
You think that with all of the work that Apple has put into x86 over the years, this would be a viable alternative?
Who said anything to that effect? How do you know what Apple's research efforts in other directions have been? Moreover, my entire comment was predicated on the assumption that if we were to be actually in the process of a Cell transition, these issues would have been resolved somehow. In which case, no, a Cell transition would not be "more disruptive." Or do you really think Apple would float some half-baked, slow alpha-level prototype crap into the market?
No recompile and hand tweaking would get this to work for Apple. In later years, IF IBM were to rework the chip for Apple, possibly. But we all know that they wouldn't go to all that work.
You continue to fail to mention why current code wouldn't be able to run on this technology, if it were properly factored for desktop use.
I'm just tired of people looking at the (theoretically) high number of flops this is supposed to do, and cry out- "This is it!"
It's not.
So that's what this entire rant and arrogant posturing has been about? About something I never even said? All I've said is this has potential, and that if the technology were currently in a state where it were actually a viable option for Macs that it would not be "more disruptive" than an x86 move. And I continue to maintain that. Adding VMX to the G4 didn't stop software from working, utilizing the GPU for image/video processing didn't stop software from working, and there's no reason to believe that a Cell processor with a PPC core couldn't run current software. Nothing would force a developer to refactor everything into cells, in fact that wouldn't even make sense for all software.
Originally posted by aplnub
I have had a chance to calm down and I am now excited about the news. If the DEVELOPERS don't get pissy (big if in my opinion) and support the PPC for 3 years past 2007, all is good.
Steve's reputation will be made upon the Developers for the Mac platform. I guess most people are worried, including myself, that our machines we have now will run the latest software in 2008 & 9. That is what it comes down to, right? As long as the developers are in for supporting PPC a few years past 2007, I'm in.
I am ready for a 12" PowerBook with dual 2.0 Ghz processors that will last 4 - 5 hours on one charge and render video like my G5 iMac has never seen.
You got the idea!
HERESY! ... and more!
So, what has happened with IBM?... "INTEL inside" ??? Nooooooo!!!!
No more Altivec? No 64 bit? Inferior hardware? FAT? God knows what else...
Sorry Steve, but I believe that the trust has been broken for a lot of Mac users. We can now look forward to tons of bloatware and maybe even some tasty viruses too. Maybe I'm being paranoid but this decision just does not feel or sound right at all. Sounds like Apple have spat the dummy on this one.
I fear IBM making a major breakthrough and trouncing Intel in the future with some super PowerPC technology, then Apple will really be screwed. Dell anyone?
Hows those AAPL shares doing?
Originally posted by MiMac
HERESY! ... and more!
So, what has happened with IBM?... "INTEL inside" ??? Nooooooo!!!!
No more Altivec? No 64 bit? Inferior hardware? FAT? God knows what else...
Sorry Steve, but I believe that the trust has been broken for a lot of Mac users. We can now look forward to tons of bloatware and maybe even some tasty viruses too. Maybe I'm being paranoid but this decision just does not feel or sound right at all. Sounds like Apple have spat the dummy on this one.
I fear IBM making a major breakthrough and trouncing Intel in the future with some super PowerPC technology, then Apple will really be screwed. Dell anyone?
Hows those AAPL shares doing?
Is this a silly post, or a real one?
Originally posted by rageous
I, too, have my mac for the OS, not the processor.
On the other hand, I have my Apple stock because of the whole ball of wax. And a large chunk of that ball (harware sales) are totally unpredictable right now. I don't think anyone expects a sudden surge in sales, and most don't reasonably expect sales to remain at their current levels. Most likely, we're looking at pretty weak hardware sales for the foreseeable future.
Unless of course they slash the prices, but in that case increased sales doesn't mean shit.
So yeah, many people who are nervous don't necessarily pin those feeling on the fact that OS X is going to be on Intel. It's about the state of their stock until that transition is well underway...
2 thoughts this morning (it's 7am now where i am, i have to go teach a Flash class in a few hours
1.
apple stock
i commend those that sold their stock, and also those that are hanging on. my guess (i've never owned any AAPL) is that i wouldn't get in now, if i had stock i'd bail at $35, and if i sold all my APPL i'd get back in at $25 or $30 if i wanna be in for a 1-2 year long run ~ this requires a level of faith though that many still do have....
2.
sales
yes, sales, revenues, profit, are generally expected to take a hit in the medium term. short term (within 1 month), most people outside of geekLand won't know what's going on, many are still pestering their parents/spouses/bosses/whatever to get a new iPod. medium term (2months - 1 year) is where many see sales suffering.
okay, 3 thoughts.
2.(b)
steve job's strategy to 'weather teh storm'
i am starting to believe that stevie J has some very sweet powerPC stuff up his sleeve for the next 12 months. i for one am not going to buy anything apple until the dust settles. but for the july-sep 2005 quarter, christmas quarter, then macworld sf2006, you can be sure he's going to bring out some big guns to hold revenues and profits at a reasonable level. if there's one thing about companies, even if they have a very strong and visionary long term view (as in this switch to intel), they are very very very reluctant to let their quarter-over-quarter profits slide in the hopes of 'future promises' of 'things getting better'. there'll be some sweet deals in the works for the rest of 2005 because
2.(b)(i)
any powerPC you buy now runs mac os X sweet
(ii)
said powerPC is robust, works, no gamble on intel
(iii)
large existing library of software that just works (well, at least on Panther)
2.(c) (i'm on a roll here
my guesstimate is that apple execs will choose to show the market strong sales and revenues if at the cost of profit margins. they could even pull off something like offering 1GB standard in teh Powerbooks towards the christmas season. profits might take a slight hit because they can't have their traditionally killer margins, but their priority is
2.(c)(i)
showing continued growth in iPod, iTMS and Mac sales
(ii)
showing clear progress on Intel at macworld SF 2006
2.(d) (oh man i'm never going to shut up am i)
unfortunately for us wanting to buy a new mac now, you can forgettabout it for a few weeks unless you find a really good deal or you really need one (i'm going to hog my dad's iBook g4 as much as i can). however, there are two no-brainer sales drivers in the rest of the year where we can shoot to lock in some nice powerPC-mac deals
2.(d)(i)
back-to-school, july-sep quarter, paris: new iBooks, mac Mini, etc etc
(ii)
run-up to christmas, oct-dec quarter: updated iMac g5, updated powerbooks, etc etc
2.(e) this is the last point i swear
standard warranty may be extended to 2 years, or some sort of much cheaper 'Applecare lite', or 'Applecare Xtra' which is a form of insurance covering spills, drops, theft
2.(f) shit i've gotta go get ready. peace. yes good things, but we need a mourning space. maybe i'll set up something on blogger.com