Intel's Plans for Low-Power Chips

[deputy_doofy]

But where would Apple go with their PowerBook? 3 more 167MHz speed bumps in the next 2 years?

This is so true. Also, what good would a 2GHz G4 be running on a 167MHz front-side bus (or a 200MHz front-side bus)?
The G4 was a good chip but it is time to put it to rest.

The G5, if it were given the attention it deserves, could probably destroy most of the x86 line in performance, but IBM refuses to put their energies into it. Hell, I still think the G5 is nothing more than IBM's "version" of the G4. A PPC chip based off the Power-5 would probably be amazing, but alas, we can only hope that Intel has gotten their own sh...tuff together and stop cranking out nuclear chips that need to go 5,000,000,000 GHz to prove their "superiority" to PPC.

 

[rolandf]

Stiff competition on all fronts

As we don't know the exact reason behind the move from IBM to Intel, a lot remains speculation.

But, certainly, serving MS and Sony, it is hard for Apple, to be the third (if we neglect Nintendo). Therefore,
for IBM it makes sense to push Apple to use one of the chips, the others use.

The change to Intel probably will not remedy the situation, as there is already a lot of potential
conflict between the two parties.

First, there are much bigger Intel customers than Apple, e.g. Dell, and so to develop and produce
special chips for Apple alone will imply the same difficulties as with IBM.

Second, Intel is pushing itself into the home-entertainment market, copying the Mac Mini, and probably
equip them with MS software.

So, a close look reveals, that at the moment, MS the best positioned company, as they have PPC and Intel based
platforms for their software.

I can just repeat, it would make sense for Apple to serve the AMD / Intel platform for the low-mid end, and use
Cell for the high-end.

Anyway, as so often, IBM has been at the forefront of chip design developments, that have been copied several
years later by others, e.g. dual-core processors. The Cell will be no exeption, but with the difference, they already have it.

 

[shamino]

For x86, it isn't the case. More registers, you can assume the processor has SSE2 (MMX, SSE1, etc) at least rather than making it i386 compatabile, the memory space is flat, etc.

I've got an Opteron box at work which speeds-up 25% when running the x86-64 binary over the i386 binary.

Comparing Opteron-optimized code against 386-optimized code isn't fair. There are huge architectural differences that have nothing to do with being 64-bit.

How about comparing Pentium-4-optimzied code against Opteron-optimized code. At least that would be close.

Or even better, compare Pentium-4-optimized code against 64-bit Intel x86 code. (I realize that this may be a hard test to run before the Intel 64-bit systems become more commonplace.)

 

[clownie]

5X not 5 Watts

The graph says 5x more performance per watt. That could be five times the performance at same watts, or same performance at 1/5 the watts, or anywhere in between.

 

[sjk]

I think it would be more accurate to speculate that some buyers may be attracted to Intel Macs if they're capable of running Windows (and...Linux?) in addition to Mac OS X.

Okay, I thought "some" was implied in what I wrote.

If I had to guess how many buyers, I would say not very many. It may have obvious geek appeal, but not much real world interest beyond that. There is at least empirical evidence (lol) that the majority of average, bread and butter OS X users choose OS X because they want Windows off their machine (and thier lives), not to have Windows back on it again - like some nagging in-law that just won't go away.

I agree it wouldn't interest most "typical" home computer buyers but could attract certain businesses and schools (which I mentioned) and geeks (which you mentioned). But support-related complications could nullify the advantages.

And I think Apple would prefer to address Windows compatibility by designing OS X to be easily interoperable with existing Windows systems as opposed to having Apple users install a competing OS (and given the present state of Windows, the security risks and extra OS management that come with it) on an Apple box along side it's flagship OS.

That may be Apple's preference but if they're profitably selling computers to customers running Windows, et.al. on them (without Apple's support) I doubt it would matter. That's coming from the current perspective of Apple still doing very little (if anything) to make some key advantages of OS X over Windows clearer to the average public.

And who is going to offer support for Windows on a Mac? Apple? Not a chance. If you go out and buy a Dell box, you can at least get support for Windows by calling Dell. A Mac user with Windows installed on his Intellimac is going to be an orphan; no one to turn to for support should something go wrong with Windows - which is all but guaranteed these days if it's hooked up to the net. Not a good state of affairs for the average user.

I've speculated elsewhere that there could be new business opportunities for supporting Windows on Apple hardware (scary as that sounds ). Still, the negatives of multiple OS support (excluding self-supporting geeks), even with a single OS running on each box, might overwhelm any benefits.

Putting Windows (or Linux) on an Apple Intellimac just doesn't jive with Apple's "it just works" mission for the Mac platform, which is a major reason why folks use OS X. And they are Windows-free at last.

Yep, plenty of "switchers" we know and have heard about are relieved if/when they can completely abandon Windows.

Personally, the idea of Apple knowingly selling Intel-based computers to customers who intend to run Windows on them seems unappealing (even appalling) on the surface but I could accept that as a necessary evil if it were part of a longer-term OS X migration strategy. Sell enough OS X compatible hardware and people will eventually figure out the best OS to run on it.

 

[heisetax]

Actual Speeds

I think we should wait and see. IBM promised 3 GHz, which seemed like a really good thing, and everyone thought that the switch from Motorola to IBM needed to happen. I agree that it made things better, but we're still not at 3 GHz, more than a year later than we were supposed to have it. My point: Just because Intel promises great things, that doesn't mean they are going to happen. Didn't Intel also promise 4 GHz P4's?

Remember that the speeds of these new processors will not be as fast as the old P4s. The Pentium M tops out at 2.1 GHz now. The indications to me are that the new chips will come in at those speeds or slower. Right now they are concentrating on power usuage & not MHz speed. We still won't see 3 GHz until 2008, if even then.

Bill the TaxMan

 

[Mitch1984]

Apple wants Intel chips early

Wanted to Post This in a New Thread But the forum wouldn't let me.

http://www.theinquirer.net/?article=26433

THE POPULAR MEDIA KEEPS reporting that Apple plans to release its X86 based Macs in June next year, probably with a big fanfare and many black turtlenecks.
So far so good. The problem is that the chips that Apple is going to base its systems on - Woodcrest and Merom - are not due until the third quarter of next year.

This would make it hard for Apple to release computers if there are no CPUs for them, unless it has decided to use Yonah processors. Apple is big, but not nearly big enough to bully Intel, unless it has to do with iTunes DRM. So just what can Apple do? It could attempt to put pressure on Intel to give it the necessary chips early. So Apple is doing just that, and sources claim that it isn't flying as well as it had hoped. Any tale of cool, need or woe won't fly very far against the argument of "Dell sells 20 times what you do, why should we give you preferential treatment again?". Those same sources claim that this what Intel is saying.

 

[RobHague]

Hmmmmm

What would happen if Intel turned around and said "Stuff you Apple"? They won't but just hyperthetically speaking... they already cut ties with IBM.... where would there be left to go.

But im thinking, Apple must have known about the Intel Roadmap, they must know the timing of the Intel CPU's... they didnt just decide "Oh! Lets go to Intel" one day. It says "unless it has decided to use Yonah processors" so thats probably what they are going to do then.

Edit:

However since INTEL are in PC's and as they say, DELL sell 20x more at least than Apple's 'nitch' market at the moment... its if Intel will think its worthy of special treatment (as in the longrun will it benifit Intel). It's said that IBM didnt make much profit from the G5 from Apple... if that profit was 'small' to IBM then what of it to Intel??? Hmmm thought provoking stuff.

 

[ScubaDuc]

Wanted to Post This in a New Thread But the forum wouldn't let me.

http://www.theinquirer.net/?article=26433

THE POPULAR MEDIA KEEPS reporting that Apple plans to release its X86 based Macs in June next year, probably with a big fanfare and many black turtlenecks.
So far so good. The problem is that the chips that Apple is going to base its systems on - Woodcrest and Merom - are not due until the third quarter of next year.

This would make it hard for Apple to release computers if there are no CPUs for them, unless it has decided to use Yonah processors. Apple is big, but not nearly big enough to bully Intel, unless it has to do with iTunes DRM. So just what can Apple do? It could attempt to put pressure on Intel to give it the necessary chips early. So Apple is doing just that, and sources claim that it isn't flying as well as it had hoped. Any tale of cool, need or woe won't fly very far against the argument of "Dell sells 20 times what you do, why should we give you preferential treatment again?". Those same sources claim that this what Intel is saying.

That is why everybody is speculating that the first intelmacs will be the mini/i-Book using a Yonah processor. In fact, I seem to recall reading that Apple has asked developers to code for SSE 2 which is supported by Yonah

Powerbook are expected come later with a dual-core Merom, except in those PB G5 threads where it is expected for ......yesterday????

 

[~Shard~]

That is why everybody is speculating that the first intelmacs will be the mini/i-Book using a Yonah processor.

Which makes sense. The iMacs are arguably the strongest in the Mac line-up right now when considering power, features and price, so there is no need to update them to Intel in the near future. As for the PowerMacs, they will no doubt be the last Mac to go Intel, as Apple will be waiting for the Conroe/Woodcrest chips, which aren't out until later in 2006.

Powerbook are expected come later with a dual-core Merom, except in those PB G5 threads where it is expected for ......yesterday????

Oh, it was due long before yesterday according to some people, trust me...

I'm not sure what Apple is going to do with the PowerBook, but if it, too, will be receiving an Intel chip later rather than sooner, i.e. the Merom, then Apple needs to release a substantial PPC upgrade to the line to tide people over - none of this 32.8 MHz speed bump crap. If this is the case, then I would see the next PowerBooks having a Freescale 7448 in it, or something like that, with better video cards and HDs as well.

 

[ammon]

none of this 32.8 MHz speed bump crap.

What are you talking about? That is faster than my first 486 computer!!

 

[~Shard~]

What are you talking about? That is faster than my first 486 computer!!

And you got that computer, when exactly, 10 years ago? Yeah, back then it would be something impressive, nowadays, it just doesn't cut it I'm afraid...

 

[Val-kyrie]

It has probably been pointed out already, but Merom is a 35W processor, not a 5W processor.

However its typical power consumption is 3-5W or thereabouts. I think the current Pentium M has an average power consumption of around 10W and it idles at 5W whilst the TDP is 21W or 27W depending on the speed and FSB. http://en.wikipedia.org/wiki/Pentium_M

Note that the AMD Turion is a 25W processor at 2.2GHz at the high end, but probably also has an average power consumption of around 10W. AMD will be releasing dual-core Turion processors next year too.

You do realize that Intel's power consumption ratings do not include the memory controller whereas AMD's do, don't you?