But, to me, it was an important part and Apple's move to Intel is an indication of a failure at some (or many) levels within IBM.

The PPC will continue, but without an important segment that could have significantly delivered increasing orders if IBM had delivered increasing performance.

Maybe its a failure at Apple. Don't know why IBM is getting all the blame for this transition. Wait because Steve Jobs says so? Microsoft, Nintendo and Sony jumping on the IBM bandwagon doesn't seem like any failure at IBM to me. Chances are next next gen consoles will use some variant of the PPC to me.

Quote

Chances are next next gen consoles will use some variant of the PPC to me.

Next Gen game consol does not equate to desktop computer.

Oh boy I can't wait for Nov. 22nd!

Quote

Anonymous wrote:
Maybe its a failure at Apple. Don't know why IBM is getting all the blame for this transition. Wait because Steve Jobs says so? Microsoft, Nintendo and Sony jumping on the IBM bandwagon doesn't seem like any failure at IBM to me. Chances are next next gen consoles will use some variant of the PPC to me.

Ummm, no. Actually, everyone doesn't say that the mistake was from IBM because SJ said so, but because IBM really failed to bring the G5 to its full potential, and to unleash good enough technologies. Had they done that, Apple would never have switched to iNtel. They would have stayed with IBM, who would be developping superb new processors. There's always a reason, you know. Oh, and yes, There were certainly failures from Apple at some point, but iBM is still the responsable for the (relative) failure of the G5 chip.

Perhaps Apple made cost demands IBM refused to accept? People forget, without the clone business, IBM basically customizes POWER cores for Apple, and fronts all the R&D costs. Yes, the PowerPC 970 (GPuL) is basically a single core POWER4 processor minus some cache and power consumption. IBM tailored this chip to Apple's desktop needs. Steve Jobs has been known to make ridiculous cost demands on parts to benefit Apple's bottom line (as any good CEO should, see how Apple cut costs on 68000 in the 80s). IBM decided it didn't make sense to further pipeline the PPC 970 and modify to get it above the 3GHz threshold. Especially when IBM was preparing the PowerPC 980 specs (which like the PowerPC 604 would have been a ground up design, not a customization). Of course, Steve being Steve announced the 3GHz mark, and thought his RFD would move IBM along. It didn't. The work still continues on the PowerPC 980 (which will all be dual core), and will showcase how scaleable POWER can be (designed to be in low and high end portables and workstations). With this generation, IBM wanted Apple to either pay part of the R&D (both Microsoft and Sony paid IBM for Xbox360 & CELL), or commit to a large order (Steve has been known to make small orders upfront, hoping for further discounts when the lines fully ramp up. Fishkill has not even reached 25% fab capacity because of this). Steve thought IBM needed Apple for POWER to remain high profile, but with three new major customers buying 10x as many cores, the suits in Armonk had enough. They made their demands. Steve decided to announce his Intel transition plans at the final moment to IBM out of spite. Intel probably made some concessions on core pricing in hopes of landing the high profile iPod for XScale sales. People are quick to point out IBM missing the 3GHz mark, but only Steve promised this. Intel failed to deliver its announced 4GHz Pentium 4 (followed by announcing a future based on Pentium M cores). Of course, now that Apple and Intel are together (for good and bad), this is ignored. Sadly, IBM's low-power PowerPC 970 "LP" is suitable for a PowerBook. In fact, the PowerBook G5 chipset was finalized save for the MPU (the 970FX was simply too hot for the housing). With the Intel transition full steam ahead, and Apple mostly disbanding the PowerPC portable team, the PowerBook G5 will not be. Heck, the most Apple will muster is higher density screens. They won't even use updated Freescale cores (higher speed e600 and new e700 cores). It's just tiring to hear people go on and on with FUD about IBM being bad.

Perhaps Apple made cost demands IBM refused to accept? People forget, without the clone business, IBM basically customizes POWER cores for Apple, and fronts all the R&D costs. Yes, the PowerPC 970 (GPuL) is basically a single core POWER4 processor minus some cache and power consumption. IBM tailored this chip to Apple's desktop needs. Steve Jobs has been known to make ridiculous cost demands on parts to benefit Apple's bottom line (as any good CEO should, see how Apple cut costs on 68000 in the 80s). IBM decided it didn't make sense to further pipeline the PPC 970 and modify to get it above the 3GHz threshold. Especially when IBM was preparing the PowerPC 980 specs (which like the PowerPC 604 would have been a ground up design, not a customization). Of course, Steve being Steve announced the 3GHz mark, and thought his RFD would move IBM along. It didn't. The work still continues on the PowerPC 980 (which will all be dual core), and will showcase how scaleable POWER can be (designed to be in low and high end portables and workstations). With this generation, IBM wanted Apple to either pay part of the R&D (both Microsoft and Sony paid IBM for Xbox360 & CELL), or commit to a large order (Steve has been known to make small orders upfront, hoping for further discounts when the lines fully ramp up. Fishkill has not even reached 25% fab capacity because of this). Steve thought IBM needed Apple for POWER to remain high profile, but with three new major customers buying 10x as many cores, the suits in Armonk had enough. They made their demands. Steve decided to announce his Intel transition plans at the final moment to IBM out of spite. Intel probably made some concessions on core pricing in hopes of landing the high profile iPod for XScale sales. People are quick to point out IBM missing the 3GHz mark, but only Steve promised this. Intel failed to deliver its announced 4GHz Pentium 4 (followed by announcing a future based on Pentium M cores). Of course, now that Apple and Intel are together (for good and bad), this is ignored. Sadly, IBM's low-power PowerPC 970 "LP" is suitable for a PowerBook. In fact, the PowerBook G5 chipset was finalized save for the MPU (the 970FX was simply too hot for the housing). With the Intel transition full steam ahead, and Apple mostly disbanding the PowerPC portable team, the PowerBook G5 will not be. Heck, the most Apple will muster is higher density screens. They won't even use updated Freescale cores (higher speed e600 and new e700 cores). It's just tiring to hear people go on and on with FUD about IBM being bad.

Perhaps Apple made cost demands IBM refused to accept? People forget, without the clone business, IBM basically customizes POWER cores for Apple, and fronts all the R&D costs. Yes, the PowerPC 970 (GPuL) is basically a single core POWER4 processor minus some cache and power consumption. IBM tailored this chip to Apple's desktop needs. Steve Jobs has been known to make ridiculous cost demands on parts to benefit Apple's bottom line (as any good CEO should, see how Apple cut costs on 68000 in the 80s). IBM decided it didn't make sense to further pipeline the PPC 970 and modify to get it above the 3GHz threshold. Especially when IBM was preparing the PowerPC 980 specs (which like the PowerPC 604 would have been a ground up design, not a customization). Of course, Steve being Steve announced the 3GHz mark, and thought his RFD would move IBM along. It didn't. The work still continues on the PowerPC 980 (which will all be dual core), and will showcase how scaleable POWER can be (designed to be in low and high end portables and workstations). With this generation, IBM wanted Apple to either pay part of the R&D (both Microsoft and Sony paid IBM for Xbox360 & CELL), or commit to a large order (Steve has been known to make small orders upfront, hoping for further discounts when the lines fully ramp up. Fishkill has not even reached 25% fab capacity because of this). Steve thought IBM needed Apple for POWER to remain high profile, but with three new major customers buying 10x as many cores, the suits in Armonk had enough. They made their demands. Steve decided to announce his Intel transition plans at the final moment to IBM out of spite. Intel probably made some concessions on core pricing in hopes of landing the high profile iPod for XScale sales. People are quick to point out IBM missing the 3GHz mark, but only Steve promised this. Intel failed to deliver its announced 4GHz Pentium 4 (followed by announcing a future based on Pentium M cores). Of course, now that Apple and Intel are together (for good and bad), this is ignored. Sadly, IBM's low-power PowerPC 970 "LP" is suitable for a PowerBook. In fact, the PowerBook G5 chipset was finalized save for the MPU (the 970FX was simply too hot for the housing). With the Intel transition full steam ahead, and Apple mostly disbanding the PowerPC portable team, the PowerBook G5 will not be. Heck, the most Apple will muster is higher density screens. They won't even use updated Freescale cores (higher speed e600 and new e700 cores). It's just tiring to hear people go on and on with FUD about IBM being bad.

Very nice of you to jump in defense of (your employer ?) Big Blue …
Only a few problems with your presentation of the facts:
1. If SJ hadn't been given assurances, he probably wouldn't have mentioned "a 3 GHz next year". If anything, the man is far from stupid …
2. In 2 years, the top of the line G5 went from 2 Ghz to 2.7 GHz. And the 2.7 runs so hot that you need a very complex cooling system for it …
3. Yeah ! IBM finally announced (but isn't producing!) a G5 chip that can be fitted in the PowerBook line. I hope we will see at least one or two in production before PBs are switched to Intel.
4. IBM seems to have lost interest in improving the G5 as it was busy getting on the game console platforms—which aren't using G5s, but a modified Power-based chip and the Cell …
5. How do you account for a whole quarter last year with no iMacs to sell because IBM couldn't deliver G5s for the new model ?? Is that also due to SJ's ego according to you ??? If it is, I would like to know how you can rationalize this line of thinking …
Don't you think that if IBM had been able to deliver, the whole Mac line would be equipped with G5s, and no mention would have been made of Intel processors ? If you don't, I think you're way off base …
The simple truth is that the G5 is an awesome chip, but its power requirements are excessive and IBM hasn't been able to control that. I, like most longtime Mac users, deeply regret it, but a company can't be run on promises without delivery, which seems to be what IBM, as a supplier, seems to have been all about for Apple for the past three years. Sorry if I'm bursting your bubble, but it's a fact …
Their interest in game consoles is a good business decision, by the way, from their perspective, but excuse us all for thinking that the change of focus shouldn’t have happened: the G5 shouldn't have been put on the back burner. The two types of processors should have had equal time … I can't blame SJ for resenting NEGLECT !

Hey, I'm a Mac consultant in DC and its funny how all the Mac people are jumping on the Intel is BEST and PowerPC sux bandwagon.

I am sick of all the no 3GHZ PowerPC BS. You don't need 3GHZ to achieve performance and is performance really what Apple is looking for? Uh not really. If so why are they going with Intel? AMD Athlon 64s and X2s outperform all their current P4/Pentium D counterparts and its not likely to change in the future. It hasn't changed since AMD has introduced the Athlon.

Get educated on processor performance people and you will see that Apple will get cheap processors from Intel. Performance is a far secondary consideration. FireWire will be dead, Intel will take care of designing the North bridge and South Bridge chipsets and the logic board for Apple. Money savings for Apple and reduced R&D.

This MacIntel thing is to increase Apple's margins. Thats all!

IBM is very capable of manufacturing high performance CPUs. Even next year the G5 will be the only high volume quad processor system in production even after Apple releases the Intel Macs.

Quote

Anonymous wrote:

I am sick of all the no 3GHZ PowerPC BS.
IBM is very capable of manufacturing high performance CPUs. Even next year the G5 will be the only high volume quad processor system in production even after Apple releases the Intel Macs.

The 3 GHz Bs is a fact … not a very interesting one, but still an anecdote that illustrates the problems Apple had with IBM.

There were serious problems with the production of G5s, as any Apple consultant should know. The 3 most serious ones were
1. Failure to deliver for the iMac last summer …
2. Failure to deliver for the PB from the very start …
3. Too damn hot …

Obviously IBM is able to produce great chips, but just as obviously, they can't be counted on to deliver on their promises …

I personnally love the G5 and will keep using it for years, but that doesn't change the business aspect of Apple's supply problems, and THAT's no BS …

Quote

kenaustus wrote:
But, to me, it was an important part and Apple's move to Intel is an indication of a failure at some (or many) levels within IBM.

The delays (Xserve, iMac, high-end Power Mac) were really painful but I think Apple wanted to switch in the first place. Apple has been maintaining Mac OS X on PPC and x86 since the beginning. The Marklar project was revealed in aug. 2002, at MacWorld NY '02 Steve Jobs said that once the transition to OS X was finished another transition to x86 would be a real option.

Quote

Anonymous wrote:
AMD Athlon 64s and X2s outperform all their current P4/Pentium D counterparts and its not likely to change in the future. It hasn't changed since AMD has introduced the Athlon.

The current P4/Pentium D has nothing to do with the switch, Apple will use the next Intel chips, Merom/Conroe/Woodcrest.

Quote

Anonymous wrote:
Even next year the G5 will be the only high volume quad processor system in production even after Apple releases the Intel Macs.

A quad Power Mac is not really a high volume product, how many of those will they sell?

hollywoodindustry.com: (8/4/05) In a technology demonstration at SIGGRAPH 2005, AMD and Boxx Technologies unveiled a four-way dual core Opteron workstation running Microsoft Windows Server 2003 and a beta 64-bit version of LightWave 3D animation software.

Dear Apple gods, the ones who make things happen ,

Please keep your options open. Now that you have convinced all the developers to compile binarys for both chip-sets, have Xcode doing 99% of the work for them there, and have people open to Mac boxes with intel chips, keep your options open.

If PowerPC chips are better and readilly available, use PowerPC.
If intel chips are better and readilly available, use intel.

If you have to, use them both, I don't care, just don't make me use inferior chips, no matter who makes them.

On that note, i like the new quad core macs - when they ship. Have you considered shipping multi-processor multi-core Power5 or Power6 macs? <drool> i'm sure they would cost more, but someone would buy 'em.

AAPL friend... keep that stock ticking up, and a hefty dividend would be nice if it stalls a little .

To the which chip is better crowd... from looking to the past... the ones they can actually deliver.

As to which platform is better... the one that produces the RAW numbers the best.

And finally, SPEC/WATT is important.

... oh, and one more thing ... I see in the near future, Apple has a new series of adds... that end with 'do-ding-do-dang... intel inside'

Maybe Apple just wants some free Advertising Dollars

peace friends

PS Steve Jobs... no matter how many mistakes you have made over the past few years, Congradulations on turning a company around. Best wishes with Pixar too. Look foward to hearing about those deals once you get things ironed out. =+) Might have to buy a little of that stock too. Stay healthy. We love you man.

Lets get some facts straight here. Apple were going to go to intel once it became obvious Moto weere not interested with developing the ppc. IBM then desperate to stop them promised the G5. SJ gave them a chance and yes he dictated the 3GH dictat. They failed to meet that and then they failed to make a cool enough chip for the power books. What customer in their right mind would feel confident thet future promises could be met?
Then SJ seees that IBM is bending over backwards to help out MS and that there is even less likelyhood thet promises will be met or indeed with big new customers that IBM will even be interested in Apple now that they used them to get to larger chip output targets which will now benefit other customers. Apple had no real choice but to go to Intel. AMD would have been better short term in theory but Apple can't actually afford to upset intel and commiting to AMD would have tied them to yet another chip supplier with a suspect future through no fault of their own Equally price and lower power chips would still be a question mark with AMD. fact is that SJ did what was right for business, future markets and developments and for Apple's future. can't blame him for that and maybe thats why he is up there and we are down here whining.

Quote

engrish wrote:
The current P4/Pentium D has nothing to do with the switch, Apple will use the next Intel chips, Merom/Conroe/Woodcrest.

They will apparently be using the new Yonah chip for the 'Books, according to the latest Mac Format issue.

Quote

macinnerd wrote:


They will apparently be using the new Yonah chip for the 'Books, according to the latest Mac Format issue.

Great ! Does SJ himself know it yet ? Maybe we should send him an issue to let him know !

dc here

Everyone likes to buy into SJ "facts" that G5 is a lousy performer, the same chip that put VA Tech in the top 5 supercomputers in the world and built in just a few months.

Now Intel is superior. Please everyone take a look at the idle temps of Celerons and you will find that they are higher than AthlonX2s at work. Intel in the X86 world is an underperformer. Its future offerings are nothing special. Turion and Mobile Athlons already outperform Intel's future offerings.

As far as the Quad G5 being a mass market desktop system I believe that I will be able to go into any Apple store and buy them I can't exactly by a BOXX at Best Buy or Circuit City can I?

G5 in a flat panel iMac is a totally dumb idea. Apple could of had duals in their via e600 or e700 cores. Its not like anybody uses 64 bit code/OS or needs access to 16GB of RAM sheesh

Same thing with PowerBooks. People insist on G5s in PowerBooks. Whaddaaya going to do with a 64bit processor running a 32bit OS?

Its all a bit nuts really. My 667 dvi tiBook is not much different in speed from a friends 1.67 ghz albook - after all, humans arent really capable of doing much at all at 1.67 ghz.....
Typing is still around 20-40 wpm for most people, and printing takes 5-15 ppm.
Watching a dvd takes exactly the same time, no matter what processor you have.
No-one moves platform because of 'speed' issues - its mostly availability of popular software that affects sales.
Who knows why Steve Jobs wants intel? Who knows why my wife insists on undercooking chicken?
Who cares?
I dont really feel sorry for those millions who wish to render full length feature films on their laptops whilst flying from LA to London.
Its all just marketing bollocks, when you really think about it.
And its amusing to see people getting all hot about 3.0 ghz G5 - most cant afford the machines at the top end, and almosts all have zero need for it.
Hey! Dont let that stop you buying one - it desnt stop idiots from buying huge SUV's that only one person will ever drive in and that goes from home to work and back on paved roads.
I remember a time when Apple people werent such dolts as they are today.
C'est la vie.

Quote

Guest wrote:
dc here

G5 in a flat panel iMac is a totally dumb idea. Apple could of had duals in their via e600 or e700 cores. Its not like anybody uses 64 bit code/OS or needs access to 16GB of RAM sheesh

Same thing with PowerBooks. People insist on G5s in PowerBooks. Whaddaaya going to do with a 64bit processor running a 32bit OS?

Exactly! I thought that 64 bit was going to be like 64 x 64 times faster or some such nonsense. If 64 bit is that much faster, then surely we can run a 50mhz chip faster than a pentium 100 ghz chip. No?
Its all lies and marketing.

Well, yes, marketing is the point. That's how every big company is supposed to make you want their newest product.

Quote

macinnerd wrote:
Well, yes, marketing is the point. That's how every big company is supposed to make you want their newest product, not like supposed innovation from Apple, Apple tricks us into buying their junk.

I'd just like to state for the billionth time that the fact that IBM hasn't hit 3GHz means absolutely nothing. In the two years after IBM released the G5, they boosted its clock speed by more than any other desktop chip in the industry. So what difference does it make if they missed the magical 3GHz mark? None. They still outperformed Intel in every regard. The entire industry hit a slump. I certainly understand (and share) your frustrations as Mac users, but if you've been following the technology world at large (instead of just the Mac world), you should know that IBM has not been failing.

The G5 still outperforms Intel's best. And there's really no sign of that changing.

If the reason for the switch was performance (which is really not a valid assumption), then the blame belongs on Motorola/Freescale, who has never been able to adequately push the G4 line forward. From day 1, Moto has been a huge liability.

SJ never made any comment that the G5 was not a good performing processor. On the contrary, at his June keynote, he emphasized that there would still be exciting PowerPC products coming until the eventual migration to Intel. His point was that "performance-per-watt" was an important consideration as Apples product roamap goes forward. This implies, at least to me, that the focus of Apple will be on portable computing. And while the real world impact of a 3.0 ghz G5 versus a 2.7 ghz unit is negligible, the psychological impact for marketing purposes is undeniable. Most consumers have no idea about the difference between CISC and RISC. All they know is that "more ghz means a faster computer." Perhaps one day when the performance/watt issue is worked out by IBM (and I believe with the PowerPC's wide acceptance in consoles this will eventually happen) and the dawn of universal binary encoding, we may eventually see some portable G5 machines, or a transition to a PowerPC based G6 or G7 in the future.