Slashdot Mirror
IBM Officially Unveils Dual-core PowerPC Chips
PM4RK5 writes "Today at the Power Everywhere Forum in Japan, IBM officially unveiled its rumored dual-core PowerPC line of chips, the 970MP. Code-named Antares, these chips have been rumored to be under development since 2004. It is believed that Apple has been working with prototypes and is likely to use them in forthcoming updates to the PowerMac G5 line. The press release is in Japanese; as of this writing, IBM has not released an English version. Some of the slides from the presentation given by IBM are available.
The processors pack some impressive specs, ranging from 1.4 to 2.5 GHz and including 1MB L2 cache per core; the chips also include the ability to power down the extra core when it is not needed. Alongside the 970MP, IBM also announced its low-power 970FX chips, ranging from 1.2 to 1.6 GHz, with power consumption ranging from 13 to 16 Watts, respectively."
Why would Apple want to waste any more time with PowerPC? I thought Intel had the most appealing "roadmap".
Now that Apple has ditched PowerPC for Intel, where is this line of chips going?
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
Still and all, Apple has been harping on about the superiority of PowerPC for so long that I'm even more surprised to see them switch when IBM has these things, which look like the answers to a couple of Apple's problems, coming up.
I'd be interested in seeing what Steve Jobs saw on Intel's roadmap for the next few years that convinced him...
Whoever corrects a mocker invites insult;
whoever rebukes a wicked man incurs abuse.
--Proverbs 9:7
This is definately not the same chip. For starters the Freescale chip is a 32-bit processor and the IBM chip is 64-bit.
Lasers Controlled Games!
We'll see. IBM promised Steve 3 GHz in one year and didn't deliver. They've announced these chips, but give no indication as to when they'll be shipping in quantity. Could be the same as before.
Also, we don't know how compelling the roadmap looks in the future. Apple will get to use these chips in the short term and then switch to Intel by the time these chips have completed their "lives." Steve may be getting what he wants now, but he knows as well as you and I that it is not necessarily an indication of things to come.
Per Square Mile, a blog about density
I'm quite sure that Apple has had full knowledge that these chips were coming for a long time.
What IBM is announcing today, this year, or even next year is not what drove Apple to go x86. Their decision was based on what's coming many years down the road. While the delay in PowerBook-usable G5 processors was a factor I'm sure, I bet Apple was more worried about what's coming after the G5.
Also, another major factor in the switch is that IBM can't seem to keep up with Apple's demand, especially at the top end of the GHz range.
So no, Apple will not be flip-flopping. (Though I bet they keep their options open and actively support PPC compiling well after x86 has fully taken over, even if only internally, in case IBM comes up with something spectacular and worthwhile from Apple's point of view.)
IBM had their chance.
Seriously, you don't think they gave Steve this news at least 6-12 months ago?
A new product announcment does not a deep roadmap make.
I think Steve saw this among a number of other bits in a meeting with Big Blue, saw it was a very weak pipeline, didn't get what he wanted in terms of pricing and development cost sharing, and was still pissed off over the 3ghz fiasco. IBM also probably wasn't terribly forthcoming, thinking they had Apple as a captive customer, probably not noticing the writing on the wall that was obvious as soon as Darwin x86 was compiled, linked, and booted. So IBM's trying to save a little face, but the horse has already won the Kentucky Derby and they're just now closing the barn door?
They boned it. The interesting question is, do they really care? I doubt it.
Speaking of Roadmaps... Why not would Apple not switch to AMD? AMD's chips run with less power consumption and way less number of transistors. When comparing the Dual core chips from AMD and Intel, AMD wins on power consumption. But I thought Jobs said Intel had the best Performance per watt? ADA4800DAA6CD (AMD Dual core 64-bit): 110W Intel® Pentium® Processor Extreme Edition: 130W These are the latest and greatest from Intel and AMD right?
From the roadmaps and rumor mill, even the Pentium EE 130 W(clocked at what, 3.8ghz?) and the AMD Athlon 110 W and too high power and not good enough on performance.
It appears Intel plans on dropping the P4 line and going to enhancing the Pentium M edition. It is expected that Apple will be going with the Pentium Ms (which apparently have dual core slated in their lineup) instead of with the Pentium EE.
In summary, Apple won't touch the Pentium EE due to high power consumption. However, they do like the Pentium M with has much better performance per watt/clock cycle and much lower power consumption.
From that I would guess that either AMD could not give Apple the same deal as Intel could. Either that or Apple expects Intel to have much better performance than AMD by that time. Also, as far as I know the Pentium Ms are much better than AMDs mobiles in power and performance.
Fly me to the moon Let me sing among those stars Let me see what spring is like On jupiter and mars
If these had been around a year ago we could be talking about Apple innovation etc, but the fact is the x86 market is ahead and Motorola/IBM have their eyes on high end servers and the embedded market.
:)).
But still, the power use of these chips is very impressive. Always liked Motorola but AMD64 is where I'm at now (it's close in name to CBM64 too
Why use a 13-16 Watt PowerPC chip when you can use a 27-watt [tomshardware.com] Pentium M?
Your batteries will last longer. It'd be nice if your laptop could last the length of long plane flights. This may not matter as much now as it did previously, what with some airlines offering outlets, and wireless.
FalconShould there be a Law?
The stated reason for the switch doesn't have to be the only reason or even the real reason. If Apple wants the option of someday competing head to head against Windows on other vendors' Intel-class hardware, they've got to get the compatible applications built and do so without killing their installed base. They're on a roadmap that does just that.
Having some life ahead in the current line of CPUs and still switching isn't without precedence either. Apple made the transition from the 68040 to PPC even though the 68060 was on the horizon.
Why is IBM unvailing it now? There are no known potential customers for this chip.
Well, IBM is a customer as will be Apple. The PowerMacs aren't planned to dump PPC for another 18 months, so you should see dual-core PowerMacs for some time here.
This announcement also helps illustrate why Apple jumped - 2.5GHz at the high-end means that Apple remains topped out on performance, and the low-power chips are okay, but really aren't low-power enough, nor fast enough to give Apple a significant gain on laptops.
Short answer, Microsoft in thier Xbox. Which is the better business opportunity, iMac or Xbox?
PowerPC is in many places you wouldn't think of. Many blade servers and storage boxes use PPC. Since IBM isn't branding "PowerPC Inside" on with thier customers, it's a little harder to tell who is using it and who isn't.
IBM wont miss Apple too much. Apple really wont grow the PowerPC business much. There's more growth for the PowerPC elsewhere, and that growth is occuring.
Also, according to the link you posted, the 27-watt Pentium M only goes up to 2.13 GHz. That is not much more than the 970FX chips which use _half_ the power for up to 1.6 GHz. I would rather have a 1.6GHz processor at 16 watts than a 2.13 GHz Pentium M at 27 watts. That is almost _twice_ the power consumption for only 0.53 more GHz. I think I will pass.
Almost half the power consumption, _plus_ the ability to run Mac OS X on sweet Apple hardware? Gee, let me think? Which one should I pick!
If Tyranny and Oppression come to this land,
it will be in the guise of fighting a foreign enemy. -James Madison
Sure they do. They have their own line of products, such as the JS20 blades -- the 5th fastest supercomputer is built with these.
They also have OEMs such as Momentum (nowadays Mercury Computer Systems) that make motherboards based on them.
In other news, nobody really cares because Apple is switching to x86 based hardware sooner than these will make it into the hardware stream.
Part of what makes the Mac experience what it is is that Apple doesn't try to cram legacy support into every product they make. With Apple it's out with the old and in with the new; PPC will be a dead end like 68k.
this is my sig
the Power4, which the power970 is a derivative of, was dual core. They were put into AS/400 (iSeries) and pSeries (think RISC/AIX) boxes years ago.
Apple got the plain jane 970 version, single core out of this chip from IBM. So the question that stands out is, why did it take so long to offer a 970 version that was dual core?
What I don't understand most about the switch Apple is making is that everyone harped on megahertz yet the AMD64 chips have great performance "ratings" with low megahertz. My current chip is only a 1.8G and many Power chips are just that as well, so where is the big boost for Apple except in the powerbook line? A well designed chip, and PowerPCs are very well designed, can run circles around faster chips as AMD has proven with the AMD64 series.
* Winners compare their achievements to their goals, losers compare theirs to that of others.
Alongside the 970MP, IBM also announced its low-power 970FX chips, ranging from 1.2 to 1.6 GHz, with power consumption ranging from 13 to 16 Watts, respectively.
This sounds exactly like what Apple needed for a G5 powerbook. Did Steve just get a little too impatient? Had he waited another month maybe he would have found the answer for a G5 powerbook? Did Apple threaten IBM that they would go to Intel if something didn't change soon? (and now IBM has delivered, but perhaps a bit too late)
I don't understand how some people are saying "OMG Apple switched at the wrong time oh noes!@#!!"
Does everything HONESTLY think Apple didn't know the exact release date of the 970MP BEFORE they announced their switch?
Apple knew when and where this was going to be released, and they know when and where Intel will release their next series. They switched because they wanted to, this isn't a surprise to them.
"I wonder if Apple will reconsider the decision regarding the migration."
After the WWDC and the trauma it inflicted on some devs, I find it highly unlikely that Apple is going to suddenly decide tomorrow that they've made a bad move and are going to stick with the PPC path in the future. Apple knew this G5 development was coming, hence the comment that has been repeated numerous times that the next 2 years are going to produce some interesting developments in the PPC platform, but by 2007, things will be at a point where Intel will overtake them and that the PPC roadmap does not offer anything that can keep up with the pace of Intel. Jumping hardware platforms is hard enough as it is, jumping back would work to obliterate the confidence that Steve Jobs has tried to instill in those who support the Mac. He and his fellow execs are trying very, very hard to appear as if this is really worth it and that they have a solid plan that will not leave 3rd parties burned.
Nobody wants to have another Amiga situation, where every week there is a new roadmap to follow, dramatically different than the one before. That is the perfect way to scare off the community that keeps a platform going.
Looks like Steve Jobs is getting all the things he said he couldn't have. ... a year late (*). That's why he's switching.
It's "put a lot of effort and money in to be a year or more late" vs. "get crazy R&D for free and be guaranteed to be current". Tough choice.
(*) Ignoring for a moment that he was also promised 3GHz by mid-2004!
AMD has approximately 20% of the PC Processor market.
Apple has 3% of the PC Market.
3% of the PC market is 15% of AMD's market.
AMD's market is normally capped not by distribution but rather by production. If AMD won the Apple contract, they would EITHER need to increase their production by 15% (not historically AMD's strong suit), or increase prices to the PC market...
If AMD picks up the Apple contract and CANNOT increase production...
Then AMD has to reduce their PC market-share by 15% of their production, which means increasing prices.
Either way, Apple would be a HUGE account for AMD, and would require a substantial portion of AMD's manufacturing resources.
Alex
Please :-)
Why wouldn't it be feasible for them to support products with both processors? They kept an up-to-date version of OS X on x86 for 5 YEARS! Now they have a developers kit that can produce binaries for both PowerPC and x86 without any extra work. It would be very easy for them to support both.
But laptop CPU upgrades are too expensive to even develop. There isn't even a market for PC laptop upgrades.
If you paid $2500 for a 2003 G4 PowerBook, and used it for 3 years until you feel it needs an upgrade, it makes a lot more sense to sell it for $900 and buy a new 2006 $2500 PowerBook than to pay $500 to upgrade your old PowerBook and still have a 2003 PowerBook with a slightly faster CPU, that is still worth $900.
It sometimes makes sense to upgrade components in the PC world, but CPU accelerators for Macs haven't made any sense since the days when new Macs cost $8000.
The G5 PowerMac was so much better than the G4 PowerMac on so many levels, that it simply doesn't make sense to buff up a G4 rather than getting a new G5.
PowerBooks upgrades make even less sense. They depreciate faster. Laptop parts are small and light and get banged around more than a desktop, so they don't hold up as well.
An upgraded PoweBook CPU wouldn't be nearly as fast as a new one, it would likely not work right, and the rest of your PB would still be 3 years old.
If you can't afford a new PowerBook, stick with what you got, and maybe throw in a new HD and more RAM if it makes sense. But the idea of putting in a faster CPU is just throwing money down a hole.
Anyway, the original comment sounded more like they wanted to cling to a PowerPC out of misplaced emotional attachment, not because they wanted to make the most of an investment.
How are they "topped out for performance"?
There's nothing here significantly stronger that what we've had for the last several months. Sure, dual core vs. dual G5 has benefits, but without a clock boost or anything else, it's not much more than packaging.
I would say IBM's offerings are competitive. Steve Jobs and his "wah I wanted a 3GHz chip!" is all bullshit when you look at it; he wanted to compete with the Intel marketing machine, and still hasn't noticed that AMD Opteron chips top out at 2.6GHz - and have done for some time. The G5 is competitive in that it matches or outperforms the AMD Opteron (that frontside bus helps).
Yah, and Apple can have pretty much the same performance with Intel chips. It's not the top end that drove the decision, it's in part the laptop chips. 1.6GHz G5 is good, but the Pentium M looks like the place to be for a while.
Ultimately, I suspect its Intel DRM that drove the decision. If the media giants want DRM, they're going to go with Intels, and anyone else is going to have a uphill fight. Better to be in the game from the start there.
Their dual cores top out at 2.2GHz and also probably will for some time. Apple still have the potential to create a 2.4GHz PowerMac and an XServe with up to 4 processors which competes with AMD's most expensive and little used 400 and 800 processor lines.
Apple has always been able to do 4-way with the G5. Dual dual-core isn't *that* much easier than 4-way. If the market was there for 4-way we'd have it.
I don't see why it "illustrates" anything except that Steve Jobs is a nut job who lied himself through a developer conference. It's a damn shame Apple has gone so low and a damn shame the developers are so loyal that they keep so quiet.
How did he lie? Everybody knew about the MP before the conference. The transition lasts until 2007. The decision surely wasn't made based on any product that will ship between now and then. In fact, the MP is likely the main reason why the PowerMacs will be last to move - the performance is there with PPC in the near term. Will IBM make a 980, though, or is this the end of the line?
There is a lot of dissent in private quarters. All that PowerPC hype Apple pushed down our throats - some of it actually real as it turned out - has left 1000s of developers with a lot of AltiVec code and not a lot of choice. They are mighty
pissed about rewriting their apps again, especially to bridge the gap between now and the 10 years in the future that Steve's Intel Roadmap says they will have better integer performance.
Oh, come on. There aren't 1000s of Altivec apps. There are thousands of apps using Apple's SIMD libraries (which are already ported to SSE3) and using other OS X libraries, but there are *maybe* hundreds of apps using Altivec directly. Apple isn't worried about them being pissed. Developers will chase the money like everyone else and Apple isn't a bad ride right now. They'll suck it up, diddle their code and start selling product again.
The biggest dissent seems to be inside IBM. Not long after the 970 came out IBM merged their semiconductor and server groups. The semiconductor group that was quietly making small profits (due to low pricing to Apple) on 970s was publicly costing the server group big profits by taking a bunch of HPC contracts that could have gone to IBM 970 or POWER systems (IBM execs would of course claim that every Apple HPC sale was a lost IBM server sale since it was the CPU that carried the day).
Once those groups merged, Apple's sweet deals went away and Apple was expected to pay their way. The contract came up, Apple didn't want to pay the enormous development costs of a CPU line, so they saw their opportunity to further commoditize the product line and struck a deal with Intel. In the end, consumers and most developers really won't have a problem, and it *might* (who really knows) portend good things for Apple.
It is common to analyze the revenue that IBM lost with the apple x86 switch. But don't forget about the huge market that was recently announced by Microsoft and Sony. The game console market is a super-tanker compared to the bathtub-toy macintosh market, and just as Intel is more hurt by the xbox 360 going to IBM, IBM has gained more. IBM is running a revenue surplus in recent PowerPC commitments.
Why was this modded off topic? Hellooooo, moderators. The topic is about IBM PPC chips, not Apple! The poster has a good point...