PowerPC RoadMap from Motorola

[MacRumors]

Several readers have noted this PowerPC core roadmap (http://www.motorola.com/mediacenter/news/detail/0,,4126_3482_23,00.html) announced by Freescale Semiconductor, a wholly owned subsidiary of Motorola.

According to the press release, Freescale detailed newly announced e300, e600 and e700 PowerPC cores. The e600 processor is described as "an enhanced version of the high performance G4 core used i nthe award-winning, high performance MPC74xx family of PowerPC host processors" and is expected to scale beyond 2 GHz as well as implement Altivec. The e700 core is expected to scale beyond 3GHz.

Freescale's implementation of these processors appears to be for specific system-on-chip (SoC) implementations, but presumably the technology could find its way into future Apple products as well. Apple presently uses MPC74xx processors to power their portable line.

[SuperChuck]

A day late and a dollar short, IMHO.

[Laslo Panaflex]

See, competition does drive inovation, or at least evolution. It took Apple to drop Motorola from thier high end machines for them to see that they need to make newer and faster chips. I can't wait to see what this holds for future Apple portables, 2ghz g4 ibooks anyone?

[virividox]

unless motorola comes out witha 4 ghz NOW then i don't think its worth it. give me g5!!!

[Koodauw]

unless motorola comes out witha 4 ghz NOW then i don't think its worth it. give me g5!!!

I agree somewhat, but as Laslo Panaflex some 2.0 Ghz portables would be very nice. If they game out fairly soon that is.

[fukuhela]

Freescale Semiconductor also disclosed today at SNDF its plans to develop the next-generation 32/64-bit e700 PowerPC core and corresponding e700 platform.

I know it isn't developed yet but are actually planning on making a 64-bit processor??? Or is it just me who doesn't know what they are talking about :D :D ?

[dead_parrot]

I'm thinking that Apple is looking solely to IBM these days. I think they want to distance themselves away from Motorola as soon as possible. The real winner out of the Power PC relationship seems to be the 970.

buh bye Motorola

[gekko513]

Freescale Semiconductor also disclosed today at SNDF its plans to develop the next-generation 32/64-bit e700 PowerPC core and corresponding e700 platform.

I know it isn't developed yet but are actually planning on making a 64-bit processor??? Or is it just me who doesn't know what they are talking about :D :D ?
Or, more likely, it's they who don't know what they are talking about :D

[rastalin94]

Actaully keeping the G4 line alive and increasing the speed could help Apple keep the consumer and pro line different.

Pro Line = G5
Consumer Line - G4

a 2GHz iBook would be a very nice little machine.

[crenz]

There've been many roadmaps posted by Motorola. I'm curious to see whether they'll actually be able to come up with the new processors, or whether they are vaporware.

[Frisco]

I hope they do make these processors and Apple uses them. Competetion is good thing. We need choices. Hopefully this will help lower the prices on low end Macs.

Motorolla has been unreliable in the past. Apple putting all of its eggs in one basket is not a good idea, whether Motorolla or IBM. Most seem to have confidence in IBM, but so far they haven't been reliable either--in terms of upgrades to their chips.

We the consumer need as many companies making chips as we can get!

[sabbath999]

I for one am VERY concerned about the "all the eggs in one basket" thing Apple is getting into with the Fishkill processors. Already we have seen major delays in the G5 line. Even though I hate Motorola with a passion, I hope they keep developing the G4 through the next couple of years "just in case",

[minstryoffunk]

I'm thinking that Apple is looking solely to IBM these days. I think they want to distance themselves away from Motorola as soon as possible. The real winner out of the Power PC relationship seems to be the 970.

buh bye Motorola

I disagree…while I'd put much more stock in IBM's development projections, I think that Apple realizes the importance of leaving their options open in order to always have the best hardware possible. Look to their implementation of the new G4 in the pBook revision - while people on the boards dismissed it as too little too late, its stil progress. An 80mhz increase and further refinement is, after all, better than nothing. Any chance Moto has gotten their act together? I guess we'll see. As for distancing, I wouldn't expect that any time soon.

Personally, I'm hoping for the continued success and improvement of the 970, but if Moto can deliver, even if its far down the line, than I look forward to it.

[mrsebastian]

come on, it's the same ol' [bleep] from moto. always a day late and a dollar... err, ghz short. g5 across apple's entire line, with the high end chips in the towers of course.

[minstryoffunk]

I for one am VERY concerned about the "all the eggs in one basket" thing Apple is getting into with the Fishkill processors. Already we have seen major delays in the G5 line. Even though I hate Motorola with a passion, I hope they keep developing the G4 through the next couple of years "just in case",

Chalk the delays up to the new process. Intel has had a hell of a time with 90 nano and their technology doesn't shrink as well. Unless something radical happens in the semiconductor industry, every subsequent process migration will get much more difficult. From what I read, Intel tech has reached its limit sizewise, but IBM/Moto have another migration yet.

There is tremendous pressure on IBM, so things should be coming along soon, or I suspect they will be in trouble.

[Mr. MacPhisto]

unless motorola comes out witha 4 ghz NOW then i don't think its worth it. give me g5!!!

I'm not too sure the G5 would be the best chip for portables. Would you rather have a 1.6 or 1.8 Ghz G5 or a dual-core 2GHZ G4 with Rapid I/O and an SOC design? Some stuff I've read indicates that the 2GHZ dual core chip FreeScale is working on runs @ 25W dissipation. To me, it's pretty clear that FreeScale's chips, if they can deliver the e600s in the near future, are a better alternative than the G5. I actually think you can split up the chips over portable and desktop lines:

Apple Desktops: G5
Apple Portables: G4

And the e700 would give Apple a 64bit mobile processor when it comes out. Granted, we'll have to wait and see, but if FreeScale can deliver these things on time (and the 7447A came in ahead of schedule) then Apple will have two viable chip manufacturers.

[T'hain Esh Kelch]

Actaully keeping the G4 line alive and increasing the speed could help Apple keep the consumer and pro line different.

Pro Line = G5
Consumer Line - G4

a 2GHz iBook would be a very nice little machine.
Nice to know that the Powerbook is a consumer product. You should tell Steve that, 'cause he doesnt know it!

[Mr. MacPhisto]

Chalk the delays up to the new process. Intel has had a hell of a time with 90 nano and their technology doesn't shrink as well. Unless something radical happens in the semiconductor industry, every subsequent process migration will get much more difficult. From what I read, Intel tech has reached its limit sizewise, but IBM/Moto have another migration yet.

There is tremendous pressure on IBM, so things should be coming along soon, or I suspect they will be in trouble.

Apple and Motorola actually have two migrations left in their fabs. The new Crolles plant, as well as Fishkill, are set up to go down to 45nm. Crolles is set to start at 90nm (this summer) and is set up for a migration to 65nm and to 45nm. Fishkill will also go down this far.

http://www.semiconductors.philips.com/products/asic/partners/jdp/

Plus, Crolles has allied Moto with Phillips & STM. We'll have to see what FreeScale comes up with at this new fab. I think we'll be impressed.

[rdowns]

I'm thinking that Apple is looking solely to IBM these days. I think they want to distance themselves away from Motorola as soon as possible. The real winner out of the Power PC relationship seems to be the 970.

buh bye Motorola

Would be Apple repeating a mistake again. Not that they haven't done that before.

Competition is good and means innovation. Would be nice to have Motorola and IBM trying to leap frog each other. (Of course, that would mean Motorola taking a huge leap forward) Apple would be smart to have 2 chip suppliers, especially if Motorola can modernize the G4. Look what having a single supplier did to Apple's line and Motorola's innovation.

[rdowns]

Nice to know that the Powerbook is a consumer product. You should tell Steve that, 'cause he doesnt know it!

Nice to know people love to post things without any regard to what was actually written.

He said a 2GHz iBook would be nice. An iBook is not a PowerBook.

[fukuhela]

Or, more likely, it's they who don't know what they are talking about :D

LOL :D :D :D

Actually I think you're on to something there. In their advertizing dep. they've read that the new thing is 64-bit processing. So they just say that what they gonna make without knowing what it involves.... This way they are on to the 'beat' :eek: :D :p :p

[LaMerVipere]

I think that apple needs to diversify their portable market the same way they have, of late, attempted to with their desktops.

They should have at least 4 different types of portables to choose from:

•Desktop Replacement - Not necessarily 1-inch thin, it is allowed to be physically larger and at the same time deliver far more power, as it is for users who probably won't be taking it outside the home or office much, and it could be used to implement technology such as the G5 into the portable line much faster

•Mid-Range System - The size and power of the current PowerBooks is fine for this

•Budget System - The current iBook is fine for this

•Sub-Notebook - Less than 1-inch thin, ultra-portable, speed is not the overriding issue here, but rather size and weight

I think Apple needs to stop putting "all of its eggs in one basket" if you will, not necessarily with the processor makers, so much as their own product line, because as it is right now, it doesn't allow for much flexibility.

[Windowlicker]

Actaully keeping the G4 line alive and increasing the speed could help Apple keep the consumer and pro line different.

Pro Line = G5
Consumer Line - G4

a 2GHz iBook would be a very nice little machine.

I didn't read the roadmap article, but I could think there's a good use for the coming chips in portables.. g5 isn't gonna be on a powerbook for a while. that's a fact. and the point when it reaches the iBooks (and iMacs/eMacs) is even further. also we don't want the portables stay as they are now forever.

this can only be a good thing. i can't understand why some people have rated this as negative :Q

[Rower_CPU]

Nice to know people love to post things without any regard to what was actually written.

He said a 2GHz iBook would be nice. An iBook is not a PowerBook.

But if you're differentiating pro and consumer lines based on the processor, PowerBooks are "consumer" machines.

I don't think he was referring to the iBook part.

[areyouwishing]

I have never had a problem with Moto, I mean, they are slow (in producing faster chips), but it seems that they are the tried and true horse at this point. Their BIGGEST problem is bus speed.

What everyone isn't realizing is that moto has gotten really close to the G5 with the new portables... a 1.5ghz chip is only 100mhz slower than the 1.6 apple is currently selling in their desktops. Coupled with the more efficient use of altivec than the g5... and you have a rather nice processor.

Its almost refreshing to see that everyone is realizing that its quite hard to produce faster chips, and moto isn't the only one thats not good at it.

[rdowns]

But if you're differentiating pro and consumer lines based on the processor, PowerBooks are "consumer" machines.

I don't think he was referring to the iBook part.


The original poster said G5s in the Pro line and G4s in the Consumers line. He didn't say a G4 in PB, he referred to a 2GHz G4 iBook.

Ergo, PM/PB=Pro and eM/iM/iB=consumer.

Does any of this really matter? :D

[Rower_CPU]

The original poster said G5s in the Pro line and G4s in the Consumers line. He didn't say a G4 in PB, he referred to a 2GHz G4 iBook.

Ergo, PM/PB=Pro and eM/iM/iB=consumer.

Does any of this really matter? :D

If you're going to accuse someone of not reading the post they're responding to: yes.

If G4 = consumer and PB = G4 then PB = consumer.

[chasingapple]

Well, since these new Moto chips have a better chance of getting into the iBook before the G5 does im all for it. Bring it on Moto! :)

[macFanDave]

to have significant numbers of chips from both IBM and Motorola in their various product lines. If those two suppliers could compete fairly (no blackmail, extortion, bribes or kickbacks (you hearing this, MicroShaft?)), all parties concerned could be big winners.

I'm convinced that AMD was the best thing that ever happened to Intel. After it was clear that Moto wasn't able to keep with them, Intel could have easily relaxed into Monopolist mode (notice how much Internet Explorer has improved ever since it "killed" Netscape?) But AMD showed up and lit a fire under Intel's butt and has driven steady progress.

If going with IBM's G5 is what it took to get Moto to move out its keester, then it was a great move. And if Moto's new offering makes IBM try even harder to get the 3GHz G5 chips out even faster, well, that's really great! That's how competition is supposed to work! That's when it drives people to do their best.

Having said all that, let's see if Moto can follow their own road map -- history has shown that they're not really good at that, but I wonder if anything will be any different with IBM eating their lunch.

[shyataroo]

is it just me or is everyone buying into the 64-bit myth? granted the G5 dual 2.0ghz runs circles around a Dual 1.25G4 however would it still run circles around a dual 3.0ghz G4? probbably not because if you look at the chart comparing the 1.6ghz G5 to the dual 1.25Ghz G4 you see the 1.6ghz G5 barly beats it now than you get a Dual 2.5ghz G4 and right there you have a processor just a hair shorter than the Dual 2.0 G5 and for about $1,000 less and than there is that new super processor that kicked the pipelines out of the 2.0ghz celeron and the processor was only at 300mhz! (more proof of the megahertz myth or now the Gigahertz myth) and was incredibly cheap you get that things speed upto 1ghz and it will burn a dual 3.0ghz G5 and possibly a quad G5 you make it a 64 bit core too and it will be 20 times faster than todays top of the line computer.

[shyataroo]

I want to see IBM and mortorlla and Intel and AMD work together on the single fastest processor ever just put aside thier differences and aligences and make the first Terrahertz processor for unix. (and since OS X is basically unix with a GUI...)

[ITR 81]

I have a feeling the next Moto speed bump will be 1.8GHz or 2.0GHz for the iBooks, iMacs, and maybe PowerBooks if they haven't squeezed in a G5 yet.

Next upgrade??
PM: 2.5-3.0Ghz?(G5)
PB: 1.8-2.0Ghz?(G5 or G4)
iBook: 1.8-2.0Ghz?(G4)
iMac: 1.5-2.0Ghz?(G4)
eMac: 1.33?(G4)

[miloblithe]

I think that apple needs to diversify their portable market the same way they have, of late, attempted to with their desktops.

They should have at least 4 different types of portables to choose from:

•Desktop Replacement - Not necessarily 1-inch thin, it is allowed to be physically larger and at the same time deliver far more power, as it is for users who probably won't be taking it outside the home or office much, and it could be used to implement technology such as the G5 into the portable line much faster

•Mid-Range System - The size and power of the current PowerBooks is fine for this

•Budget System - The current iBook is fine for this

•Sub-Notebook - Less than 1-inch thin, ultra-portable, speed is not the overriding issue here, but rather size and weight

I think Apple needs to stop putting "all of its eggs in one basket" if you will, not necessarily with the processor makers, so much as their own product line, because as it is right now, it doesn't allow for much flexibility.

Apple doesn't really have enough market share for that much product diversity. The more similar the portables are (note they all have the same keyboard, for example) the cheaper they are to make. Also, the advantage of the iBooks and Powerbooks over PC laptops is that they combine most of the best qualities of the categories you list above. They are smaller and lighter than most PC laptops, yet more powerful than sub-notebooks (and have optical drives).

Desktop replacements are, for the most part, just cheaply made, incredibly heavy laptops. Someone in one of my classes brought in a desktop replacement Dell one day. When he opened it up, I thought I heard the sound of a croud roaring at a baseball game. It took me a few seconds to realize that it was the computer's incredibly load fan. 9lbs, and as loud as an aircraft carrier. No thanks. The other end of desktop replacements, the incredibly powerful laptops for gamers, is really a niche market. In my book, the 14" iBook is enough of a cheap desktop replacement and the 17" Powerbook is enough of a high-end desktop replacement.

As for subnotebooks, the 12" Powerbook is really quite small. They could make a model without the optical drive and save a pound and a little thickness, but I don't think they'd sell enough to make it worthwhile.

On the other hand, I would love to see more options for customers. Like a 1.5Ghz G4 in the 12" powerbook as a custom option.

[swdrumcp]

If the G4s fsb ran at the speeds of the G5, then watch out. If a G5 and G4 ran at the same speed say 2ghz with the same fsb then the G4 looks very attractive.

I see no reason not to use G4s in portables if things keep progressing.

Should be interesting to see how it plays out.

[rastalin94]

My post of G4 consumer and G5 pro was based on the fact that power books will move to the G5 as soon as they can. I would expect to see faster G4's in the iBook and eMac line if they become available instead of them switching over to the G5. The iMac would be an exception, I would expect that to be a G5 some time soon.

Continuing development of the G4 can only be a good thing. But I agree this is nothing more than talk, I will believe it when it happens.

[stockscalper]

I have never had a problem with Moto, I mean, they are slow (in producing faster chips), but it seems that they are the tried and true horse at this point. Their BIGGEST problem is bus speed.

What everyone isn't realizing is that moto has gotten really close to the G5 with the new portables... a 1.5ghz chip is only 100mhz slower than the 1.6 apple is currently selling in their desktops. Coupled with the more efficient use of altivec than the g5... and you have a rather nice processor.

Its almost refreshing to see that everyone is realizing that its quite hard to produce faster chips, and moto isn't the only one thats not good at it.

According to speed tests by Barefeat.com the 1.6 mhz G5 blows the doors off the new 1.5 ghz powerbooks. The powerbooks use fast ram, fast hard drives and fast video cards. So, what's the difference. Bus speed primarily. The G5 has an 800 mhz frontside bus while the G4 has a 167 mhz bus. Moto just doesn't get it.

Another consideration is that the new 970 FX G5's put out about half the heat those new G4's will produce. The new G5 pulls 12 watts while the G4's in the latest powerbooks draw twice that. Has anybody used one of the new powerbooks on battery? I have and got just a little over 2 hours on it. I can get almost 5 on my old TI 500 (it has the new battery design in it) that draws only 7.5 watts.

[thatwendigo]

I hope they do make these processors and Apple uses them. Competetion is good thing. We need choices. Hopefully this will help lower the prices on low end Macs.

Fat chance. Motorola's charged more for the G4 than IBM did for the G5, and I find it unlikely that they're going to change their policy when they're still strong in the defense embedded industry, where you can charge $10,000 for a hammer.

Motorolla has been unreliable in the past.

This is the key problem with this "roadmap." Motorola promises the world, and then deliver DeMoines, Iowa.

Apple putting all of its eggs in one basket is not a good idea, whether Motorolla or IBM. Most seem to have confidence in IBM, but so far they haven't been reliable either--in terms of upgrades to their chips.

What lack of reliability is this that you're talking about? Many people say this, but I don't recall any promises about chips that would have been broken yet. All that Jobs said was that Apple and IBM would hit 3.0ghz in a year.

We the consumer need as many companies making chips as we can get!

No, we need companies making good chips. I'll take one good chip over eight bad ones any day.

I'm not too sure the G5 would be the best chip for portables. Would you rather have a 1.6 or 1.8 Ghz G5 or a dual-core 2GHZ G4 with Rapid I/O and an SOC design?

I'd rather have a 750vx running at 2.0ghz and up, since it's being produced by a company with a real stake in the computer market. Motorola is an embedded company that's been selling to Apple for a while now. I don't believe for a moment they're going to make a dual-core processor that would be at all cost-effectice for Apple.

Some stuff I've read indicates that the 2GHZ dual core chip FreeScale is working on runs @ 25W dissipation...Granted, we'll have to wait and see, but if FreeScale can deliver these things on time (and the 7447A came in ahead of schedule) then Apple will have two viable chip manufacturers.

I just don't trust Motorola anymore, especially not after some of the stunts they pulled with the 74xx core over the years. FreeScale's chips plans sound nice, and I'd prefer even those to a single G5 in the PowerBooks. They're just not a good bet until they're on a loading dock and being brought in to slot into machines.

--

Oh, and while I mostly agree with Miloblithe, he also left out a key aspect of not going hog-wild on product lines... Apple's done it before, right around the time the company was starting to spiral out of control. Jobs put a stop to that.

[Mord]

i hope apple will use this e600 chip just so OSX supports it

then powerlogix can roll out the dual dual core 3GHz e600 cube upgrade oh yhea :cool:

when they deliver the goods i will be happy untill then go ibm

[Some_Big_Spoon]

Well, this should pan out in 10 or 12 years..

[PPC970FX]

Well I do not care who makes the CPU. As long as it is fast, and good. But in the long run IBM has a LOT more $$$ than moto so IBM is in the long run a better CPU maker.

Read more: http://www.technewsworld.com/perl/story/33566.html

[Durendal]

I'll believe it when I see it. The G4 had a ton of potential, but quite frankly, Motorola sucks. The G4 should be at 2+ghz right now, but Moto has their head too far up their ass to get it out the door. The G5 bit is nothing. We will see 3ghz by the end of summer. The reason Apple went to the PPC in the first place was because Motorola couldn't get the 040's out and ramped up fast enough. This new chip wont clock at 2ghz. It'll clock initially at about 1.5, and it'll take Motorola 2 years to get it up to 2ghz. History has shown us that much. Screw Motorola.

[Dont Hurt Me]

I'll believe it when I see it. The G4 had a ton of potential, but quite frankly, Motorola sucks. The G4 should be at 2+ghz right now, but Moto has their head too far up their ass to get it out the door. The G5 bit is nothing. We will see 3ghz by the end of summer. The reason Apple went to the PPC in the first place was because Motorola couldn't get the 040's out and ramped up fast enough. This new chip wont clock at 2ghz. It'll clock initially at about 1.5, and it'll take Motorola 2 years to get it up to 2ghz. History has shown us that much. Screw Motorola.I agree..............and still feel sorry for those new G4 buyers who are Macgamers :( Durendal has spoken the truth. Screw Moto and its Last Place G4.

[thatwendigo]

I agree..............and still feel sorry for those new G4 buyers who are Macgamers :( Durendal has spoken the truth. Screw Moto and its Last Place G4.

You know, for once, I'm just not going to argue with DHM. Motorola has had plenty of opportunity to scale the G4 by now, and yet they're hanging around at 1.5ghz at the current design. Somehow, I just don't see a 2.0ghz enhanced G4 any time in the near future, and when it comes to the desktop... Well, the G5 is a better system in all respects.

[kenaustus]

I think a lot will depend on how free Freescale is from Moto. If Moto keeps screwing things up then IBM has a wide open field. If Freescale can do their own thing (be totally free of Moto) then there is a chance to do some good.

I think that their main challenge, however, will be to try to catch up with IBM's Power 5 chip - which might hit the Mac this summer - and the neat things that will come after that, and come rather fast. It's very difficult to go against a company now producing chips at 90 nm and with the facilities to migrate to 65 nm and then 45 nm.

[Penman]

I have never had a problem with Moto, I mean, they are slow (in producing faster chips), but it seems that they are the tried and true horse at this point. Their BIGGEST problem is bus speed.

You should note that even on the 'short' bus the new PB's benchmark withing a hair of the 1.6 G5's. Bus bandwidth isn't the choke point 90% of the time. I want a faster bus too but the G5 architecture is not any faster than G4 at equivalent speeds with todays applications. The G4 PB's the first time we've seen a G4 close to G5 speeds and the gap dissapeared.

I'd live the G5 mobile to be a G4 with improvements. As long as the power requirements stay level or drop and the performance improves I don't care about the age of the architecture. Look at PC's - tons of legacy and as fast as the best new designs through sheer ingenuity.

[nmk]

I wouldn't mind a 2Ghz G4 in powerbooks at all. I've seen benchmarks that show that the G5's arene't much faster than the G4's Mhz for Mhz. The G5's are just clocked much higher. As I'm quite doubtful that Apple would be in a position to introduce anything higher than a 2 Ghz G5 in the Powerbooks by the beginning of next year, I really don't see that as a big jump. Motorolla will have the G4's up to at least 1.8 Ghz (if not 2 Ghz) by the same time. So if they came up with a G5 powerbook, you would basically end up with a Powerbook that runs hot, consumes a lot of power, and isn't any faster than its G4 counterpart. If you look at the barefeats comparison below, you will see that the 1.6 G5 is only very marginally faster than a 1.5 G4 powerbook at rendering FCP. This is only one of the benchmarks I have seen illustrating this same point. I think people going on about the G5 Powerbook are just getting cought up in the hype and not necessarily in touch with reality.

[Mr. MacPhisto]

I'd rather have a 750vx running at 2.0ghz and up, since it's being produced by a company with a real stake in the computer market. Motorola is an embedded company that's been selling to Apple for a while now. I don't believe for a moment they're going to make a dual-core processor that would be at all cost-effectice for Apple.




Problem is that the people I know at IBM have said the VX project was dumped by Apple in January - due to IBM not being able to ramp up 90nm and the new offerings Motorola would have before the VX would see production (pushed back to late summer with the delays). The VX project is dead. IBM and Apple are working on a SOC project, but that may also be cancelled before the year is out.

[Bendit]

I hope they do make these processors and Apple uses them. Competetion is good thing. We need choices. Hopefully this will help lower the prices on low end Macs.

Motorolla has been unreliable in the past. Apple putting all of its eggs in one basket is not a good idea, whether Motorolla or IBM. Most seem to have confidence in IBM, but so far they haven't been reliable either--in terms of upgrades to their chips.

We the consumer need as many companies making chips as we can get!

Unless Motorolla makes the g4 line 64bit compatible they won't keep using both processors for long. The sooner will all go 64 bit the sooner we can get a 64bit OS and 64 bit software. Or else the using 64 bit G5s will have been useless.

Staying with 32 bit G4s will just make the transition longer and harder.

[Dont Hurt Me]

Just a comment, every bench i have seen shows a single 1.6 G5 kicks G4 butt and most of the time is neck and neck with a dual on SMP aware apps. Lets not start thinking a G4 is the same clock for clock as a G5 because it just is not so. that benchmark is one of the closest. G5 is still running G4 software, when G5 stuff comes up there will be no comparison. a 1.5 G4 does not make for a 1.6 G5. just had to say that for those who think these 2 animals are about the same performance wise. one more thing and that is it took moto over 1 year to go from 1.42 to 1.5. I dont plan on seeing any G4 speed bumps anytime soon. G4 does not compete with Intel/AMD. Apple needs a Cpu that can compete and win. G5 will be that chip. Im waiting for a G5 iMac.

[zync]

I want to see IBM and mortorlla and Intel and AMD work together on the single fastest processor ever just put aside thier differences and aligences and make the first Terrahertz processor for unix. (and since OS X is basically unix with a GUI...)

I realize you're joking but the only problem there is that of architectures. Depending on which way they went they could either develop for x86 or PPC. Well, since all of those companies produce x86 chips for PCs AFAIK, they'd probably develop for x86 and we'd be screwed. OS X is basically a Unix (FreeBSD) with a special display manager (Quartz) that was written to support the PPC architecture...

[Neuro]

PowerBook G5s will take months as already stated by everyone in te know.
PowerBook G4 2Ghz+ are interesting, as they will compete directly with Intel Centrino notebooks, and may actually be faster(!). Good stuff.

[nmk]

DHM, i really don't know what benchmarks you're talking about, but all the benchmarks I've seen show the DP G4's beating the single 1.6 G4 by large margins. I have already posted one benchmark that shows the 1.6 G4 barely beating a 1.5 powerbook in FCP rendering. Mhz for Mhz, I really don't believe there is that big a difference. Why don't you post links to these supposed benchmarks that show the superiority of the G5. I can provide plenty of benchmarks that will show that the 1.6 is not competitive with most of the faster DP powermacs.

[hose this!]

Who cares??!?!??!?! Bring on the new Studio Displays dammit!!!!!!!

[thatwendigo]

Problem is that the people I know at IBM have said the VX project was dumped by Apple in January - due to IBM not being able to ramp up 90nm and the new offerings Motorola would have before the VX would see production (pushed back to late summer with the delays). The VX project is dead. IBM and Apple are working on a SOC project, but that may also be cancelled before the year is out.

If Apple bought that load of crap, then they're not half the company they ought to be right now. Motorola took how long to make those last weak jumps? Until Freescale establishes themselves as being past the shortsighted failures of their forebears, they're suspect and complicit in the performance gap that IBM and Apple closed considerably.

If the VX is actually dead, then that's a sad, sad thing to have had happened. IBM is far better than Motorola, in my opinion, and even if there are issues with 90nm, the 750vx would still have been better at 130nm than the G4 was at the same process.

Just a comment, every bench i have seen shows a single 1.6 G5 kicks G4 butt and most of the time is neck and neck with a dual on SMP aware apps.

There are no 1.6G5 duals. What are you talking about? Also, as someone just up this thread pointed out, the new PowerBook 1.5s are pretty close to the G5 1.6s in performance. Not a 1:1, but not bad for only being 100mhz behind.

Lets not start thinking a G4 is the same clock for clock as a G5 because it just is not so. that benchmark is one of the closest. G5 is still running G4 software, when G5 stuff comes up there will be no comparison. a 1.5 G4 does not make for a 1.6 G5. just had to say that for those who think these 2 animals are about the same performance wise.

Actually, there is G5-aware and optimized software, but it's almost universally expensive pro-level apps like Photoshop CS.

one more thing and that is it took moto over 1 year to go from 1.42 to 1.5. I dont plan on seeing any G4 speed bumps anytime soon. G4 does not compete with Intel/AMD. Apple needs a Cpu that can compete and win. G5 will be that chip. Im waiting for a G5 iMac.

While I don't expect this amazing G4 revision, the one thing that does make me pause is that they're not using the older 74xx core. The ex00 line is a modified G4 core, and so I might end up eating my words if Motoroloa does create a chip somewhat like the 750vx. Taking the slightly older tech and adding advances since then might yield the power saving features with higher performance, the same way that the Centrino ends up competing with P4s that are double its clock.

If that kind of computing efficiency could be achieved, I'd look at Freescale favorably and try to think of them as separate from their roots. Until it does, though, this is just smoke and mirrors in the form of a press release.

[aldo]

My god, why does everyone think G5 optimized software is going to be so much faster - it wont.

For a great example of this, look at Linux and AMD64. 64 Bit Linux has been out for a hell of a long time now, and it's not twice as fast. Infact, it's basically the same but better at complex mathematical operations.

This isn't like going from 16 bit to 32 bit. 16 bit was hugely flawed and was restraining computing. 32bit will do us just fine - the only reason we need to go to 64 is for extra memory addressing.

Also, I think it would be better if Apple just switched to x86-based arch. However, as we all know that will not happen and instead we will be left with CPUs that seem great at first but seriously start to wane when they have problems bumping up the speed. I would actually be moderately surprised if they hit the target of 3GHz by summer - look at the problems IBM is having with the 970FX and personally I doubt they are going to be able to mass produce hundreds of thousands of these.

[h'biki]

According to speed tests by Barefeat.com the 1.6 mhz G5 blows the doors off the new 1.5 ghz powerbooks. The powerbooks use fast ram, fast hard drives and fast video cards. So, what's the difference. Bus speed primarily. The G5 has an 800 mhz frontside bus while the G4 has a 167 mhz bus. Moto just doesn't get it.
.

Really?

According to their FCP rendering speeds, the 1.5 G4 Powerbook is only slightly slower than the 1.6 G5.

Which is not surprising. If you go through their benchmarks, clock for cock, the dual G4s aren't *that* much slower than the dual G5s. (A 2ghz G5 is 1.4 times as fast (clock speed wise) than a 1.4ghz G4, and performs around 1.5-1.6 times faster.)

I'm sure if Moto fixed the FSB and scaling issues, the G4 would compete favourably.

The G4 is a good chip with a choking fsb.

[areyouwishing]

Don't Hurt Me, your theory about the G4 going from 1.42->1.5 in one year is flawed.

First, is that it is generally accepted that the 1.42 was an overclocked 1.33 (or something of that nature) when it went into the PowerMacs.

The next... in 10 months IBM will has gone from 2.0ghz to what? OH WAIT, they haven't done a speed bumb since introduction... i almost forgot.

According to speed tests by Barefeat.com the 1.6 mhz G5 blows the doors off the new 1.5 ghz powerbooks. The powerbooks use fast ram, fast hard drives and fast video cards. So, what's the difference. Bus speed primarily. The G5 has an 800 mhz frontside bus while the G4 has a 167 mhz bus. Moto just doesn't get it.

What benchmarks? Game Benchmarks? A videocard on a laptop is far inferior to most modern video cards... even with the same model number.

The 1.5ghz "has" fast RAM, it doesn't "use" fast ram.

The only benchmarks i saw on barefeats were for games, and one MP aware photoshop test, in which the single 1.5 is compared to duals.


and a couple of notes DIRECTLY FROM barefeats.com...
When we ran Bryce, the 1.5GHz PowerBook was 2.7% faster than the 1.42GHz MP Power Mac! Why? Because Bryce is NOT dual processor "aware." Ditto for FileMaker Pro. It ran 7.6% faster on the 1.5GHz PowerBook.


To put it another way, when we disabled the second processor in the G4/1.42GHz MP Power Mac and re-ran our Final Cut Pro test, the advantage over the G4/1.5GHz PowerBook dropped from 54% to 6%.


Of course, you would never want to disable the second processor in the real world. But don't expect your SINGLE processor Power Mac to overwhelm a PowerBook running at a similar clock speed....

...All PowerBooks tested had 64MB video memory while the graphics cards in the Power Macs had 128MB. We're not convinced that would have made much difference, since Quake3 Arena is dual processor aware and Unreal Tournament 2004 uses the second processor for sound. However, we will bring you test results on the 128MB version of the Mobility Radeon 9700 in a few days.


Also, keep in mind that the Mobility Radeon 9700 in the PowerBooks is NOT the equal of the Radeon 9700 Pro AGP card in the Power Mac. The first has 4 pipelines and 128bit memory bus. The second has 8 pipelines and 256bit memory bus.

[Dont Hurt Me]

valid remarks but IBM has some legs and we will see them the next few years this is why Apple is moving that way. Moto the G4 could be a kick ass chip if the FSB problem could be fixed. i guess it cant Moto and all. though solid cant stand eye for eye with Intel or AMD and G5 still needs two to stand with them. G4 is a hasbeen and G5 is a newborn getting its legs. sort of like os9 and 10 or rather X.

[dongmin]

If the VX is actually dead, then that's a sad, sad thing to have had happened. IBM is far better than Motorola, in my opinion, and even if there are issues with 90nm, the 750vx would still have been better at 130nm than the G4 was at the same process.

1. Motorola roadmaps are way too optimistic. It's either vaporware or 1-2 years behind. So unless Freescale can actually deliver this e600 in large volume, it doesn't exist, as far as I'm concerned.

2. The 7447A is at the end of the road, no? Unless Apple does something radical to increase the battery power and keep the temp. down, I can't see the G4 going past 1.6 ghz or so. And outside of some wild wild rumors, I haven't heard of any further 74xx revisions.

3. 970fx in a PB? I'm doubtful of that. The 970fx seems pretty close to the 7447A in terms of power consumption at the same clock. Add to that, the potentially hotter-running controller, etc. of the new architecture, the 970fx is too hot. I'm putting my money on something else, a 970-derivative tailored for laptops. Sometime in the spring of 2005.

4. Probably one very modest PB upgrade until then. Minor changes to graphics, RAM, HD, etc. with maybe a 100 mhz increase. Nothing worth writing home about.

[sonyrules]

I have a feeling the next Moto speed bump will be 1.8GHz or 2.0GHz for the iBooks, iMacs, and maybe PowerBooks if they haven't squeezed in a G5 yet.

Next upgrade??
PM: 2.5-3.0Ghz?(G5)
PB: 1.8-2.0Ghz?(G5 or G4)
iBook: 1.8-2.0Ghz?(G4)
iMac: 1.5-2.0Ghz?(G4)
eMac: 1.33?(G4)

I think the iMac will have a higher speed than the iBook. anyways, 1.7-2.0 iBook would be a nice little machine... Smoke an intel chip (if it comes out in time)

[fabsgwu]

Kind of like Intel's "Centrino." Of course don't tell them MHz matters there ;)


I'm not too sure the G5 would be the best chip for portables. Would you rather have a 1.6 or 1.8 Ghz G5 or a dual-core 2GHZ G4 with Rapid I/O and an SOC design? Some stuff I've read indicates that the 2GHZ dual core chip FreeScale is working on runs @ 25W dissipation. To me, it's pretty clear that FreeScale's chips, if they can deliver the e600s in the near future, are a better alternative than the G5. I actually think you can split up the chips over portable and desktop lines:

Apple Desktops: G5
Apple Portables: G4

And the e700 would give Apple a 64bit mobile processor when it comes out. Granted, we'll have to wait and see, but if FreeScale can deliver these things on time (and the 7447A came in ahead of schedule) then Apple will have two viable chip manufacturers.

[ltgator333]

I have to agree with everyone who talked about the 'putting all your eggs in one basket' thing. Having alternatives is ALWAYS a good thing. It had seemed to me that Motorola had announced in the past that they were working on a 64-bit chip but it fell through, I don't know if they are anywhere near getting that back on track or not, but either way some higher clocked 74xx based 32 bit chips with higher FSB's would be excellent for iBooks, eMacs, iMacs, and possibly PBooks until a G5 is crammed into a PBook..

[Mr. MacPhisto]

If Apple bought that load of crap, then they're not half the company they ought to be right now. Motorola took how long to make those last weak jumps? Until Freescale establishes themselves as being past the shortsighted failures of their forebears, they're suspect and complicit in the performance gap that IBM and Apple closed considerably.

If the VX is actually dead, then that's a sad, sad thing to have had happened. IBM is far better than Motorola, in my opinion, and even if there are issues with 90nm, the 750vx would still have been better at 130nm than the G4 was at the same process.


A large part of the shortsightedness of Motorola was due to their old fab. The Crolles2 facility is able to go down to 32nm and is as advanced, if not moreso, as Fishkill. Moto is also teaming with Phillips and STMicroelectronics in France. There is good reason to believe that Moto has turned the corner. Crolles is already producing chips at 90nm and should be pumping them out in volume this summer.

Apple invested millions of dollars in VX - and they killed it off when it's production had to be pushed back. They had to have had some very good reasons for terminating this project, and I don't consider Apple a stupid company.

The VX (it was not a 750) was slightly better than the current G4, although the 7447A could give it a run for its money clock-to-clock, even with its crippled bus.

I can also tell you that IBM considers Motorola to be a very formidable competitor once more and expects some excellent chips from them this summer. It's helping to drive IBM's own chip designs - and could help them also improve.

Plus, there's been talk that FreeScale will deliver something this summer. I'm sure Apple knows precisely what's coming - and has good reason to believe that it will be delivered. I understand the skepticism - but I'm sure we all hope that FreeScale delivers bigtime and Apple has two solid chipmakers to turn to.

[ltgator333]

whoa.. hey what happened to the MPC 7457 chip? I thought that was the newest rev 13 micron G4 variant? From what I heard that chip used less power/ran cooler than the 7447 which was a 18 micron chip..

[Mr. MacPhisto]

whoa.. hey what happened to the MPC 7457 chip? I thought that was the newest rev 13 micron G4 variant? From what I heard that chip used less power/ran cooler than the 7447 which was a 18 micron chip..

No. The 7447 is a 130nm chip. The 7457 is the G4 with L3 cache capability. The 7447 is that chip with fewer pins - basically the embedded version. The 7455 is the 180nm chip that the 7447/57 replaced. I'm not sure if the 7457 was ever produced because of the switch to the G5. Apple isn't using the L3 versions currently.

[ddtlm]

nmk:

I think people going on about the G5 Powerbook are just getting cought up in the hype and not necessarily in touch with reality.
Quite true, but that's not gona stop Apple from rolling out a G5 PB at their first oppurtunity. People think its way faster and Apple will be able to find enough benchmarks to make that impression stick.

Dont Hurt Me:

one more thing and that is it took moto over 1 year to go from 1.42 to 1.5
Yeah that was a 1.42ghz chip in a desktop, under a huge copper heatsink, at the hairy edge of combustion, vs a 1.5 in a laptop.

Mr. MacPhisto:

The VX (it was not a 750) was slightly better than the current G4, although the 7447A could give it a run for its money clock-to-clock, even with its crippled bus.
See thats why I never belived a word about the 750vx. Why on earth would IBM design a fancy new chip to go head to head with Moto in a crowded little market... and then not even have it be significantly superior? People just aren't connecting with the fact that IBM isn't getting rich selling processors to Apple: the 970's are just diluting the R&D of their big PPC's, and the 750's have a huge market other than Apple.

[zync]

Mr. MacPhisto:


See thats why I never belived a word about the 750vx. Why on earth would IBM design a fancy new chip to go head to head with Moto in a crowded little market... and then not even have it be significantly superior? People just aren't connecting with the fact that IBM isn't getting rich selling processors to Apple: the 970's are just diluting the R&D of their big PPC's, and the 750's have a huge market other than Apple.

Yeah, seriously. I don't think many people realize that IBM makes a lot of money on it's enterprise solutions/servers.

[JFreak]

G4 is a good chip, it just suffers from the bottleneck called the front side bus. make it a +400MHz bus and re-introduce the L3 cache, and we'll have pretty good performance.

G5 is a killer because of the fsb throughput. there's nothing magical in the chip itself.

[MikeAtari]

I think that apple needs to diversify their portable market the same way they have, of late, attempted to with their desktops.

They should have at least 4 different types of portables to choose from:

•Desktop Replacement - Not necessarily 1-inch thin, it is allowed to be physically larger and at the same time deliver far more power, as it is for users who probably won't be taking it outside the home or office much, and it could be used to implement technology such as the G5 into the portable line much faster

•Mid-Range System - The size and power of the current PowerBooks is fine for this

•Budget System - The current iBook is fine for this

•Sub-Notebook - Less than 1-inch thin, ultra-portable, speed is not the overriding issue here, but rather size and weight

I think Apple needs to stop putting "all of its eggs in one basket" if you will, not necessarily with the processor makers, so much as their own product line, because as it is right now, it doesn't allow for much flexibility.


What about a Dual Processor G5 Transportable?
- Apple's got the Motherboard built with the XServe
- It's got the 20 display
- It's got the design capability.
Put it all together, and you've got a cool machine.
No need to wait for cooler ( 65nm G5's ) for the portable line.

[Phinius]

A large part of the shortsightedness of Motorola was due to their old fab. The Crolles2 facility is able to go down to 32nm and is as advanced, if not moreso, as Fishkill. Moto is also teaming with Phillips and STMicroelectronics in France. There is good reason to believe that Moto has turned the corner. Crolles is already producing chips at 90nm and should be pumping them out in volume this summer.

Apple invested millions of dollars in VX - and they killed it off when it's production had to be pushed back. They had to have had some very good reasons for terminating this project, and I don't consider Apple a stupid company.

The VX (it was not a 750) was slightly better than the current G4, although the 7447A could give it a run for its money clock-to-clock, even with its crippled bus.

I can also tell you that IBM considers Motorola to be a very formidable competitor once more and expects some excellent chips from them this summer. It's helping to drive IBM's own chip designs - and could help them also improve.

Plus, there's been talk that FreeScale will deliver something this summer. I'm sure Apple knows precisely what's coming - and has good reason to believe that it will be delivered. I understand the skepticism - but I'm sure we all hope that FreeScale delivers bigtime and Apple has two solid chipmakers to turn to.

That's some good informative insight into what Motorola and IBM are doing.

Here are some links to backup what you are saying about Motorola:

A June 2003 interview with the chief technology officer of Motorola's semiconductor products. In the article Motorola's Claudine Simson states in the last paragraph that a 90-nm PowerPC will be in production in 2004.

http://www.siliconstrategies.com/article/printableArticle.jhtml?articleID=10802061


A shorter article which reiterates some information from the above interview:

http://www.siliconstrategies.com/article/printableArticle.jhtml?articleID=10802058


Statements from the director of device/integration at Motorola's MOS-13 wafer fab:

http://www.siliconstrategies.com/article/printableArticle.jhtml?articleID=10802309

In the fourth paragraph he states that Motorola's goal is "a frequency doubling every 18 months or so, and get into the 2-GHz range for PowerPC, but at very low power consumption of say 20 watts"

If you add about 33% frequency improvement to the 1.5GHz G4 from a process shrink at 90-nm, that takes it to about 2GHz. Moving from 180-nm to 130-nm produced about a 33% increase in frequency.


In another article, Motorola's manager of PowerPC products platform division mentions incorporating a memory controller on the PowerPC and why that would be beneficial (the next G4 is expected to have a onboard memory controller):

http://www.siliconstrategies.com/article/printableArticle.jhtml?articleID=10808952

[Mr. MacPhisto]

nmk:

Mr. MacPhisto:


See thats why I never belived a word about the 750vx. Why on earth would IBM design a fancy new chip to go head to head with Moto in a crowded little market... and then not even have it be significantly superior? People just aren't connecting with the fact that IBM isn't getting rich selling processors to Apple: the 970's are just diluting the R&D of their big PPC's, and the 750's have a huge market other than Apple.

IBM wasn't really designing it for their embedded customers. They could have used it, but the chip was being designed for Apple - and Apple was fronting most of the cost for research and development of it. I'm sure IBM would have sold it to others, but they weren't the ones that killed it. Apple killed it because Apple was paying for it.

[ltgator333]

No. The 7447 is a 130nm chip. The 7457 is the G4 with L3 cache capability. The 7447 is that chip with fewer pins - basically the embedded version. The 7455 is the 180nm chip that the 7447/57 replaced. I'm not sure if the 7457 was ever produced because of the switch to the G5. Apple isn't using the L3 versions currently.

Yeah that's really kinda sad.. when looking at the specs between different procs the one glaring *ouch* on 74xx procs is the bus speed.. 167Mhz is terrible- how you combat terrible bus bandwidth is having a lotta L2/L3 cache... and I don't see why they couldn't have (I didn't look really close so I may be eating my words here...) bumped the FSB frequency up to 200Mhz for the new PBooks, it just don't seem to me like when it comes to the current G4 based systems like anyones trying all that hard.

[ddtlm]

JFreak:

G4 is a good chip, it just suffers from the bottleneck called the front side bus. make it a +400MHz bus and re-introduce the L3 cache, and we'll have pretty good performance.
Even when G4's had L3's and a much for favorable core-to-FSB ratio they got their butts kicked by the competition. I maintain that the FSB is less of a problem for most tasks than people think.

G5 is a killer because of the fsb throughput. there's nothing magical in the chip itself.
I'd say your putting way to much wieght on FSB clock speeds. Sure a high bandwidth FSB helps, but the other number people should look at is the memory access latency through a FSB. Although infomation has been spotty, everything I've read agrees with the edjucated guess that the G5's FSB and memory system is pretty high latency, compared to what the G4 is running on. High bandwidth helps "steaming" performance (compression, decoding, etc), low latency helps everything else.

The PC world contains many examples of higher clocked busses not offering much return due to latency problems, for example there have been countless chipsets that offer poor returns when RAM and FSB are run asyncronously. Numerous times a new chipset arrived on the market supporting a new faster FSB or new faster RAM... and failed to impress anyone until version B showed up that ran RAM and FSB at the same clock speed. And lemme tell ya, the G5's FSB is far more asyncronous with RAM than anything the PC world has ever had. Two one-way 32-bit 1ghz busses with a packet-based protolcol and no separate address lines must talk to 128-bit wide 400mhz RAM and a separate address line. And Apple's chipset has to support two of those FSBs in simultaneous operation. Youch! Apple should be dreaming of the day that they can get on-die memory controllers such as AMD has, that would probably unlock a lot of non-streaming performance in the G5 designs we already have.

Edit:
After thinking about it, I think I should be more clear about what I mean be FSB's and RAM at the "same" clock speed. You see, they often arent really the same, for example Vias old KT133 vs KT133A where going to a 266mhz (DDR) FSB from a 200mhz FSB helped add significant performance, when coupled with PC133 RAM. In the KT133A they even removed the ability to run the FSB and RAM at a different "base clock", and it paid dividends. Really what matters is the "base clock", which is often (or always?) the address clock speed. Intel's quad-pumped FSB has the same address clock as DDR at half the data clock, so you'll see that arrangement a lot.

Mr. MacPhisto:

I remain sceptical of the whole thing. :)

[cr2sh]

While my first response went something like this:

"The true g5.... orgasm"

I quickly realized.. so what? It might be news to us, but I'm quite sure Steve knew this was coming. We've had our g5's for a year or so now... let's not look back. If it makes it into the i-family.. then awesome, but this isn't good for much else.

[Mr. MacPhisto]

JFreak:


Even when G4's had L3's and a much for favorable core-to-FSB ratio they got their butts kicked by the competition. I maintain that the FSB is less of a problem for most tasks than people think.

Mr. MacPhisto:

I remain sceptical of the whole thing. :)

I've never seen any evidence of this. I remember some tests of AfterEffects that showed the Dual 1.42 GHZ PowerMacs getting toasted by a 3.2GHZ P4, BUT AfterEffects is not optimized for the Mac at all (G4 or G5) and is well optimized for the Pentium, so it's not a fair comparison at all. I also believe those G4s only had 256K of L2 cache compared to the P4's 512K. Jack up the L2 to equal levels, jack up the bus speed to 400MHZ+ (Moto chips are more efficient with their bus than Intel chips are, btw. MaxBus isn't spectacular anymore, but a 167MHZ FSB on a G4 is far more efficient than the same on a Pentium. Also remember that it's running without clock doubling, quadrupling tech. The Pentiums run on buses that have been quadrupled usually), and put an on-chip memory controller - with the chip running at 2GHZ, and my guess is the G4 will keep up with the new 3.6GHZ Prescott - one of the most inefficient hogs of a chip ever fabbed. Make the G4 a dual-core chip that draws 1/4 the power of the Prescott and you've got yourself a great chip that will run circles around the Pentiums and Centrinos of the world - all the while being more energy efficient and much faster due to SOC design.

Also note that Moto already is working on 65nm, 45nm, and 32nm. Who knows then Intel will get down to 65nm - and thus far there's no indication they can go beyond that. I've talked to enough people so I'm definitely not skeptical. Mark my words, sometime between this summer and fall you will be astonished by what Moto puts out. You also will be floored by the price. Crolles will make their chips much, much cheaper. No longer does Moto have to make due with a dirty fab - they've got one of the world's best fabs, and some of the top engineers in the world working there. It's not just Moto working on these new chips - they're doing it with Phillips and STMicroelectronics. You may not believe it, but with IBM and Moto on board Apple's CPU future looks much brighter than the PC's.

[ddtlm]

Mr. MacPhisto:

Well I'm not gona argue with you about relative performance of computers past, but ...

Make the G4 a dual-core chip that draws 1/4 the power of the Prescott and you've got yourself a great chip that will run circles around the Pentiums and Centrinos of the world - all the while being more energy efficient and much faster due to SOC design.
Forget the Prescott, its clear now that its a failiure. The x86 designs to beat are the Pentium-M and the Athlon-64, against which the G4 offers a much smaller power edge, and doesn't offer a performance edge. While a 2ghz dual core G4 would be nice, by the time such a chip arrives it won't be competing against anything that is currently on the market. It might cause problems for single-core compeditors in certain programs, but it would be up against some much higher clocked 970/980/whatevers and A64's.

Also note that Moto already is working on 65nm, 45nm, and 32nm. Who knows then Intel will get down to 65nm - and thus far there's no indication they can go beyond that.
You can bet Intel will defeat Moto at every transition. Moto hasn't got the money, people, or facilities that Intel has, and AMD is pushing Intel hard, punishing them for every misstep. IBM and AMD are the only companies out there that can be considered compeditors to Intel.

I've talked to enough people so I'm definitely not skeptical. Mark my words, sometime between this summer and fall you will be astonished by what Moto puts out.
Yeah whatever, I've heard it all before. Not even the G5 was truly astonishing, and there's no way that Moto will beat that.

It's not just Moto working on these new chips - they're doing it with Phillips and STMicroelectronics. You may not believe it, but with IBM and Moto on board Apple's CPU future looks much brighter than the PC's.
Damn right I don't believe it. :) Phillips and STMicroelectronics are not exactly known for expertise with high performance CPU production.

[FoxyKaye]

Several readers have noted this PowerPC core roadmap (http://www.motorola.com/mediacenter/news/detail/0,,4126_3482_23,00.html) announced by Freescale Semiconductor, a wholly owned subsidiary of Motorola.

According to the press release, Freescale detailed newly announced e300, e600 and e700 PowerPC cores. The e600 processor is described as "an enhanced version of the high performance G4 core used in the award-winning, high performance MPC74xx family of PowerPC host processors" and is expected to scale beyond 2 GHz as well as implement Altivec. The e700 core is expected to scale beyond 3GHz.

Wind blowing, wolf howling in the distance, tumbleweed rolling by, and crickets chirping in the dead of night. :rolleyes:

If this roadmap were released two years ago, I'd be a little more excited. It's a neat premise that Apple would keep the G4s kicking around in low-end/low-price stuff, but it makes no sense: the future releases of OSX are rumored to be 64-bit only, the G5 chips are getting smaller (presumably not requiring a tornado to keep them cool in the future), and Apple needs to maintain some sort of new edge to its products to keep up with Intel/AMD.

Although the G4 is a solid chip, its time is ending for Apple. I think Motorola will look to move its PPC chips into other consumer products (not computers/laptops) that would benefit from the PPC architecture with some form of embedded OS like Linux. The G4 is a good thing governed by a bad company, and I suspect they might promulgate elswhere now that Apple and Motorola have gone J'Lo and Ben. :D

Something tells me that Steve wants every Apple lineup running G5s by the end of next year (i.e. eMac/iMac/iBooks with 1.6, 1.8, & 2.0 single G5s, PowerBooks with 2.0+ G5s, and PowerMac systems with single/dual 3.0+ G5s). Not to say that IBM hasn't already done its fair share of screw-ups delivering the G5s :mad:, but I think that discussion is being held elsewhere.

[maxterpiece]

When is apple gonna just buy motorola?

[sjk]

Just a comment, every bench i have seen shows a single 1.6 G5 kicks G4 butt ...Where do you see that in the FCP4 benchmark chart (http://forums.macrumors.com/attachment.php?attachmentid=13386) from barefeats that nmk posted in this very thread?

Whoops, I guess he already mentioned that. :)

[Mr. MacPhisto]

Mr. MacPhisto:

Well I'm not gona argue with you about relative performance of computers past, but ...


Forget the Prescott, its clear now that its a failiure. The x86 designs to beat are the Pentium-M and the Athlon-64, against which the G4 offers a much smaller power edge, and doesn't offer a performance edge. While a 2ghz dual core G4 would be nice, by the time such a chip arrives it won't be competing against anything that is currently on the market. It might cause problems for single-core compeditors in certain programs, but it would be up against some much higher clocked 970/980/whatevers and A64's.


You can bet Intel will defeat Moto at every transition. Moto hasn't got the money, people, or facilities that Intel has, and AMD is pushing Intel hard, punishing them for every misstep. IBM and AMD are the only companies out there that can be considered compeditors to Intel.


Yeah whatever, I've heard it all before. Not even the G5 was truly astonishing, and there's no way that Moto will beat that.


Damn right I don't believe it. :) Phillips and STMicroelectronics are not exactly known for expertise with high performance CPU production.

I say bullhockey to most of this dribble. I know engineers in the industry and every single one of them has said that since Crolles has gone online that Motorola and IBM have the best fabs in the industry - Intel is behind both of them. They've managed to gain a whopping 200MHZ in transitioning to 90nm. I'm in awe of Intel's greatness!!!

And before Intel has even moved to 90nm, Motorola was producing chips at 90nm. Moto's fab at Crolles was the second to begin pumping out anything at 90nm, second behind IBM - and only by a few months.

As for the Pentium M, the answer is simple. If you load up a chip with enough L2 cache, it'll rock. I'd like to see how the Pentium M would fair against a G4 with 1MB of L2 cache. The G4 doesn't have to compete with the Athlon64 if it's a mobile chip - the A64 has been crammed into a "laptop" (and I use the term very, very loosely), but it's not a power efficient chip at all.

I also will submit that the 7447A was delivered 3 months ahead of schedule - and that Apple had the redesigns ready because they know everything that is going on. I know Apple has several engineers that work in the Fishkill plant and monitor production, design, etc. I'm positive that Apple does the same with Moto. That's why they want to IBM well in advance of Moto's dropoff and asked for help. They've got people who've been watching them like a hawk for the last year - I'm positive. IBM has their eyes on them too.

It's certainly you're choice not to believe it. It'll make it much better when they deliver and shock you.

[sjk]

It's certainly you're choice not to believe it. It'll make it much better when they deliver and shock you.Regardless of their validity, many of your "speculative" comments and followups have generated some provocative discussion. More worthy of attention and consideration than uninspired, misinformed, unresearched commentary. :)

[thatwendigo]

A large part of the shortsightedness of Motorola was due to their old fab. The Crolles2 facility is able to go down to 32nm and is as advanced, if not moreso, as Fishkill. Moto is also teaming with Phillips and STMicroelectronics in France. There is good reason to believe that Moto has turned the corner. Crolles is already producing chips at 90nm and should be pumping them out in volume this summer.

I'm doing some reading on Crolles2, which I hadn't been aware of before I posted. If nothing else, thanks for bringing it to my attention. The idea that Freescale is operating a 300mm wafer fab and cutting 90nm chips out already gives me pause in my previous assessments. If they really do manage to get an SOI, copper interconnect chip out to replace the G4, and honestly manage to repair the damage done to the line by holding back for so long, then I'll welcome Freescale with open arms.

It won't hurt us to have two processor suppliers, provided both are pushing for advancement in technology. The building of Crolles2 makes me believe that there might be some actual meat to the press release, after all.

Hmmm.

Apple invested millions of dollars in VX - and they killed it off when it's production had to be pushed back. They had to have had some very good reasons for terminating this project, and I don't consider Apple a stupid company.

The VX (it was not a 750) was slightly better than the current G4, although the 7447A could give it a run for its money clock-to-clock, even with its crippled bus.

I'm just curious, since I asked rdowns on another thread... Where are you getting all this from? A lot of what you say can be found in public documents and such, but I've never seen any kind of confirmation that the VX project was anything other than a modified 750, nor have I seen it directly stated that it was being developed solely for Apple.

I can also tell you that IBM considers Motorola to be a very formidable competitor once more and expects some excellent chips from them this summer. It's helping to drive IBM's own chip designs - and could help them also improve.

Motorola wouldn't have been competing the same space as IBM, except in the lower-end embedded market (like the PPC 4xx) before the spinoff. Now? I don't know. You've honestly brought some doubt to my outright distaste for the newer company.

Plus, there's been talk that FreeScale will deliver something this summer. I'm sure Apple knows precisely what's coming - and has good reason to believe that it will be delivered. I understand the skepticism - but I'm sure we all hope that FreeScale delivers bigtime and Apple has two solid chipmakers to turn to.

Before? I'd have probably tried to argue this.

Now? Yeah. I think I'd like to see something solid come out of Freescale, if only to keep the 970 out of the PowerBook.

See thats why I never belived a word about the 750vx. Why on earth would IBM design a fancy new chip to go head to head with Moto in a crowded little market... and then not even have it be significantly superior? People just aren't connecting with the fact that IBM isn't getting rich selling processors to Apple: the 970's are just diluting the R&D of their big PPC's, and the 750's have a huge market other than Apple.

Why do you think that anyone who was arguing in favor of the VX thought it was supposed to compete with Motorola's "crowded market?" If it's even pin-compatible with the G3, then odds are that IBM's own market for the 750s as embedded chips would snap them up, too. I was never arguing that a processor would just be a gift from Big Blue.

In another article, Motorola's manager of PowerPC products platform division mentions incorporating a memory controller on the PowerPC and why that would be beneficial (the next G4 is expected to have a onboard memory controller):

This alone would make me want the revised G4/e600 to be in the PowerBooks and considered for other products. On-die memory control would solve quite a few of the bottlenecks with the G4, especially if it's properly fed through a wider bus.

[ingenious]

I want to see IBM and mortorlla and Intel and AMD work together on the single fastest processor ever just put aside thier differences and aligences and make the first Terrahertz processor for unix. (and since OS X is basically unix with a GUI...)

thats not going to happen anytime soon. besides intel and amd produce one type of processor: x86, while IBM and Moto produce another: PPC. Mac OS X would have to be completely ported and recoded altho darwin would not. :rolleyes:

[thatwendigo]

Here's something a bit interesting. From Freescale's PowerPC core site (http://e-www.motorola.com/webapp/sps/site/overview.jsp?nodeId=02VS0l72156402) we can pull this graphic and see that they're using the 603e and G4 cores as basis for two designs, but bringing two new lines in as well.

http://e-www.motorola.com/files/graphic/other/PPCCORES.jpg

Delivering Higher Performance: The e600 and e700 Cores and Platforms

Taking a step up in performance, customers may leverage the e600 core and corresponding e600 platform. The e600 core is instruction set and pin compatible with the G4 core used in the award-winning, high-performance MPC74xx family of PowerPC processors; however the e600 core is planned to scale beyond 2 GHz and to support Chip Multiprocessing (CMP). Like the G4 core, the superscalar e600 core is designed to issue four instructions per clock cycle (three instructions plus one branch) into eleven independent execution units, and to include a full 128-bit implementation of Freescale's advanced AltiVec Single Instruction Multiple Data (SIMD) vector processing technology.

For a huge performance leap, stay tuned for the next-generation e700 core and corresponding e700 platform. Semiconductor products engineered around Freescale Semiconductor's forthcoming e700 SoC platform are planned to be capable of running both 32-bit and 64-bit software and scaling to 3 GHz and beyond in next-generation process technologies.

[ingenious]

You know, for once, I'm just not going to argue with DHM. Motorola has had plenty of opportunity to scale the G4 by now, and yet they're hanging around at 1.5ghz at the current design. Somehow, I just don't see a 2.0ghz enhanced G4 any time in the near future, and when it comes to the desktop... Well, the G5 is a better system in all respects.

OMGosh! did thatwendigo and DHM just agree???? LOL :P :D!

On a side note, i do not want another G4. Moto had its chance and they majorally blew it! I want the G5 to come screaming down into at least PBs and iMacs! Also, and G5 Cube would be nice! :p

[ingenious]

I think a lot will depend on how free Freescale is from Moto. If Moto keeps screwing things up then IBM has a wide open field. If Freescale can do their own thing (be totally free of Moto) then there is a chance to do some good.

I think that their main challenge, however, will be to try to catch up with IBM's Power 5 chip - which might hit the Mac this summer - and the neat things that will come after that, and come rather fast. It's very difficult to go against a company now producing chips at 90 nm and with the facilities to migrate to 65 nm and then 45 nm.

so would the Power5 be the g6? or the G5 extreme or no apple name change?

[zync]

thats not going to happen anytime soon. besides intel and amd produce one type of processor: x86, while IBM and Moto produce another: PPC. Mac OS X would have to be completely ported and recoded altho darwin would not. :rolleyes:

I already said this however OS X has supposedly been ported already as the XBox developers received copies of OS X that work on their systems and they're developing x86 architecture. I'm unsure of exactly what has been given to whom as I'm not an XBox dev but I'm sure you could search and find info on it.

Oh and the last time I checked IBM also made x86 chips, which is why I said that if anyone got together to do anything, they'd probably end up making an x86 chip.

[Mr. MacPhisto]

I'm doing some reading on Crolles2, which I hadn't been aware of before I posted. If nothing else, thanks for bringing it to my attention. The idea that Freescale is operating a 300mm wafer fab and cutting 90nm chips out already gives me pause in my previous assessments. If they really do manage to get an SOI, copper interconnect chip out to replace the G4, and honestly manage to repair the damage done to the line by holding back for so long, then I'll welcome Freescale with open arms.

It won't hurt us to have two processor suppliers, provided both are pushing for advancement in technology. The building of Crolles2 makes me believe that there might be some actual meat to the press release, after all.

Hmmm.

I'm just curious, since I asked rdowns on another thread... Where are you getting all this from? A lot of what you say can be found in public documents and such, but I've never seen any kind of confirmation that the VX project was anything other than a modified 750, nor have I seen it directly stated that it was being developed solely for Apple.

Motorola wouldn't have been competing the same space as IBM, except in the lower-end embedded market (like the PPC 4xx) before the spinoff. Now? I don't know. You've honestly brought some doubt to my outright distaste for the newer company.

Before? I'd have probably tried to argue this.

Now? Yeah. I think I'd like to see something solid come out of Freescale, if only to keep the 970 out of the PowerBook.


The VX information comes from three friends I have working at IBM who are close to the fab at Fishkill. I know them from when I worked for IBM before I went back to school. The VX was not merely a 750 with Velocity Engine tacked on. It was based on the 750 - that was the starting point. The original idea was that it was going to be available in March for mass distribution @ 90nm (there were discussions of 130nm versions, but it was designed for the 90nm process for its first run). It came about, from what I can piece together, due to Apple's lack of confidence in Motorola over a year ago (this lack of confidence had led to the development of the 970 earlier - a processor that IBM wanted to make in the first place, they just engineered it to work for Apple as well by adding the Velocity Engine). The VX was put together fairly quickly, but lost a bit of steam during the last half of '03 because of Crolles and the future of Moto's semiconductor division becoming less cloudy. It was being developed because Apple and IBM did not know if putting the 970 into a laptop was a great idea (they still don't think it's the most efficient chip to place in a laptop). Apple couldn't have kept up if the 970 was not viable in a laptop - and their consumer end would languish for about two years as they waited for the G5 to filter down to every Mac -- too long when Moto was not to be trusted.

The VX project may still see the light of day on the embedded market - heck, it's even possible Apple could be interested in it later if FreeScale doesn't deliver. Even though it was killed off, it was far enough along that it could be fabbed in a couple of months; they didn't destroy their work. But there is no fabbing scheduled for the design as of the last report I received a couple of weeks ago.

As for the 970FX - it can go into a laptop and may have been planned to if not for the 90nm delays, and the fact that a 1.6GHZ G5 is not too much faster than a 7447A running at 1.5GHZ (there was a FCP4 benchmark posted earlier that shows this - imagine if their bandwith and memory were equal).

I know nothing about Moto/FreeScale outside of what I read online and what I hear from IBM. I know IBM considers them a viable chipmaker once more and takes them seriously. The new fab at Crolles is a big help, but a lot of the people in the industry felt that Motorola's management caused many problems too - and now FreeScale is free of them, for the most part.

I'm another who will believe it when I see it, but I tend to be optimistic. FreeScale delivered the 7447A ahead of schedule. Granted, it's not a spectacular upgrade, but it is the last chip really created under old management.

The big IF is FreeScale delivering, but like you say - if Apple can have both IBM and Motorola giving them good solutions then they're in great shape.

Another thing I want to say is to those who want a firm schedule from FreeScale (this is not aimed directly at thatwendigo or anyone else) - why should they release a timeline publicly? Remember that Apple contributes a significant amount of revenue to them (my guess is a couple hundred million a year in revenue comes from Apple). Apple just introduced new PowerBooks - and PB sales were in a slump before the refresh. If FreeScale announces a few weeks later that there will be 2GHZ G4s with on-board memory controllers, true DDR support, and 1MB L2 cache fabbed at 90nm available in the middle of the summer, Apple's PowerBook sales would drop like a rock because everyone would anticipate an update not long after the introduction - and 2GHZ is a big jump, especially when combined with increased bandwith and an enlarged cache. There's also the problem of getting everyone's hopes up.

I can tell you this, though. FreeScale's future success depends on them delivering exactly what they have promised. If they do, Apple is in spectacular shape.

[ingenious]

I already said this however OS X has supposedly been ported already as the XBox developers received copies of OS X that work on their systems and they're developing x86 architecture. I'm unsure of exactly what has been given to whom as I'm not an XBox dev but I'm sure you could search and find info on it.

Oh and the last time I checked IBM also made x86 chips, which is why I said that if anyone got together to do anything, they'd probably end up making an x86 chip.

i believe that you're confused. m$ supposedly shipped g5s with WINDOWS installed. Windows ported to PPC. Now that's scary.

[zync]

i believe that you're confused. m$ supposedly shipped g5s with WINDOWS installed. Windows ported to PPC. Now that's scary.

That's right, my mistake. I knew that NT was involved in there somewhere. It was an NT version only to be run on G5s methinks, unless I'm wrong again :)

[nek]

If IBM is making a PowerPC version of the Power5 as I'm sure they are, then it will likely be designed as a 90nm part, eliminating any problems in moving from 130nm to 90nm as they are having with the PPC970. I realize that IBM's problems with 90nm likely go beyond this issue, but it seems quite possible that there will be a sudden and large increase in speed in the Powermacs this summer.

This will then allow the iMac to get a G5 a couple months later. Or the iMac and Powermac could get speed increases at the same time since the Powermac will be using a newer 3+GHz 97x and the iMac will get the current 2GHz 970FX. I think the PowerBook will begin using the G5 in january.

If motorola/freescale actually create their e700 within the next year, then the iBook and eMac will likely use that processor. Or they might move to the e600 first, then the e700.

Did anyone else notice that freescale named their processor very similar to intel with the 300, 500, and 700? (Except that they also have the 600.) Thats not likely a coincidence. They are just trying to make their processors seem the same speed.

[Mr. MacPhisto]

If IBM is making a PowerPC version of the Power5 as I'm sure they are, then it will likely be designed as a 90nm part, eliminating any problems in moving from 130nm to 90nm as they are having with the PPC970. I realize that IBM's problems with 90nm likely go beyond this issue, but it seems quite possible that there will be a sudden and large increase in speed in the Powermacs this summer.

This will then allow the iMac to get a G5 a couple months later. Or the iMac and Powermac could get speed increases at the same time since the Powermac will be using a newer 3+GHz 97x and the iMac will get the current 2GHz 970FX. I think the PowerBook will begin using the G5 in january.

If motorola/freescale actually create their e700 within the next year, then the iBook and eMac will likely use that processor. Or they might move to the e600 first, then the e700.

Did anyone else notice that freescale named their processor very similar to intel with the 300, 500, and 700? (Except that they also have the 600.) Thats not likely a coincidence. They are just trying to make their processors seem the same speed.

I think the e700 would be a better chip for the PowerBooks. The SOC design is perfect for laptops, I think better than the Power5 derivative would be.

Also, note that FreeScale is stating the clockrate with the designs. These are model number and chip series, not naming conventions based on speed or use to hide the clockrate.

I think the IBM chips will be used for desktops and the FreeScale chips will find their way into the laptops.

[thatwendigo]

OMGosh! did thatwendigo and DHM just agree???? LOL :P :D!

You want to know the kicker, the one that makes this even more amusing? I was agreeing because of past prejudice, and since have slightly changed my opinion because other people have shown me some new information.

Go figure. :rolleyes:

On a side note, i do not want another G4. Moto had its chance and they majorally blew it! I want the G5 to come screaming down into at least PBs and iMacs! Also, and G5 Cube would be nice! :p

The G5, as things stand right now, is not much faster at the lower clockspeeds that would be necessary to cram it into the smaller formfactors. This is why something like the VX or e600/7000 is going to be a better choice for a while. Most of the advantages of 64-bit computing are denied in a system with limited space for RAM, in any case.

The VX was not merely a 750 with Velocity Engine tacked on. It was based on the 750 - that was the starting point. The original idea was that it was going to be available in March for mass distribution @ 90nm ... Apple couldn't have kept up if the 970 was not viable in a laptop - and their consumer end would languish for about two years as they waited for the G5 to filter down to every Mac -- too long when Moto was not to be trusted.

Interesting stuff.

Do you know what alterations they made to the chip, in general? The Centrino, which I've been using as a comparison in terms of back-stepping to older technology and then adding new features, keeps the Pentium-3 execution core and a lower staged pipeline in order to drive heat down, but has the faster Pentium-4 memory bus and what, by my reading, is a fairly interestingly power managed L2 cache. Fixing up the memory and front-side
buses in the G3, tacking some extra cache on, and tuning the power management sounds like it would have been the way to go with the VX. After all... Apple's been against the megahertz myth for years, so it shouldn't be too hard to sell a chip to professionals that is clocked the same as the curent ones, as long as it can be shown to perform better.

Even though it was killed off, it was far enough along that it could be fabbed in a couple of months; they didn't destroy their work. But there is no fabbing scheduled for the design as of the last report I received a couple of weeks ago.

Good to know it's at least waiting in the wings, if it turns out that Freescale is going to pull the same mistakes and unfortunate prejudices that their predecessor did. I'm of the opinion that Apple's better with one good manufacturer supplying their stuff, but if we get the kind of competitive pushing that the PC world has with Intel/AMD, then that can only benefiet us as consumers.

As for the 970FX - it can go into a laptop and may have been planned to if not for the 90nm delays, and the fact that a 1.6GHZ G5 is not too much faster than a 7447A running at 1.5GHZ (there was a FCP4 benchmark posted earlier that shows this - imagine if their bandwith and memory were equal).

I spent a not-inconsiderable time pointing this out and extrapolating the numbers, actually. I've never been in favor of the G5 in a portable. :D

Another thing I want to say is to those who want a firm schedule from FreeScale (this is not aimed directly at thatwendigo or anyone else) - why should they release a timeline publicly? Remember that Apple contributes a significant amount of revenue to them (my guess is a couple hundred million a year in revenue comes from Apple). Apple just introduced new PowerBooks - and PB sales were in a slump before the refresh. If FreeScale announces a few weeks later that there will be 2GHZ G4s with on-board memory controllers, true DDR support, and 1MB L2 cache fabbed at 90nm available in the middle of the summer, Apple's PowerBook sales would drop like a rock because everyone would anticipate an update not long after the introduction - and 2GHZ is a big jump, especially when combined with increased bandwith and an enlarged cache. There's also the problem of getting everyone's hopes up.

Point conceded. That's actually a damn good set of reasons not to say anything about when they'll be available. I don't think that we'll see major updates until after the summer, at the very earliest. Whatever way they go, Apple's going to have to redesign boards for a chip transition to anything but the MPC 74xx series.

If IBM is making a PowerPC version of the Power5 as I'm sure they are, then it will likely be designed as a 90nm part, eliminating any problems in moving from 130nm to 90nm as they are having with the PPC970.

In a word, no.

At least one of the reported problems with the switch to 90nm is the usage of SSOI and the failure of a layer-bonding agent at operating temperatures in the completed chips. If IBM sticks with the older SOI technology and avoids their tendency to integrate new materials, then you might be right that a Power5 would leap in without issue. At this point, the problem is in fabbing, not the design.

I realize that IBM's problems with 90nm likely go beyond this issue, but it seems quite possible that there will be a sudden and large increase in speed in the Powermacs this summer.

This I agree with. They've been ironing out the issues, and if anyone in the industry has the money to throw at a problem like this, it's IBM. Odds are they're patenting as they go, too.

This will then allow the iMac to get a G5 a couple months later. Or the iMac and Powermac could get speed increases at the same time since the Powermac will be using a newer 3+GHz 97x and the iMac will get the current 2GHz 970FX. I think the PowerBook will begin using the G5 in january.

Hardly. There are issues besides the chip that will need to be solved before the G5 can go into an iMac that's even remotely the formfactor that's currently in use. Maybe it will end up being resolved in a some unforseen way, maybe it's possible that Apple's got something up their sleeves and I'm about to eat my words about the heat being too great when they release a 2.0ghz iMac in the gumdrop white machines.

If motorola/freescale actually create their e700 within the next year, then the iBook and eMac will likely use that processor. Or they might move to the e600 first, then the e700.

The e700 looks quite a bit more appealing than the current crop of 970s, and unless there's some revolutionary addition of power management and bus slewing that cuts the overall system down on usage, the e-series could prove to be a better line to follow with portables. A pin-compatible enhanced G4 that loses the restricting bus, has on-die memory, at least 1MB of L2 cacde, and advanced power control, staging over 2.0ghz, all under 20watts of heat at peak operation? Where do I sign up, since it's unlikely that the 2.0ghz G5s are going to get under 20watts?

[Mr. MacPhisto]

Interesting stuff.

Do you know what alterations they made to the chip, in general? The Centrino, which I've been using as a comparison in terms of back-stepping to older technology and then adding new features, keeps the Pentium-3 execution core and a lower staged pipeline in order to drive heat down, but has the faster Pentium-4 memory bus and what, by my reading, is a fairly interestingly power managed L2 cache. Fixing up the memory and front-side buses in the G3, tacking some extra cache on, and tuning the power management sounds like it would have been the way to go with the VX. After all... Apple's been against the megahertz myth for years, so it shouldn't be too hard to sell a chip to professionals that is clocked the same as the curent ones, as long as it can be shown to perform better.

Good to know it's at least waiting in the wings, if it turns out that Freescale is going to pull the same mistakes and unfortunate prejudices that their predecessor did. I'm of the opinion that Apple's better with one good manufacturer supplying their stuff, but if we get the kind of competitive pushing that the PC world has with Intel/AMD, then that can only benefiet us as consumers.

I spent a not-inconsiderable time pointing this out and extrapolating the numbers, actually. I've never been in favor of the G5 in a portable. :D

Point conceded. That's actually a damn good set of reasons not to say anything about when they'll be available. I don't think that we'll see major updates until after the summer, at the very earliest. Whatever way they go, Apple's going to have to redesign boards for a chip transition to anything but the MPC 74xx series.



The good news for Apple is that a large portion of their market ignore the megahertz numbers. They'd have to if they've purchased Macs the last few years.

As for the VX, I don't know too many specifics. I know it was designed with 1MB L2 cache and could support bus speeds up to 400MHZ with the possibility of stretching it a bit further (not sure how - I know it never has had an on chip controller). I believe they did lengthen the pipeline so it might equal the length of the 74xx series. Most of the stuff I get is not overly detailed on architecture - even friends don't want too give out too much info. Engineers could get canned really quick if they do stuff like that.

I agree about the PowerBooks. I think Apple will stick with their usual update cycles and update the laptops in September or October. I wouldn't be surprised to see them debut in France in September and ship inn October. The two machines that should be updated this summer are the iMacs and PowerMacs, and both of those will likely feature G5s.

It's good for Apple to have options. I think they may have learned their lesson with Moto slumping the past several years. I know they watch everything at IBM like a hawk; they've got a vested interest in what IBM is fabbing at Fishkill.

I also am not a proponent of the G5. I think Joswiak's comments about it being a while before they come out is primarily a stall tactic. Apple may put it in a laptop if they're forced to, but I don't think they want to. They've been involved in designing an SOC portable chip with IBM (also to be used for IBM's embedded market), so there's definitely a desire to go in that direction. That design won't be ready for another year. If FreeScale can deliver nowthen I'm sure Apple will be happy as a clam. As for the IBM SOC chip, I don't know anything about it specifically. I just know they're working on it.

[mbs]

I'm just wondering when Mot will apply the gallium arsenide on silicon technology to their CPUs - they have several patents for that. GaAs is known to switch much faster than silicon and should allow clock frequencies way beyond 3 GHz.

e.g. http://v3.espacenet.com/textdoc?DB=EPODOC&IDX=US2003034508

[ewoh24]

I'm having a debate with a co-worker. He says that the FSB of the G5 hardly makes a difference when it comes to video rendering and that RAM is the biggest factor when it comes to how many apps you can open while doing CPU intensive tasks like that. I agree with him that RAM is super handy as far as open apps are concerned, but add that the large FSB of the G5 allows all those open apps to be responsive at the same time because of the amount of bandwidth available. He disagrees.

Who is right?

[Hattig]

I'm having a debate with a co-worker. He says that the FSB of the G5 hardly makes a difference when it comes to video rendering and that RAM is the biggest factor when it comes to how many apps you can open while doing CPU intensive tasks like that. I agree with him that RAM is super handy as far as open apps are concerned, but add that the large FSB of the G5 allows all those open apps to be responsive at the same time because of the amount of bandwidth available. He disagrees.

Who is right?

You'll want a lot of RAM, to avoid swapping which is bad.

Beyond that, good OS design is important for reponsiveness - Windows doesn't have this even with gob****e loads of RAM and uber-fast processors because of poor GUI systems.

For application speed, processor speed is foremost. If the application is processing data (heh, most will be) then fast, low latency memory is good to have. For example, the G4 might have 3 Altivec units, but the FSB bandwidth is only good enough to feed one of them at full rate (IIRC, this could have been at 700MHz as well). Luckily a lot of problems can be contained within the processor's caches.

Low-latency memory is handy for applications that don't stream memory as access it on a more random basis. High bandwidth memory is good for streaming applications. It is little wonder that Intel concentrates on bandwidth and media applications for its benchmarks, because the P4 just happens to be good at these.

The Opteron/A64 has the best design - low latency memory at high bandwidth, and no FSB to worry about as it is all integrated on the same chip.

I think the best (quickest) way forward for the G4 would be to put a memory controller on-chip, and leave the FSB as is, or up to 200MHz. the 745x series are almost there, with an L3 controller. 1.6GB/s FSB would be enough for laptop quality AGP and I/O controllers.

[ddtlm]

Mr. MacPhisto:

I say bullhockey to most of this dribble.
Dribble? I dont care what engineers you know, Intel's dropping the ball on 90nm doesnt make Moto better than them. Hell, its easy to critise from the sidelines at 130nm.

Intel is behind both of them. They've managed to gain a whopping 200MHZ in transitioning to 90nm.
Any increase beats IBM's 0mhz increase (so far), and long shipping delay. Anyway, it remains to be seen if the Prescott design or the process is the problem. Intel insists their process is fine and yielding well, people can overclock Prescotts a lot if they have monster cooling systems.

And before Intel has even moved to 90nm, Motorola was producing chips at 90nm. Moto's fab at Crolles was the second to begin pumping out anything at 90nm, second behind IBM - and only by a few months.
Wake me up when they make a 90nm processor.

As for the Pentium M, the answer is simple. If you load up a chip with enough L2 cache, it'll rock.
Like the G4's with 2mb L3's? Your making 1024k L2 sound huge, its only twice what Moto has on their G4's, its what IBM is putting on the next 750. 10% performance boost maybe.

I also will submit that the 7447A was delivered 3 months ahead of schedule - and that Apple had the redesigns ready because they know everything that is going on.
Heh, so a good year behind the market is "ahead of schedule"?. Nice schedule. Moto was embarrassingly slow getting a 130nm G4 out.

The VX was not merely a 750 with Velocity Engine tacked on. It was based on the 750 - that was the starting point. The original idea was that it was going to be available in March for mass distribution @ 90nm
I have a pretty hard time believing anyone would go for this, by the time this was supposed to show up the 90nm G5 should have been available, seems way way cheaper to have just planned on using that in as many places as possible while limping along with G3's in the iBooks and old-school 180nm G4's in iMacs/eMacs.

If FreeScale announces a few weeks later that there will be 2GHZ G4s with on-board memory controllers, true DDR support, and 1MB L2 cache fabbed at 90nm available in the middle of the summer
Oh so I see, you figure a G4 wrapped with modern L2 and memory is gona be a shock. Didn't they have something like this on the roadmap for a long time, something to do with "rapidIO"? Well I'll be impressed if they get this thing fabbed on 90nm without the delays and problems most other people seem to be having.

I believe they did lengthen the pipeline so it might equal the length of the 74xx series.
Well then it sounds like a pretty major, expensive rework then. Less believable still.

thatwendigo:

Why do you think that anyone who was arguing in favor of the VX thought it was supposed to compete with Motorola's "crowded market?" If it's even pin-compatible with the G3, then odds are that IBM's own market for the 750s as embedded chips would snap them up, too. I was never arguing that a processor would just be a gift from Big Blue.
The 400mhz DDR FSB rules out pin compatibility, and I found the performance estimates pretty far out. Other than that, yea, I could see a vector unit once they are fabbing at 90nm when its real cheap to add features like that. So far though, they seem more interested in just tacking on L2.

[thatwendigo]

Dribble? I dont care what engineers you know, Intel's dropping the ball on 90nm doesnt make Moto better than them. Hell, its easy to critise from the sidelines at 130nm.

Crolles2 Foundry produces first 90nm samples in 2002 (http://www.semiconductors.philips.com/acrobat/various/final_crolles_overview.pdf)


Any increase beats IBM's 0mhz increase (so far), and long shipping delay. Anyway, it remains to be seen if the Prescott design or the process is the problem. Intel insists their process is fine and yielding well, people can overclock Prescotts a lot if they have monster cooling systems.

Centrino 1.7ghz @25-30w roughly equals P4 2.8-3.0ghz.
P4 90nm 3.4ghz @90-105w roughly equals P4 130nm 3.4 @75-80w

Looks like a design flaw to me.

Like the G4's with 2mb L3's? Your making 1024k L2 sound huge, its only twice what Moto has on their G4's, its what IBM is putting on the next 750. 10% performance boost maybe.

It's part of what makes the Celeron so much faster than the P-III.

I have a pretty hard time believing anyone would go for this, by the time this was supposed to show up the 90nm G5 should have been available, seems way way cheaper to have just planned on using that in as many places as possible while limping along with G3's in the iBooks and old-school 180nm G4's in iMacs/eMacs.

The VX != a mere 750.

Like the Centrino, which I have repeatedly compared it to, it was a redeisgn that used a more efficient core design and new factors from later processors. The "P4-M Centrino" is the latest label for the P-III Mobile core with a P-4 memory bus and a higher cache. Compare it to the other laptop PC processors that get anything more than an hour of battery life while also providing performance roughly on par with a 2.8-3.0ghz desktop P-4.

It's pretty impressive, isn't it?

Oh so I see, you figure a G4 wrapped with modern L2 and memory is gona be a shock. Didn't they have something like this on the roadmap for a long time, something to do with "rapidIO"? Well I'll be impressed if they get this thing fabbed on 90nm without the delays and problems most other people seem to be having.

I'm skeptical, too, but if this is real and coming down the pipe, then it can only be good for Apple. As was shown in some previous posts, the G4 can still hold its own in some dual systems, and a comparatvely clocked system with no improvement in performance per-clock would come damn close to the G5 in specs, but at lower heat.

The 400mhz DDR FSB rules out pin compatibility, and I found the performance estimates pretty far out. Other than that, yea, I could see a vector unit once they are fabbing at 90nm when its real cheap to add features like that. So far though, they seem more interested in just tacking on L2.

The e600 is pin-compatible with the G4 and backwards compatible with the processor's instruction set. Whether or not that means the FSB is ruled out is something I'm not fully qualified to speculate on, but I'd love to see your justification for that claim.

Also, the e600 (G4e-based) and e700 (new design) will both have 128-bit dual-precision SIMD (AltiVec) engines. The e500 is already being fabbed and supplied for embedded purposes, and that's being sold as the PowerQUIC II.

[ClimbingTheLog]

Who cares??!?!??!?! Bring on the new Studio Displays dammit!!!!!!!

Why do we care again about the Apple displays? Yeah, they're pretty. A few years ago Apple put out the best displays in terms of pixel density and viewing angle, but the market caught up and how much beyond 180 degrees do we need? ADC is interesting but not revolutionary.

I pay Apple's premiums for their computers as they bring alot to bear to the equation, but with displays? - they got out of the printer market because they were only equal with the competition, when will they do the same with displays? Pay Samsung something to make an Aluminum Ives-Edition frame and be done with it.

[ClimbingTheLog]

In the article Motorola's Claudine Simson states in the last paragraph that a 90-nm PowerPC will be in production in 2004.

Look, I sat through a Motorola NDA briefing in the summer of 1998 about their processor roadmap. They guaranteed quad-core G4's running at 900MHz (I think, maybe 800) by the 1st quarter of '01.

Need I say more?

[ClimbingTheLog]

I'm just wondering when Mot will apply the gallium arsenide on silicon technology to their CPU

GaAs tends to be much more expensive to produce. This is the stuff Cray's chips were made of.

[ddtlm]

thatwendigo:

I not concerned with samples, I care about products. I'm not aware of a 90nm Moto processor, perhaps there is one on the market. Someone should tell me about it. Unless Moto has a 90nm chip out, and a fairly complex one at that, then I don't think they can be viewed as being ahead of Intel in the tech game. Like I was saying, its easy to critisize Intels 90nm progress from the sidelines.

Centrino 1.7ghz @25-30w roughly equals P4 2.8-3.0ghz.
P4 90nm 3.4ghz @90-105w roughly equals P4 130nm 3.4 @75-80w
Looks like a design flaw to me.
Hey I'm not gona defend the Prescott. I'm just saying this isn't proof that Intel botched 90nm. Wait to see a 90nm Pentium M, then we can look at process and design separately.

It's part of what makes the Celeron so much faster than the P-III.
Well clearly cutting cache below a certain point is a severe problem. Adding "extra" cache is totally different. Going to 2MB of on-die cache (from 512k) didn't keep the P4EE ahead of A64.

The VX != a mere 750. Like the Centrino, which I have repeatedly compared it to, it was a redeisgn that used a more efficient core design and new factors from later processors.
Precisely why I don't believe it ever existed. Why would IBM work so hard for a design which, apparently, is primarily going to go into low-margin Macs? It could go into the embedded market but I'm thinking that most all IBM's embedded customers that want features like AltiVec are already running Moto chips. IBM would have a fight on its hands, no easy money there.

The e600 is pin-compatible with the G4 and backwards compatible with the processor's instruction set. Whether or not that means the FSB is ruled out is something I'm not fully qualified to speculate on, but I'd love to see your justification for that claim.
Isn't the e600 a G4? Perhaps I'm confused. But in any case, I'm not aware of any processor which has significantly redesigned its FSB without changing pinouts. Even if no new pins need to be added, the DDR signalling isn't gona work with an old chipset, so what's the point of being pin-compatible? Pin-compatible is all about drop-in replacement, such as 7455 to 7457.

Also, the e600 (G4e-based) and e700 (new design) will both have 128-bit dual-precision SIMD (AltiVec) engines.
Really? Dual-presion SIMD floats I could get excited about. However I've read informative articles that suggests it would be hard to do cause the execution units would be huge.

[Dont Hurt Me]

Look, I sat through a Motorola NDA briefing in the summer of 1998 about their processor roadmap. They guaranteed quad-core G4's running at 900MHz (I think, maybe 800) by the 1st quarter of '01.

Need I say more?Not a word, those of us long time Mac users know all about Moto's so called fictional dreams and road maps, we know all about the year at 500 or 450 rather and we know about the year at 1.42 or should i say 1.33. enough talk, show me a chip that can hang with AMD's Fx53 or even a stale yet still much faster P4. so Moto has a chip that equals a 2.0 P4 from 2 years ago. BFD! Moto is still Last place when you look at Intel,AMD & IBM. Still amazed Apple went with this looser. Biggest reason for Apples lost marketshare is this slow cpu and Apples piss pot poor marketing.

[sjk]

There are issues besides the chip that will need to be solved before the G5 can go into an iMac that's even remotely the formfactor that's currently in use. Maybe it will end up being resolved in a some unforseen way, maybe it's possible that Apple's got something up their sleeves and I'm about to eat my words about the heat being too great when they release a 2.0ghz iMac in the gumdrop white machines.Seems clear Apple's got something up their sleeves relative to the current iMac. Otherwise wouldn't you think they'd have gotten a refresh update by now since all other G4 systems recently got one? Waiting until WWDC now in light of that would be a disappointment, which the current models can't afford. Regardless of what happens, it sure looks to me like WWDC will be a pivotal time for the destiny of the iMac. Or I'll eat my words when it gets a refresh update next Tuesday. :)

[Phinius]

Look, I sat through a Motorola NDA briefing in the summer of 1998 about their processor roadmap. They guaranteed quad-core G4's running at 900MHz (I think, maybe 800) by the 1st quarter of '01.

Need I say more?

Your stating you sat in on a Motorola briefing that gave a talk about a roadmap two and one half years into the future. My links were to statements about chip designs that were already being tested in 2003 and were scheduled to go into production sometime this year.

In order for Motorola or Freescale to continue producing PowerPC processors that can be used in the embedded market (or even Apple's computer market), advancements must be made to the chips performance. At some point design changes must be made in order for the chip to stay even remotely competitive. Furthermore, the G4 has had at least yearly updates in performance since August of 1999. It just that many Mac users are disappointed that it's performance has not kept up with Intel's speed advances. Seeing how Apple's marketshare has been sliding since a peak of 12% in 1992, it seems inevitable that at some point Motorola and or IBM would find it difficult to keep up with Intel in performance processors used in personal computers.

To believe that the G4 or a derivitive will never make it to a 90-nm process in the near future is ignoring the facts that have happened in the last year and a half. First, Motorola (or Freescale) now has two major chip manufacturing partners and use of a state-of-the-art chip making in France. Second, Motorola's chip manufacturing division is now making money after going several quarters of heavy losses and cutbacks. Third, the G4 has advanced about as far as it can realistically go in frequency using a 166MHz bus and a a 130-nm process. It quite simply would need a change in bandwidth in order to move the core up to 2 GHz on a 90-nm process, and to stay competitive in the embedded market.

Motorola or Freescales projections of a 3GHz+ processor are also a natural progression for the G4. Motorola has stated numerous times that they intended to keep the average power use at a peak of 20 watts. In the last two process shrinks the top average power use of the G4 has moved below 20 watts. Moving to a 65-nm process would shrink the power use enough to add more pipeline stages and advance the frequency to 3GHz. This has happened once before with the move from 4 pipeline stages to it's current 7 pipeline stages. I'd expect to see about 10 pipeline stages for the e700 at the 65-nm process level. Seeing how Freescale has chip manufacturing partners, I'd expect the move to a 65-nm process to occur as early as the beginning of 2006 or late 2005.

[thatwendigo]

Look, I sat through a Motorola NDA briefing in the summer of 1998 about their processor roadmap. They guaranteed quad-core G4's running at 900MHz (I think, maybe 800) by the 1st quarter of '01.

Need I say more?

As much as I'm hoping we don't get burnt again, Freescale is already moving faster than Motorola was on the PowerPC arcitecture. Their new e500 core is on the line right now, runs ramping up for the embedded market.

On top of that, a huge portion of the Crolles2 budget is into research into alternative materials, process shrinks, and alternatives to traditional COMD techniques. This isn't just Motorol doing this... Phillips and ST are heavily invested in the line, and AMD, Sony, and others are buying into the PowerPC's newer, more open standards as well.

If nothing else, we ought to see a broadening of the technology over the next few years. That, and all these licensees are paying into IBM's coffers. They do own PowerPC, after all.

GaAs tends to be much more expensive to produce. This is the stuff Cray's chips were made of.

As I mentioned above, I've been doing some research into the purposes of the fab, the R&D space, and their outlined projects for the next couple of years. Depending on just how much success they have, Motorola might start creating a little IP of their own on the PowerPC, something that they don't necessarily need to license from IBM. It may not directly help Apple, but the platform needs to expand in general if there's going to be the kind of economy of scale that would start to favor us.

I not concerned with samples, I care about products. I'm not aware of a 90nm Moto processor, perhaps there is one on the market. Someone should tell me about it. Unless Moto has a 90nm chip out, and a fairly complex one at that, then I don't think they can be viewed as being ahead of Intel in the tech game. Like I was saying, its easy to critisize Intels 90nm progress from the sidelines.

There is no Motorola chip actively being sold at 90nm that I am aware of. Granted, I'm new to seiving their site for information and they seem to be in a transitional phase. However, their entire top-end line is set to be transferred down to 90nm, with a whole wing of the fab dedicated to the 300mm wafer 90nm process.

Hey I'm not gona defend the Prescott. I'm just saying this isn't proof that Intel botched 90nm. Wait to see a 90nm Pentium M, then we can look at process and design separately.

Dothan. 10th of May, unless they slip the deliver date.

Well clearly cutting cache below a certain point is a severe problem. Adding "extra" cache is totally different. Going to 2MB of on-die cache (from 512k) didn't keep the P4EE ahead of A64.

That's because the Athlon 64 FX series (which is the only Athlon 64 to beat the P4EE, as of the last time I looked) uses a lowered version of the Opteron's memory controller, which AMD licensed from Cray. It's got a far better FSB and memory fabric than the Pentium ever has. Also, the Athlon 64 is a newer chip, not just a further ramping of the same execution cores and pipelines that were introduced three or four years ago.

Precisely why I don't believe it ever existed. Why would IBM work so hard for a design which, apparently, is primarily going to go into low-margin Macs? It could go into the embedded market but I'm thinking that most all IBM's embedded customers that want features like AltiVec are already running Moto chips. IBM would have a fight on its hands, no easy money there.

The G3 is still used in embedded designs, and if IBM could offer their G3 customers extra value and function on the processor they've been using, it wouldn't hurt them. Having something competitive with the G4 would give them a chance at Motorola's market, and unless I'm vastly mistakenm IBM's got a much bigger warchest than Motorola. They could weather the storm much better than even Intel could.

Also, unless I'm mistaken again, IBM is the only manufacturer of 750s.

Isn't the e600 a G4? Perhaps I'm confused. But in any case, I'm not aware of any processor which has significantly redesigned its FSB without changing pinouts. Even if no new pins need to be added, the DDR signalling isn't gona work with an old chipset, so what's the point of being pin-compatible? Pin-compatible is all about drop-in replacement, such as 7455 to 7457.

I linked to the Freescale PowerPC core site earlier in this post, but just to save you time, I'll confirm that their documents say the e6000 is a "G4e," or MPC74xx series derivative. I find it likely that if they're making it as backwards compatible as the site claims, the memory bus will run in older boards as well, just not with the advantages of the new bus or memory clock.

Even without them it's got a better SIMD engine and a higher clock.

Really? Dual-presion SIMD floats I could get excited about. However I've read informative articles that suggests it would be hard to do cause the execution units would be huge.

From the horse's mouth: (http://e-www.motorola.com/webapp/sps/site/overview.jsp?nodeId=02VS0l72156402)
The e600 core is instruction set and pin compatible with the G4 core used in the award-winning, high-performance MPC74xx family of PowerPC processors; however the e600 core is planned to scale beyond 2 GHz and to support Chip Multiprocessing (CMP). Like the G4 core, the superscalar e600 core is designed to issue four instructions per clock cycle (three instructions plus one branch) into eleven independent execution units, and to include a full 128-bit implementation of Freescale's advanced AltiVec Single Instruction Multiple Data (SIMD) vector processing technology.

SoC Implementation to include new buses and architectures: (http://e-www.motorola.com/webapp/sps/site/overview.jsp?nodeId=02VS0l72152529)
In addition to leveraging enhanced PowerPC™ processor cores optimized for SoC design methodologies, Freescale's scalable SoC platforms draw from the company's large and growing portfolio of intellectual property (IP). This broad portfolio includes system fabrics (RapidIO™, SerDes), network acceleration (10/100/100 Ethernet, ATM, HDLC, etc.), external buses (PCI, PCI-X, etc.), memory controllers (DDR and DDRII), general communications peripherals, and security engines. Access to Freescale's IP portfolio makes it fast, easy and cost-effective to mix and match functional blocks and develop new SoC-based products optimized for a wide range of applications.

Incidentally, the e500 embedded processor has a cool feature. It does on-chip encryption enhancements for industry standards like MD-5, RC5, and other that are listed in the specs.

Not a word, those of us long time Mac users know all about Moto's so called fictional dreams and road maps, we know all about the year at 500 or 450 rather and we know about the year at 1.42 or should i say 1.33.

Yes, we knows about a company under different management and entirely under Motorola's heel. Supposedly, Freescale is completely reorganized. It might as well be another company with the same IP and licensing rights.

I'm not saying that they're going to deliver on the promises they're making, but there might be reason to at least take them more seriously. The first of the new core designs is done.

enough talk, show me a chip that can hang with AMD's Fx53 or even a stale yet still much faster P4. so Moto has a chip that equals a 2.0 P4 from 2 years ago. BFD! Moto is still Last place when you look at Intel,AMD & IBM.

Depends on the market. Embedded? Intel and Motorola are the big players, not AMD or IBM, though IBM does sell a large number of G3s in this market.

For laptop performance? Intel all the way right now, short of a major revolution from IBM or Motorola in the next few months. In a week, the Dothan 90nm Centrino is supposed to be droppping in at 2.0ghz. Considering that the 1.7 performs like a 3.0 ghz, the 2.0ghs could very well be faster than the top of the line P4 Northwoods, let alone the Prescotts. Scaling linearly, the 2.0 would reach a theoretical equivalent of a 3.5ghz P4, and since the Athlon FX and P4EE are neck and neck at the moment... It's interesting all over, right now.

Still amazed Apple went with this looser. Biggest reason for Apples lost marketshare is this slow cpu and Apples piss pot poor marketing.

I'm sure it has nothing to do with the hundreds of PC component manufacturers and OEMs that compete on price. :rolleyes:

[thatwendigo]

Seems clear Apple's got something up their sleeves relative to the current iMac. Otherwise wouldn't you think they'd have gotten a refresh update by now since all other G4 systems recently got one? Waiting until WWDC now in light of that would be a disappointment, which the current models can't afford. Regardless of what happens, it sure looks to me like WWDC will be a pivotal time for the destiny of the iMac. Or I'll eat my words when it gets a refresh update next Tuesday. :)

I'm predicting the death of the iMac as we've known it, to be replaced by an unannounced product. The iMac concept has run for seven years and it's getting a little long in the tooth.

Barring that, we'll probably see a major revision of the machines at WWDC. My guess is that they're either going to be dual G4s or single G5s at around 1.6-1.8ghz, though either one will probably require a new form factor.

Oh, and nicely done, Phinius. I still retain my skepticism, but let's say that I'm open to possibilities at this point.

[Phinius]

Well clearly cutting cache below a certain point is a severe problem. Adding "extra" cache is totally different. Going to 2MB of on-die cache (from 512k) didn't keep the P4EE ahead of A64.

Both IBM and FreeScale are probably going to go for adding more processors on a chip for their respective 970 and G4 designs, rather than adding much more cache as Intel is doing with the Pentium 4 and Pentium-M. Motorola had already announced plans last year to make a dual-core G4 that tops out at 2GHz+ (presumably at the 90-nm level) and IBM could very well put 3-9XX processors on one chip at the 65-nm process size. A rough calculation puts a 3 core 9XX chip at about the same die size and watt use as the 970. Intel would be hard pressed to put three Pentium 4 processors on one reasonably sized chip and the Pentium-M is expected to move to dual-core in 2006.

Isn't the e600 a G4? Perhaps I'm confused. But in any case, I'm not aware of any processor which has significantly redesigned its FSB without changing pinouts. Even if no new pins need to be added, the DDR signalling isn't gona work with an old chipset, so what's the point of being pin-compatible? Pin-compatible is all about drop-in replacement, such as 7455 to 7457.

The e600 will not need a Northbridge bus, as it will have a onboard memory controller. FreeScale does state that the e600 will be pin compatable with the 74XX series. The e600 is just another name for a continuation of Motorola's 74XX series of chips. The e700 series will evidently have much more extensive changes to the architecture.

Really? Dual-presion SIMD floats I could get excited about. However I've read informative articles that suggests it would be hard to do cause the execution units would be huge.

The G4 has four independent 128-bit single precision Altivec units (simple, complex, floating, permute). Adding RapidIO and a onboard DDR/DDRII memory controller, by way of the e600, should move the G4s floating point performance up considerably.

[sjk]

That's a well formulated and sensible prediction, thatwendigo.

Not to belabor this too much, but the eMac/iMac legacy has become somewhat of a sour note in Apple's product offerings. Just today I read something about that starting in the third paragraph of this Bill Palmer article (http://www.billpalmer.net/com000188.html). His claim is basically that Apple was unable to carry out its original plan to completely replace the original iMac, thus the eMac was born. And that jives with Steve Jobs' claim around that time that the CRT was dead. Whoops.

So, it'll be interesting if an iMac replacement is announced and with it the overdue retirement of the "misplaced" iMac name (I hope; no sentimentality here).

Wish I could make it to my first WWDC this year since it's looking like an emphasis will be on desktop hardware announcements and that's what I've been waiting to get for over a year.

[Mr. MacPhisto]

I'm predicting the death of the iMac as we've known it, to be replaced by an unannounced product. The iMac concept has run for seven years and it's getting a little long in the tooth.

Barring that, we'll probably see a major revision of the machines at WWDC. My guess is that they're either going to be dual G4s or single G5s at around 1.6-1.8ghz, though either one will probably require a new form factor.

Oh, and nicely done, Phinius. I still retain my skepticism, but let's say that I'm open to possibilities at this point.

I've got to admit that I love ideas circulating around concerning a modular design that would allow a new Mac to be AIO - but would also allow it to be a stand alone unit. I agree that the iMac has reached the end of the line. The original G3 iMac was a brilliant design and necessary to the Mac's rejuvenation. The FP iMac was even more brilliant, but I really think the AIO has it the wall (although there may be a market for another LCD AIO to replace the eMac to the education market).

I'm optimistic for the future, though I feel that Apple needs to be able to offer better options on their low-end. And there's a need for better video cards across the lineup. I realize this is not Apple's fault, but they should be more proactive in getting essential hardware developed for the platform.

[mbs]

GaAs tends to be much more expensive to produce. This is the stuff Cray's chips were made of.

Yes, but now one can deposit it on silicon - you need some titanate as an interlayer to remove the strain because silicon and GaAs have different cell constants - so you need less material.

[zync]

Why do we care again about the Apple displays? Yeah, they're pretty. A few years ago Apple put out the best displays in terms of pixel density and viewing angle, but the market caught up and how much beyond 180 degrees do we need? ADC is interesting but not revolutionary.

I pay Apple's premiums for their computers as they bring alot to bear to the equation, but with displays? - they got out of the printer market because they were only equal with the competition, when will they do the same with displays? Pay Samsung something to make an Aluminum Ives-Edition frame and be done with it.

Sorry this is just a small thing. You can't have anything beyond a 180º viewing angle because the other side would have to be pure LCD as well :) You could get it to work but you'd have to work pretty hard on lighting it. Seriously though, the highest potential viewing area for an LCD display (unless it's like an LCD window) cannot be more that 180º. 180º goes from viewing parallel to the display to viewing it from the opposite side along the same parallel.

[HiRez]

1. How much heat do these things dissipate?

2. What is the current and maximum FSB bus speed?

3. How efficient are they at MP?

4. How does the performance compare with a G4 or G5 clock-for-clock?

5. How much do they cost?

[anjaki]

This whole thread has been the most interesing read I've had for months, thanks to everyone who contributed. It's certainly given me a lot to think about.

P.S. Zync, I think ClimbingTheLog was being ironic about the 180º+ viewing angle.

[sushi]

Nice to know that the Powerbook is a consumer product. You should tell Steve that, 'cause he doesnt know it!
Suggest that you read what he said.

Here is it again:

Originally Posted by rastalin94
Actaully keeping the G4 line alive and increasing the speed could help Apple keep the consumer and pro line different.
Pro Line = G5
Consumer Line - G4
a 2GHz iBook would be a very nice little machine.

He is saying the G4 would be in the Consumer Line, which is the iBook!

The PowerBook is the Pro Line, which would be a G5 in his argument.

Sushi

[sushi]

Nice to know people love to post things without any regard to what was actually written.

He said a 2GHz iBook would be nice. An iBook is not a PowerBook.
rdowns, you beat me to it. Good catch!

Sushi

[sushi]

But if you're differentiating pro and consumer lines based on the processor, PowerBooks are "consumer" machines.

I don't think he was referring to the iBook part.
You are getting the posters mixed up.

Go back to the beginning of the thread and read them again.

One person is saying:

Pro Line G5
Consumer Line G4

Another is saying:

Desktops G5
Portables G4

It helps to read the posts before posting. We have two ideas going in parallel here.

IMHO, of these two concepts, I like the first one the best (Pro=G5, Con=G4). It keeps it simple and is easy to understand/differentiate.

Sushi

[sushi]

Does any of this really matter? :D
Probably not! :eek:

Sushi

[sushi]

Well, this should pan out in 10 or 12 years..
Being a little optomistic are we?! :D

Sushi

[Phinius]

1. How much heat do these things dissipate?

Last year Motorola stated that the dual-core G4 chip would dissipate 25 watts at 1.5 GHz. That's at least 25% more than a 1.5GHz 7447A G4 uses now. Whether Apple would use it in a notebook computer is debatable. A 2GHz dual-core should use about 35 watts, by my rough estimate. That would make it's power use high for a notebook computer, but it also dispenses with a Northbridge memory controller (it will be on the processor) which uses some watts. The main memory will run at a faster MHz and probably use more watts, but if Apple uses DDR-II memory then that would reduce the memory power use somewhat.


2. What is the current and maximum FSB bus speed?

The G4 FSB currently tops out at 166MHz, but the 7457 has the capability of using up to 2MB of L3 cache. Since L3 cache is accessable to the processor much faster than going to main memory, then the use of a L3 cache probably makes up for the slow bus speed. But since most people don't seem to be aware of the heirarchy of cache to main memory, then that could be a big reason why Apple dispensed with the added expense of using it in any Mac computers.


3. How efficient are they at MP?

Don't know at this point, but the second G4 processor on the PowerMacs added up to a 50% performance boost on some applications that made use of a second processor.


4. How does the performance compare with a G4 or G5 clock-for-clock?

Can't say for sure without seeing the chip in action. Moving the memory controller onboard will reduce latency due to shortening the distance to main memory. Also there is the RapidIO chip interconnect, which has similar characteristics to Hypertransport used in Opteron. On some applications a dual-core e600 (G4) should be at least 50% faster clock-for-clock compared to the current G4s and probably faster than the G5. The question is cost and power use compared to the G5. The G5 uses a more costly and complex memory subsystem than this dual-core e600 would need. The G5 uses a northbridge memory controller and dual channel memory. The e600 should have a onboard memory controller and use single channel memory which will also reduce power use compared to the G5 memory subsystem.

Aceshardware compared the dual 1.25GHz G4 PowerMac to Athlon and Pentium 4 powered computers. The authors conclusion was that the 1.25GHz G4 was comparable in speed to at least a 1.8GHz Pentium 4. So if you add a boost of 60% to bring the G4 to 2GHz and add a 50% speed increase with the use of a second processor onboard, then that brings it up to about the performance of a 4.3GHz Pentium 4 without Hyperthreading. With two processors sitting side-by-side on the chip and using RapidIO, along with a onboard controller, it could very well be that a 2GHz dual-core e600 G4 could match the performance of a 4.3GHz Pentium 4 on at least some applications.

I'd say that the current G4 has comparable performance to the Pentium III clock-for-clock. The Pentium-M is about 50% faster than the Pentium III clock-for-clock on SPECint2000 and about 70-80% faster on SPECfp2000. So adding another processor and faster subsystem should bring the G4 performance more in line to what the Pentium-M is capable of.


5. How much do they cost?

Judging from the almost 50% die shrink that IBM achieved moving the 970 to the 90-nm process, the upcoming dual-core 90-nm G4 could be not much larger than the 130-nm G4. So a dual-core G4 could cost not much more than the $245 retail list price that Motorola states for the 7447A used in the latest PowerBooks. Compare that to the $600 list price for the topend Pentium 4 and Pentium-M chips. What IBM is charging Apple for the G5 is unknown, but I can't see IBM being stupid enough to not charge more than what the G4 is costing Apple. Afterall, the G5 is supposed to be a higher performing chip and should therefore cost a premium.

[zync]

This whole thread has been the most interesing read I've had for months, thanks to everyone who contributed. It's certainly given me a lot to think about.

P.S. Zync, I think ClimbingTheLog was being ironic about the 180º+ viewing angle.

Maybe, but it didn't read like it. Though I'd been awake since like 2 p.m. on saturday and my post was at like 8:30 sunday morning :)

[ddtlm]

thatwendigo:

That's because the Athlon 64 FX series (which is the only Athlon 64 to beat the P4EE, as of the last time I looked) ... Also, the Athlon 64 is a newer chip, not just a further ramping of the same execution cores and pipelines that were introduced three or four years ago.
Ah yes, but as you recall my point was that simply adding cache to an existing design does not make it high-performance, which was what "Mr. MacPhisto" claimed about the Pentium M some posts ago. So we are agreeing, it seems. :)

Having something competitive with the G4 would give them a chance at Motorola's market, and unless I'm vastly mistakenm IBM's got a much bigger warchest than Motorola.
Well they could do it, but I'm just thinking that the incentive for them to do it is small. They want to turn a profit. To me it seems that they would profit more by working on the 970 family, and in my opinion the fact they stuck to 512k L2 at 90nm shows that they want to make it attractive to lower-end machines.

I find it likely that if they're making it as backwards compatible as the site claims, the memory bus will run in older boards as well, just not with the advantages of the new bus or memory clock.
Hmmm well I guess its not impossible that the processor could support two different FSB modes. That would be a clever trick.

Phinius:

Intel would be hard pressed to put three Pentium 4 processors on one reasonably sized chip and the Pentium-M is expected to move to dual-core in 2006.
Yeah the way its looking, Intel is gona be hard pressed to do anything useful with P4's. :) I'm thinking that you'll be seeing Intel chips that look something like A64, in the future.

The e600 will not need a Northbridge bus, as it will have a onboard memory controller. FreeScale does state that the e600 will be pin compatable with the 74XX series.
OK, this seems like a reasonable direction to take.

On some applications a dual-core e600 (G4) should be at least 50% faster clock-for-clock compared to the current G4s and probably faster than the G5.
I'd put a lot of emphasis on some there. If there's one thing time has shown me, its that people always overestimate the performance of the "next big thing". Sure it'll be faster than a current G4, but this appears to be the same core as we've had for the past couple years, so lets not expect miracles. Its not exactly advanced, practically an in-order processor, sporting just a single unit capable of doing double-precision floating point math. If I were to characterize the estimated performance boost of this e600, I'd say "from 0% to 50%". I couldn't see much a boost in, for example, code that was mostly bottlenecked on double-precision float math. That sort of thing will definately continue to be the G5's stomping grounds.

The e700, though...

[quagmire]

Actaully keeping the G4 line alive and increasing the speed could help Apple keep the consumer and pro line different.

Pro Line = G5
Consumer Line - G4

a 2GHz iBook would be a very nice little machine.

The ibook will stay a G4 intill G6 comes out. They will make the powerbook a 2.5 Ghz+ G5. I am only taking the G4 intill G6 perdiction because when the G5 came out ibook G4's came out. With the powerbook G5's we will see a 20" screen.

[cb911]

well it's good to see that Moto is still coming out with new chips (or at least planning to).

i don't think it's possible for Apple to say definitely whether or not they're going to stick with IBM now, or what will happen. as we've seen far too often chips get delayed and there are yeild issues that prevent G5 Xserves from being released on time. ;)

it's going to be a good thing for Apple to have both Moto and IBM working on chips now.

[zync]

The ibook will stay a G4 intill G6 comes out. They will make the powerbook a 2.5 Ghz+ G5. I am only taking the G4 intill G6 perdiction because when the G5 came out ibook G4's came out. With the powerbook G5's we will see a 20" screen.

Where are you coming up with this? Tell me you're joking. I hope you're at least being sarcastic. The iBook may stay a G4 if this FreeScale deal comes about, however it will most likely move to the upper end G4's and then transition into older G5's. The PowerBook may get a FreeScale G4 but I'd bet they're going to put next gen G5's in it. A 20" screen? NO ONE want's to carry something that huge around. The 17" is big enough as it is...adding three more inches (even though it's measured diagonally) would make it far to big for many bags and everything. Unless however you suppose the OLED research Apple put money into will come to fruition and they'll put some of them into the PowerBook's and have it expand off a roller or something. I'm not exactly sure why they haven't bothered to put in OLED displays into their laptops. They're cheaper, require less power, and can have a much higher resolution due to the fact that they can be layered instead of having to have transistors for sub-pixels.

[sjk]

Where are you coming up with this? Tell me you're joking. I hope you're at least being sarcastic.A "perdiction" like quagmire's surely can't be taken seriously "intill" there's more evidence. :rolleyes:

[thatwendigo]

I've got to admit that I love ideas circulating around concerning a modular design that would allow a new Mac to be AIO - but would also allow it to be a stand alone unit. I agree that the iMac has reached the end of the line. The original G3 iMac was a brilliant design and necessary to the Mac's rejuvenation. The FP iMac was even more brilliant, but I really think the AIO has it the wall (although there may be a market for another LCD AIO to replace the eMac to the education market).

I have to confess that you've got me on terminology, and I hate to do that. :D

What do you mean by AIO? Is it anything like the idea I floated a while back, by any chance?

I hadn't been aware of that before now, but it looks an awful lot like something my dad and I used to talk about. Eventually, we think that home users will all be using a sort of modular system for their needs. It could even work pretty well in certain kinds of businesses.

IBM and a few others are working on designs that are pretty close to what we envisioned a few years ago. Basically, you have a central unit with your main drive, processor, RAM, and so on, where your work and personal stuff would stay. Then, there are form-factor units that you'd slot the module into - desktops, laptops, palmtops... It'd only be limited by what the engineers could squeeze into the space. Taking a lesson from the Sun server idea, though, you'd also have auxilliary modules. Slottable drives, extra RAM, and so on in high-badnwidth external units.

So, say that you had a single processor personal module, with 1 GB of RAM, a 250GB drive, and a low-end graphics card. It serves pretty well at home, and runs your laptop shell, too. When you're at work, though, you can drop it into your eight-processor, 20 GB RAM, 2 TB disk array system, so that your workflow is defined by your already-established preferences, and a hard-coded data lock for business users keeps you from taking any of it home.

Far fetched?

IBM's got working prototypes as we speak, and Sun does the drive/RAM/processor module in servers. It's just a matter of time before someone shrinks it.

As I like to keep reminding people: Think Different.

I'm optimistic for the future, though I feel that Apple needs to be able to offer better options on their low-end. And there's a need for better video cards across the lineup. I realize this is not Apple's fault, but they should be more proactive in getting essential hardware developed for the platform.

They're working on it, and I think we might have some pleasant surprises in the near future. The eMac is pretty competitive for the price, though it's no performance hog, and the new PowerBooks have top-of-the-line GPUs with the option to go 128MB on the VRAM. That's definitely an improvement over previous generations.

I still have my fingers crossed in hope that we'll see either the NV6800 or Wildcats and FireGLs demoed at WWDC.

[Phinius]

Well they could do it, but I'm just thinking that the incentive for them to do it is small. They want to turn a profit. To me it seems that they would profit more by working on the 970 family, and in my opinion the fact they stuck to 512k L2 at 90nm shows that they want to make it attractive to lower-end machines.

Your probably right that IBM was aiming for lower-end computers when it left the L2 cache at 512Kb. There is also the possbility that IBM is concentrating on shrinking the size of the processor in order to move it to a dual or triple core in the near future.

The size of the 750FX chip being 62mm2, and the rumor of 3 processors in the Xbox2, led me to calculate the possibity of another substantial shrink at 65-nm, with a triple-core chip using only 1MB of shared L2 cache. That would make a chip that is about the same size and power use as the 970 at 130-nm. Whether that would even be intended for Apple's market is another matter. Yet, consider the possiblities for Apple if they would use two of these chips in a computer box. That would be six processors and a whole lot of performance for servers or PowerMacs, which would certainly help justify the higher prices that Apple charges. Plus, it's overall processor power use would probably not be any higher than what the dual 2GHz PowerMacs are right now. Could it be that Apple was anticipating something like that when it made a bigger PowerMac box with 9 fans in it?


Yeah the way its looking, Intel is gona be hard pressed to do anything useful with P4's. :) I'm thinking that you'll be seeing Intel chips that look something like A64, in the future.

In the near future Intel intends to only use 64-bit Xeon or Itanium processors in servers, the desktop will remain 32-bit. Intel seems to have plans to use the Pentium 4 at least through 2005 and it should be into the 5 Ghz range by early 2005.


Sure it'll be faster than a current G4, but this appears to be the same core as we've had for the past couple years, so lets not expect miracles. Its not exactly advanced, practically an in-order processor, sporting just a single unit capable of doing double-precision floating point math. If I were to characterize the estimated performance boost of this e600, I'd say "from 0% to 50%". I couldn't see much a boost in, for example, code that was mostly bottlenecked on double-precision float math. That sort of thing will definately continue to be the G5's stomping grounds.

A dual-core e600 (G4) would have two double-precision floating point units, one on each processor. The 970FX would definitely be faster on single threaded apps clock-per-clock. But, with dual threaded applications the G5 would probably have to be at a higher frequency than the e600's topend 2GHz. The higher costs associated with the G5s memory subsystem is another drawback compared to the e600. Then there is the problem of higher power use for the G5 compared to a dual-core e600. The greater watt use should make it tougher to get the higher performing G5s into a small box than it would for a dual-core e600 chip.

Having a less advanced processor has advantages. The G4 uses less watts per Mhz on the same process size compared to the G5 and due to that it's much easier to put two of them on one chip and still maintain a reasonable watt usage. A dual-core e600 chip running at a top speed of 2GHz would probably only use about 35 watts. Compare that to the 25 watts that a single 2GHz 970FX uses.

Intel plans on putting more than two Itanium cores on one chip and they are evidently going to use a less advanced Itanium core to keep the power use down. Sun also plans to do something similar and the company terms it throughput computing. Rather than a high performance uni-processor using a high amount of watts, put several less advanced processors together on one chip. IBM's Blue Gene project to create a supercomputer in a small box uses the less advanced PowerPC SOC processor that runs at 500MHz. They were able to cram lots of processors in a very small space with a much smaller power use per performance. Less advanced processors are slower for use as a uniprocessor, but they are also much more efficient for power use/performance.

[thatwendigo]

Ah yes, but as you recall my point was that simply adding cache to an existing design does not make it high-performance, which was what "Mr. MacPhisto" claimed about the Pentium M some posts ago. So we are agreeing, it seems. :)

No, I don't think he did, though he did say that the added cache would certainly help the design. Basically, the VX and the e600 look like different takes on making a new version of the older processors. One uses the 750 core, the other the 74xx core, and both increase the cache in order to work around lower FSBs.

However, I will readily agree that merely adding cache will not fix the performance problems. The P4EE is not all that much faster than the regular P4, just to name the most obvious example.

Well they could do it, but I'm just thinking that the incentive for them to do it is small. They want to turn a profit. To me it seems that they would profit more by working on the 970 family, and in my opinion the fact they stuck to 512k L2 at 90nm shows that they want to make it attractive to lower-end machines.

The 970 is a poor choice for lower-end, at least at this stage in its development. I don't think it's really meant as more than a stopgap measure that will become the low end within a year, as the new Power5 and PPC 980 chips are rolled out. When the bigger boys are out to play, then it might see a drop in price that would make it more appealing farther down.

Hmmm well I guess its not impossible that the processor could support two different FSB modes. That would be a clever trick.

I wouldn't put it past IBM to be capable of doing it, either.

I'd put a lot of emphasis on some there. If there's one thing time has shown me, its that people always overestimate the performance of the "next big thing". Sure it'll be faster than a current G4, but this appears to be the same core as we've had for the past couple years, so lets not expect miracles. Its not exactly advanced, practically an in-order processor, sporting just a single unit capable of doing double-precision floating point math. If I were to characterize the estimated performance boost of this e600, I'd say "from 0% to 50%". I couldn't see much a boost in, for example, code that was mostly bottlenecked on double-precision float math. That sort of thing will definately continue to be the G5's stomping grounds.

The e700, though...

As with the 970, I think that the e600 is the stopgap to hold people until the bigger, badder chips (or is that smaller?) show up and start kicking ass. The e600 strikes me as an embedded solution that is supposed to appeal to people who already use the G4, while the e700 will be working towards new applications of the technology.

[Mr. MacPhisto]

AIO = All-in-One like the iMacs, eMacs, and Mac Classics, etc.

I've seen a couple concepts designed by Macheads that were pretty cool. Essentially, the new cinema displays would be able to attach directly to the iMac replacement so they would act like an AIO. You'd only need to push one button to power it on - basically attaching to the back of the machine with an ADC connector.

[thatwendigo]

AIO = All-in-One like the iMacs, eMacs, and Mac Classics, etc.

I've seen a couple concepts designed by Macheads that were pretty cool. Essentially, the new cinema displays would be able to attach directly to the iMac replacement so they would act like an AIO. You'd only need to push one button to power it on - basically attaching to the back of the machine with an ADC connector.

Actually... Yeah.

That does sound pretty interesting, though I'm not sure what you'd use the detachable screen for without there being something like 802.11z (with 10GB/s bandwidth) and and embedded graphics chip to speed handling of the data and/or cache the reusable elements. What's the point of a detachable screen?

Let me guess... It's the headless iMac crowd again.

[Mr. MacPhisto]

Actually... Yeah.

That does sound pretty interesting, though I'm not sure what you'd use the detachable screen for without there being something like 802.11z (with 10GB/s bandwidth) and and embedded graphics chip to speed handling of the data and/or cache the reusable elements. What's the point of a detachable screen?

Let me guess... It's the headless iMac crowd again.

Yes and no. The LCD display would be the new Apple Cinema display - but it could attach to these modular Macs. They'd be more than just a headless iMac; they'd be able to be upgraded fairly easily and would function with another display as well. The attachable display would serve those who want to save space and have everything in one package - but that would be only one option.

Of course, I'd be perfectly happy with a redesigned version of the G4 Tower (MDD) that's smaller but has an AGP 8x slot, PCI slots, etc. I wouldn't push for PCI-X or PCI Express on the low-end. A minitower G5. The problem is it would require extra slots on the new PowerMacs to give them much more versatility and expandability.

If I were head of design of Apple, here's what I'd do:

1. Offer the PowerMacs with only dual processors. If the chips arrive on schedule, I'd offer dual 2.5 and 3 GHZ G5 machines with PCI-Express, etc. These machine would also run quietly and have all the heat monitors, etc like the present PowerMacs. I'd hope to find a way to increase the amount of expansion slots as well. I think a tower should have 5 slots + the AGP slot. I'd also hope for high-end memory (dual channel DDR533 is available, I believe. haven't kept up with memory). Maybe the possibility of putting another optical drive in the case or up to 4 internal hard drives.

2. Offer a single chip G5 machine in a stylistic, though less sophisticated case (like the G4 casing - or a small aluminum case). If this machine had high speed expansion slots, it would have to be limited. It'd have an AGP slot and 3 expansion slots. It'd also feature slower memory (DDR400). May also not be Serial ATA hard drives. With those limitations, I could see offering single 2GHZ and 2.5GHZ G5s. Maybe it's not all that realistic. If a Moto option becomes availabe, maybe a new G4 also is available on the low-end (with a different logic board, etc of course).

3. iBooks and PowerBooks stay basically the same. If the dual-core G4s become available sooner rather than later (if at all) I could see those in the PowerBooks at higher speeds with single-core G4s in the iBooks. Faster memory and video cards in the PowerBooks.

4. A new version of the eMac with an LCD screen - but without an arm. These would have G4s. My guess would be single-core, though it'd be awesome to see dual core G4s in a desktop solution. USB and Firewire access on the front. I wish Apple would develop a way to save the LCD and move it to a new machine in the future though. That's what I hate about AIO; once the machine becomes obsolete you lose the screen with it.

I'd like to see more VRAM as an option across the lineups, like what they're doing on the 15 and 17 inch PBs.

[thatwendigo]

Yes and no. The LCD display would be the new Apple Cinema display - but it could attach to these modular Macs. They'd be more than just a headless iMac; they'd be able to be upgraded fairly easily and would function with another display as well. The attachable display would serve those who want to save space and have everything in one package - but that would be only one option.

Okay, so I think we have a different set of reference for "headless iMac," since the people who I repeatedly denegreate with the term are the ones who want some mini-tower mac. I'm against the idea for a couple of reasons, not the least of which being that it still won't be price-performance competitive with the PC world at the same form factor. Single G5s are nice, but they're going to get killed by cheap PC manufacturers packaging Athlon 64s and P4 3.4s.

In general, most of the people I refer to as the "headless iMac crowd" want a machine with a graphics card slot, PCI and/or PCI-X, and other aspects that are not traditional parts of the iMac line. It really does boil down to "I'm cheap and I want a G5, gimme," in most cases.

Of course, I'd be perfectly happy with a redesigned version of the G4 Tower (MDD) that's smaller but has an AGP 8x slot, PCI slots, etc. I wouldn't push for PCI-X or PCI Express on the low-end. A minitower G5. The problem is it would require extra slots on the new PowerMacs to give them much more versatility and expandability.

No... I think that you could make a pretty good case that, were Apple to go the way of he low-end G5 (and I hate the idea, unless it's done along certain lines), that the pro line could be pretty easily differentiated.

I want the iMac as it exists to be dead, anyways. But in it's place, if you really have to have this low-end G5, the desktop line ought to look like this:

eMac - 1.33/1.5ghz, 256/512MB PC2700 RAM, 60/80GB PATA 7200 RPM, Combo/Superdrive, Radeon 9200 64MB, $799/$999

cMac (3 PCI-X, 1 AGP 8x, 1 Optical, 2 HDs, i.e. Current Towers)
-- 2.0ghz G5, 512MB PC3200, 80GB SATA 7200 RPM, SuperDrive, Radeon 9600 Pro 128MB, $1399
-- 2.5ghz G5, 512MB PC3200, 120GB SATA 7200 RPM, SuperDrive, Radeon 9600 XT 128MB, $1799
-- 3.0ghz G5, 1GB PC3200, 250GB SATA 7200 RPM, SuperDrive, Radeon 9800XT 256MB, $2,199

PowerMac (3 PCI-X, 2 PCI Extreme, 1 AGP 8x, 4 HDs with SATA RAID or 2 Optical - Deepen the chasis, add four more fans midbay)
-- Dual 2.0ghz G5, 512MB PC4200, 2x 80GB SATA 7200, SuperDrive, NV6800 128MB, $1,999
-- Dual 2.5ghz G5, 1GB PC4200, 2x 80GB SATA 10000, SuperDrive, NV6800 128MB, $2,499
-- Dual 3.0ghz G5, 1GB PC4200, 2x 120GB SATA 10000, SuperDrive, NV6800 256MB, $2,999

The downsides: Almost certainly a cut in margins.

The upsides: Holy ****! Who's going to top the pro line in performance-to-value? If this doesn't grow pro marketshare, then there's something wrong with the buyers.

1. Offer the PowerMacs with only dual processors. If the chips arrive on schedule, I'd offer dual 2.5 and 3 GHZ G5 machines with PCI-Express, etc. These machine would also run quietly and have all the heat monitors, etc like the present PowerMacs. I'd hope to find a way to increase the amount of expansion slots as well. I think a tower should have 5 slots + the AGP slot. I'd also hope for high-end memory (dual channel DDR533 is available, I believe. haven't kept up with memory). Maybe the possibility of putting another optical drive in the case or up to 4 internal hard drives.

I've long been a proponent of the all-dual pro line, so I'm fully with you on that. Also, looking into prices on dual-channel PC4200 RAM (DDR533), you can get a double-stick kit from major vendors for around $280 retail. Looks like that's going to be what comes next on the other side of the fence.

3. iBooks and PowerBooks stay basically the same. If the dual-core G4s become available sooner rather than later (if at all) I could see those in the PowerBooks at higher speeds with single-core G4s in the iBooks. Faster memory and video cards in the PowerBooks.[quote]

Agreed on all points.

[quote]I wish Apple would develop a way to save the LCD and move it to a new machine in the future though. That's what I hate about AIO; once the machine becomes obsolete you lose the screen with it.

It's not really an all-in-one, at that point, though. It kind of defeats the purpose and you might as well buy a mini-ATX machine.

I'd like to see more VRAM as an option across the lineups, like what they're doing on the 15 and 17 inch PBs.

I think this is coming, actually, if the PowerBooks are at all an indication.

[Mr. MacPhisto]

Okay, so I think we have a different set of reference for "headless iMac," since the people who I repeatedly denegreate with the term are the ones who want some mini-tower mac. I'm against the idea for a couple of reasons, not the least of which being that it still won't be price-performance competitive with the PC world at the same form factor. Single G5s are nice, but they're going to get killed by cheap PC manufacturers packaging Athlon 64s and P4 3.4s.

Apple could match them, I think, if they decided to go the way of using off-the-shelf cases, etc - though the motherboard is an issue because Apple has to design and have it built. The biggest question is can they get a minitower in the $800-$900 range? I think that's plausible. Cheap people aren't going to buy Macs anyways when they see Gateway is offering a P4 and 17" monitor for $499. For me, the biggest question is upgradeability outside of the pro lineup - but not necessarily across the entire line.

In general, most of the people I refer to as the "headless iMac crowd" want a machine with a graphics card slot, PCI and/or PCI-X, and other aspects that are not traditional parts of the iMac line. It really does boil down to "I'm cheap and I want a G5, gimme," in most cases.


I understand that. To be honest, I don't care if it's a decent G4 or a G5. I think a decent single G5 should be able to come in at $1200 with a nice graphics card (not top of the line) and some expandibility. A lot of people do want their cake and eat it too - they want it dirt cheap (I've seen $500 mentioned) and upgradeable. This is not going to happen in the Apple world, but I could see an entry level system starting at $800-900 and running up to around $1500.

No... I think that you could make a pretty good case that, were Apple to go the way of he low-end G5 (and I hate the idea, unless it's done along certain lines), that the pro line could be pretty easily differentiated.


It really all depends on what we've been talking about in this thread (and what a great thread it has been!) - can FreeScale deliver a viable G4 alternative for the low-end and the laptops - and can they make it inexpensive enough and fast enough for it to be used instead of the G5. The G5 is not a too terribly expensive chip. The Power5 derivaitives are supposed to cost approximately the same as the 970FX. I don't know the prices exactly, but I'm positive they run cheaper than the G4s have been. If they could have a decent chip manufactured for $100-150 then it'd be easier to do something. The SOC design would cut costs for desktops too because the logic board would cost less to design and manufacture. 2GHZ+ G4s on the lower end consumer line (not dirt cheap though)?

I want the iMac as it exists to be dead, anyways. But in it's place, if you really have to have this low-end G5, the desktop line ought to look like this:

eMac - 1.33/1.5ghz, 256/512MB PC2700 RAM, 60/80GB PATA 7200 RPM, Combo/Superdrive, Radeon 9200 64MB, $799/$999

cMac (3 PCI-X, 1 AGP 8x, 1 Optical, 2 HDs, i.e. Current Towers)
-- 2.0ghz G5, 512MB PC3200, 80GB SATA 7200 RPM, SuperDrive, Radeon 9600 Pro 128MB, $1399
-- 2.5ghz G5, 512MB PC3200, 120GB SATA 7200 RPM, SuperDrive, Radeon 9600 XT 128MB, $1799
-- 3.0ghz G5, 1GB PC3200, 250GB SATA 7200 RPM, SuperDrive, Radeon 9800XT 256MB, $2,199

PowerMac (3 PCI-X, 2 PCI Extreme, 1 AGP 8x, 4 HDs with SATA RAID or 2 Optical - Deepen the chasis, add four more fans midbay)
-- Dual 2.0ghz G5, 512MB PC4200, 2x 80GB SATA 7200, SuperDrive, NV6800 128MB, $1,999
-- Dual 2.5ghz G5, 1GB PC4200, 2x 80GB SATA 10000, SuperDrive, NV6800 128MB, $2,499
-- Dual 3.0ghz G5, 1GB PC4200, 2x 120GB SATA 10000, SuperDrive, NV6800 256MB, $2,999

The downsides: Almost certainly a cut in margins.

The upsides: Holy ****! Who's going to top the pro line in performance-to-value? If this doesn't grow pro marketshare, then there's something wrong with the buyers.

I could live with those prices, though I'd like to see the eMacs bumped up in speed a bit, if there is that possibility. It comes back to FreeScale again. If they have a 90nm 74xx available for the next eMac revision with improved chip architecture, greater bandwidth, etc. then I could see the eMac going up to 1.8 GHZ - which would be a pretty sweet deal at $999. It'd be nice to see a 17" 4:3 LCD replace the CRT, but I don't think it's possible to do that yet - not for that kind of money, though prices are still coming down. I haven't seen any numbers lately on how much is costs to manufacture LCDs (usually given out per inch). A quick search indicated that Samsung hoped to manufacture 17" screens for $10 an inch by 2005, but were running ahead of schedule (this was in 2002). If they've come down that far then it would cost $170 for a 17" screen to go in an eMac. It could all hinge on chip prices.


I've long been a proponent of the all-dual pro line, so I'm fully with you on that. Also, looking into prices on dual-channel PC4200 RAM (DDR533), you can get a double-stick kit from major vendors for around $280 retail. Looks like that's going to be what comes next on the other side of the fence.


I'm pretty sure that'll be the way to go. I'm sure my bank account will take a big hit when those get released. Dual 3GHZ with PC4200 starting at $2999 with a 256MB nVidia? Like you said earlier, the margins would shrink - but what a spectactular deal! Of course, increased sales would decrease the impact of the margin reduction. I must say that Apple has done pretty well in trying to get the prices down somewhat. To be honest, I think high G4 prices have prevented bigger drops. Once again, the big IF lies with FreeScale. IF they can get costs down we may see greater reductions in the future. The PowerMac prices will be fine as they are with a revision (and will be a steal). Of Apple could find a way to slice $100 off everything else they'd be in spectacular shape - but I won't hold my breath


It's not really an all-in-one, at that point, though. It kind of defeats the purpose and you might as well buy a mini-ATX machine.


This is true. Maybe a trade-in program so Apple will give you a certain amount off a future AIO Mac if you bring in your old eMac for "recycling". Probably too much to ask again.


I think this is coming, actually, if the PowerBooks are at all an indication.

I think and hope so. I'd love to see 64MB options on eMacs and iBooks. The iBooks have very good video cards for consumer end laptops (in fact, better than anything I've seen on high-end PC laptops I've looked at, though I haven't done an exhaustive search), but the PowerBooks have even better cards (except for the 12inch). This upgrade may be limited to the 14" iBook, unless they can find a way to cram 128MB into the 12" PB.

As you referenced in your prices - 256MB options on PowerMacs. I'd probably allow for 128MB cards on mid-level machines. However, if Apple does have more machines out there that have AGP slots I can see more interest and business in aftermarket Mac videocards. That'd be essential for future upgrades on those mid-market machines.

As for the eMacs, if they're given a good enough video card to start with (they should be equal to the iBook's). They did upgrade it to the 9200 finally, but it took a while before it equalled the iBook. I wonder if it would be more cost efficient for the eMac and iBook to share logic boards? Obviously the eMac could have a larger enclosure, but would an SOC design on the G4 make this a better alternative - and possibly help to cut costs?

[thatwendigo]

Apple could match them, I think, if they decided to go the way of using off-the-shelf cases, etc - though the motherboard is an issue because Apple has to design and have it built. The biggest question is can they get a minitower in the $800-$900 range? I think that's plausible. Cheap people aren't going to buy Macs anyways when they see Gateway is offering a P4 and 17" monitor for $499. For me, the biggest question is upgradeability outside of the pro lineup - but not necessarily across the entire line.

Nope. My reasoning?

Look here (http://forums.macrumors.com/showthread.php?t=68163) amd here (http://forums.macrumors.com/showpost.php?p=818147&postcount=522)

One of the only commodity-based attempts at a PowerPC motherboard is a measly 1ghz MPC7447, and while it's better than the older Apple G4 boards, it still costs $775!

I understand that. To be honest, I don't care if it's a decent G4 or a G5. I think a decent single G5 should be able to come in at $1200 with a nice graphics card (not top of the line) and some expandibility. A lot of people do want their cake and eat it too - they want it dirt cheap (I've seen $500 mentioned) and upgradeable. This is not going to happen in the Apple world, but I could see an entry level system starting at $800-900 and running up to around $1500.

I'm dead against that approach. The G5 needs to be a speed demon. The best price/performance mix possible needs to go into every machine that's sold with that name attached to it, just as a bulwark against the years upon years of FUD that macs are slower. Even with single processors, the machines need to scream as high as economically possible, even if that means not directly competing in the sub-$1,500 space.

Apple cannot, at the current market state, compete below a certain point. I think it's best that they just make the best computers possible and let others eat the cheapskates.

Blah, and it's late. I'll cover the rest tomorrow. :P

[ClimbingTheLog]

Sorry this is just a small thing. You can't have anything beyond a 180º viewing angle because the other side would have to be pure LCD as well :)

Ummm... yeah, that was my point. Sorry, I have a tendency to be obtuse, or at least not very expansive.

The point is that Apple's screens used to have a better viewing angle than the competition. But as the competition's angle approaches 180 degrees Apple doens't have that competitive advantage. Ditto for brightness, contrast, resolution, and they still don't have foolproof color, which is about the only place they can go for cover. Once the competition has fool-proof color they're done. They have a frame, and that's it. Sure, the Mac has a nice frame, but it's much more than the competition on the inside.

[ClimbingTheLog]

As much as I'm hoping we don't get burnt again, Freescale is already moving faster than Motorola was on the PowerPC arcitecture. Their new e500 core is on the line right now, runs ramping up for the embedded market.

I think lots of us here would like Freescale to succeed, but our thrice-bitten motto has become "show me the money" (Gooding Voice).

Maybe it should be "show me the flops", but there's no movie scene to think about. :)

[zync]

Ummm... yeah, that was my point. Sorry, I have a tendency to be obtuse, or at least not very expansive.

The point is that Apple's screens used to have a better viewing angle than the competition. But as the competition's angle approaches 180 degrees Apple doens't have that competitive advantage. Ditto for brightness, contrast, resolution, and they still don't have foolproof color, which is about the only place they can go for cover. Once the competition has fool-proof color they're done. They have a frame, and that's it. Sure, the Mac has a nice frame, but it's much more than the competition on the inside.

Sorry, I was hoping you were making light of things. I agree, the displays need a revamp and they need it badly. I'd like to see a resolution increase, with a possible switch to OLED as it'll make them cheaper and afford a much higher resolution. They do have a nice frame however it's kind of large for multi-panel setups. The 23" does have some real estate on it, I just wish it were able to be filled with more pixels :D

[thatwendigo]

Funny how I said it was too late, and that's why I wouldn't do my response until later. Here I am, at 6 AM... :rolleyes:

It really all depends on what we've been talking about in this thread (and what a great thread it has been!) - can FreeScale deliver a viable G4 alternative for the low-end and the laptops - and can they make it inexpensive enough and fast enough for it to be used instead of the G5. The G5 is not a too terribly expensive chip. The Power5 derivaitives are supposed to cost approximately the same as the 970FX. I don't know the prices exactly, but I'm positive they run cheaper than the G4s have been. If they could have a decent chip manufactured for $100-150 then it'd be easier to do something. The SOC design would cut costs for desktops too because the logic board would cost less to design and manufacture. 2GHZ+ G4s on the lower end consumer line (not dirt cheap though)?

I agree that this has been an enlightening and vastly interesting thread. Thanks for the material you've brought to my attention, at the very least.

The question of whether Freescale can deliver a viable laptop processor is one that I think is almost moot. Of course they can. At issue is the timeframe, the competitive standing of the eventual core, and the cost per unit, since it could very well be that the allegations some have made (that clocking down the G5 and somehow fitting its subsystems in would be cheaper) are actually true. I find this not only unlikely, but ludicrous, without some kind of major revamp of the 970 line that yields even lower heat. Of course, we've been all over that issue already.

I could live with those prices, though I'd like to see the eMacs bumped up in speed a bit, if there is that possibility. It comes back to FreeScale again. If they have a 90nm 74xx available for the next eMac revision with improved chip architecture, greater bandwidth, etc. then I could see the eMac going up to 1.8 GHZ - which would be a pretty sweet deal at $999. It'd be nice to see a 17" 4:3 LCD replace the CRT, but I don't think it's possible to do that yet - not for that kind of money, though prices are still coming down.

Actually, I could see a form-factor revision for the eMac if it were to move to LCD. It's for education, after all, so you want stability and solidity more than anything else, right? Take a page out of the 20th Anniversary Mac and the current PC all-in-one market... You could have the screen fronting a flat enclosure with some CHUNKY and rather solid adjustable legs, with a 1.8-2.0ghz e600 (what's that, like 25-20w?), a slot-loading combo drive, and the RAM and HD accessible from lockable panels in the back. Truly make it an education machine - sturdy, reasonably specced by not a bank breaker, and with a good screen (front the LCD in glass for protection and cleaning). Price it at $999 (generic LCDs are $350 retail for 17" now) and move those things off the shelf.

I haven't seen any numbers lately on how much is costs to manufacture LCDs (usually given out per inch). A quick search indicated that Samsung hoped to manufacture 17" screens for $10 an inch by 2005, but were running ahead of schedule (this was in 2002). If they've come down that far then it would cost $170 for a 17" screen to go in an eMac. It could all hinge on chip prices.

I think it's either not quite that far down, or that insane profits are being taken, even in the PC world.

I'm pretty sure that'll be the way to go. I'm sure my bank account will take a big hit when those get released. Dual 3GHZ with PC4200 starting at $2999 with a 256MB nVidia? Like you said earlier, the margins would shrink - but what a spectactular deal! Of course, increased sales would decrease the impact of the margin reduction.

If they do 256MB, I don't expect it to be standard, and while I think a margin reduction might be a good idea, I don't know what kind of impact it would have on the Apple warchest. I'm most interested in seeing them keep innovating and providing a good experience, not in growing marketshare.

That was my wishlist, not necessarily my "this is realistic" list.

I must say that Apple has done pretty well in trying to get the prices down somewhat. To be honest, I think high G4 prices have prevented bigger drops. Once again, the big IF lies with FreeScale. IF they can get costs down we may see greater reductions in the future. The PowerMac prices will be fine as they are with a revision (and will be a steal). Of Apple could find a way to slice $100 off everything else they'd be in spectacular shape - but I won't hold my breath

As you said, the key is in Freescale. The G4 is not cheap at the kinds of speeds that we're talking about, and I'm hoping that Crolles and the new direction will help fix that situation. If not an outright correction, it would be nice to see it at least swinging in the right direction. I could accept delivery of the chip at current prices or thereabouts, with revisions as the process and design matures.

This is true. Maybe a trade-in program so Apple will give you a certain amount off a future AIO Mac if you bring in your old eMac for "recycling". Probably too much to ask again.

It's the tradeoff of AIO. You get a cheaper machine, but less expandability and you're tied down. It's not like there's a huge market for returned AIO's that they could exploit.

I think and hope so. I'd love to see 64MB options on eMacs and iBooks. The iBooks have very good video cards for consumer end laptops (in fact, better than anything I've seen on high-end PC laptops I've looked at, though I haven't done an exhaustive search), but the PowerBooks have even better cards (except for the 12inch). This upgrade may be limited to the 14" iBook, unless they can find a way to cram 128MB into the 12" PB.

I think a move to at least 64 on the eMac (I'd even support 128 if it didn't hurt the price point too badly), and at least 64 on the iBooks would be a good idea. As ATI and nVidia progress with their cards, I think we might see some significant heat savings. The stats at this review of the new Radeon X800 (http://www.hothardware.com/viewarticle.cfm?articleid=517&cid=2) show that it clocks higher than nVidia's offerings, performs better in nearly all tests, and has better power-saving features.

You might say GPUs are getting just as interesting as CPUs at the moment.

As you referenced in your prices - 256MB options on PowerMacs. I'd probably allow for 128MB cards on mid-level machines. However, if Apple does have more machines out there that have AGP slots I can see more interest and business in aftermarket Mac videocards. That'd be essential for future upgrades on those mid-market machines.

Rereading the latest reviews of graphics cards, here's my modified wish list for the machines:
eMac - 1.8ghz, 512MB PC3200 RAM, 80GB PATA 7200 RPM, Superdrive, Radeon 9600 64MB, 17" ActiveTFT LCD $999

cMac (3 PCI-X, 1 AGP 8x, 1 Optical, 2 HDs, i.e. Current Towers)
-- 2.0ghz G5, 512MB PC3200, 80GB SATA 7200 RPM, SuperDrive, Radeon 9800 Pro 128 MB, $1399
-- 2.5ghz G5, 512MB PC3200, 120GB SATA 7200 RPM, SuperDrive, Radeon 9800 XT 256MB, $1799
-- 3.0ghz G5, 1GB PC3200, 250GB SATA 7200 RPM, SuperDrive, Radeon X800 Pro 256MB or NV6800GT 256MB, $2,199

PowerMac (3 PCI-X, 2 PCI Extreme, 1 AGP 8x, 4 HDs with SATA RAID or 2 Optical - Deepen the chasis, add four more fans midbay)
-- Dual 2.0ghz G5, 512MB PC4200, 2x 80GB SATA 7200, SuperDrive, Radeon 9800 Pro 256MB or GeForceFX 5950 Ultra 256MB, $1,999
-- Dual 2.5ghz G5, 1GB PC4200, 2x 80GB SATA 10000, SuperDrive, Radeon 9800 Pro 256MB or NV6800GT 256MB, $2,499
-- Dual 3.0ghz G5, 1GB PC4200, 2x 120GB SATA 10000, SuperDrive, Radeon X800 XT 256MB or NV6800 Ultra 256MB, $2,999

As for the eMacs, if they're given a good enough video card to start with (they should be equal to the iBook's). They did upgrade it to the 9200 finally, but it took a while before it equalled the iBook. I wonder if it would be more cost efficient for the eMac and iBook to share logic boards? Obviously the eMac could have a larger enclosure, but would an SOC design on the G4 make this a better alternative - and possibly help to cut costs?

Interesting trivia fact - While not sharing logic boards, the original iMac and IBook did share many components. Try taking apart a dead Bondi sometime and count how many of the components are from latops.

[ClimbingTheLog]

I'd like to see a resolution increase, with a possible switch to OLED as it'll make them cheaper and afford a much higher resolution.

It's interesting that you mention OLED. Have they solved the problem with the OLED dyes breaking down overtime? Most OLED's tend to lose one of their RGB components (forget which one) over time.

Still, that might not be a problem if the timeframe is long enough. Old CRT's certainly lost their color over time. I'm even writing this on a 10 year-old ViewSonic 17" that is starting to lose it's color fidelity (time to switch it to the second video card). I'm even lusting after a Sony T1 digicam which uses OLED and I'm not that worried because it's not going to be with me forever.

And who really uses a 5 year old computer these days besides cheapskates like me? I can see a $799 eMac with a 5-year-rated OLED doing very well.

[Mr. MacPhisto]

Funny how I said it was too late, and that's why I wouldn't do my response until later. Here I am, at 6 AM... :rolleyes:



I agree that this has been an enlightening and vastly interesting thread. Thanks for the material you've brought to my attention, at the very least.

The question of whether Freescale can deliver a viable laptop processor is one that I think is almost moot. Of course they can. At issue is the timeframe, the competitive standing of the eventual core, and the cost per unit, since it could very well be that the allegations some have made (that clocking down the G5 and somehow fitting its subsystems in would be cheaper) are actually true. I find this not only unlikely, but ludicrous, without some kind of major revamp of the 970 line that yields even lower heat. Of course, we've been all over that issue already.


True. The two big ifs involve FreeScale delivering in a timely manner and also the speed of the chips they deliver.


Actually, I could see a form-factor revision for the eMac if it were to move to LCD. It's for education, after all, so you want stability and solidity more than anything else, right? Take a page out of the 20th Anniversary Mac and the current PC all-in-one market... You could have the screen fronting a flat enclosure with some CHUNKY and rather solid adjustable legs, with a 1.8-2.0ghz e600 (what's that, like 25-20w?), a slot-loading combo drive, and the RAM and HD accessible from lockable panels in the back. Truly make it an education machine - sturdy, reasonably specced by not a bank breaker, and with a good screen (front the LCD in glass for protection and cleaning). Price it at $999 (generic LCDs are $350 retail for 17" now) and move those things off the shelf.

This is precisely what I was thinking, and Apple should be able to get a hold of a 17" LCD for under $300, likely more in the range of 200-250 if gotten in bulk for OEM purposes. You could set it on legs or you could develop a new tilt and swivel base that would give it great mobility. I'd likely place a slot-load drive on the right hand side, behind the LCD screen with USB and Firewire access located on the front of the machine. As for the e600, I believe the 25W at 2GHZ referenced referred to a dual core version of the e600, so my guess would be that a single core version for the eMac would run slightly below 20W. At $999, this would be a great AIO solution for education and for individuals. I'm not sure if they would still be able to pull of the $799 combo and $999 Superdrive, but it would be a great value with a 64MB Radeon in there as well.



I think it's either not quite that far down, or that insane profits are being taken, even in the PC world.


It's hard to say. If people are willing to pay an inflated price then the LCDs will not drop but the profit margine will increase. My guess would be Apple could get a 17" LCD for somewhere around $250 when gotten in bulk.


If they do 256MB, I don't expect it to be standard, and while I think a margin reduction might be a good idea, I don't know what kind of impact it would have on the Apple warchest. I'm most interested in seeing them keep innovating and providing a good experience, not in growing marketshare.

That was my wishlist, not necessarily my "this is realistic" list.


I'm not willing to sacrifice the innovation in favor of marketshare, but I think there is a way of maintaining high margins while being innovative and growing the marketshare. If the quality and speed of the machine can pick up, I think OS X can win converts. It's the reason I switched over, and I've been positively delighted with it for two years - as well as all the other cool things that have hit the Mac first. Apple has slowly been reducing prices. I think the biggest factor will be how much they have to pay for the CPU. If IBM and FreeScale can get CPU cost to go down (and it should as they shrink the designs) then I can see Macs getting slightly less expensive.



As you said, the key is in Freescale. The G4 is not cheap at the kinds of speeds that we're talking about, and I'm hoping that Crolles and the new direction will help fix that situation. If not an outright correction, it would be nice to see it at least swinging in the right direction. I could accept delivery of the chip at current prices or thereabouts, with revisions as the process and design matures.

How much of the prices we due to low yields in the past? If Crolles can have a higher yield then we may see a cheaper chip. I agree that a move int he right direction would be a start. I'd be willing to pay the current price on a G4 if it delivered what FreeScale has announced. I think it would be a pretty good chip. I don't think that the prices from Intel or AMD will be able to be matched on the low-end.



It's the tradeoff of AIO. You get a cheaper machine, but less expandability and you're tied down. It's not like there's a huge market for returned AIO's that they could exploit.


This is true. The benefit of buying an Apple AIO is they tend to last longer in regards to general usefulness. I know a few people running OS X on original iMacs @ 233Mhz. Things are slower, but they can still get a lot done, and there have been upgrades offered by third parties. One friend had just recently upgraded the original iMac to a G4 running above 600MHZ. Granted, it's not as fast as it could be, but it allowed him to spend less than $200 and more than double his speed (and his own tests have shown that most things do run twice as fast - my guess is Altivec has something to do with some of that).


I think a move to at least 64 on the eMac (I'd even support 128 if it didn't hurt the price point too badly), and at least 64 on the iBooks would be a good idea. As ATI and nVidia progress with their cards, I think we might see some significant heat savings. The stats at this review of the new Radeon X800 (http://www.hothardware.com/viewarticle.cfm?articleid=517&cid=2) show that it clocks higher than nVidia's offerings, performs better in nearly all tests, and has better power-saving features.

You might say GPUs are getting just as interesting as CPUs at the moment.


The only challenge is getting them to the Mac. I think all consumer products should be at 64MB with all Pro products starting at 128MB - and I can see that in the not-so-distant future. I will say something for Mac in the iBooks and eMacs - most similar PCs have got integrated Intel video with shared RAM. I'd much rather take the Radeon 9200 w/ 32MB, thank you.


Rereading the latest reviews of graphics cards, here's my modified wish list for the machines:
eMac - 1.8ghz, 512MB PC3200 RAM, 80GB PATA 7200 RPM, Superdrive, Radeon 9600 64MB, 17" ActiveTFT LCD $999

cMac (3 PCI-X, 1 AGP 8x, 1 Optical, 2 HDs, i.e. Current Towers)
-- 2.0ghz G5, 512MB PC3200, 80GB SATA 7200 RPM, SuperDrive, Radeon 9800 Pro 128 MB, $1399
-- 2.5ghz G5, 512MB PC3200, 120GB SATA 7200 RPM, SuperDrive, Radeon 9800 XT 256MB, $1799
-- 3.0ghz G5, 1GB PC3200, 250GB SATA 7200 RPM, SuperDrive, Radeon X800 Pro 256MB or NV6800GT 256MB, $2,199

PowerMac (3 PCI-X, 2 PCI Extreme, 1 AGP 8x, 4 HDs with SATA RAID or 2 Optical - Deepen the chasis, add four more fans midbay)
-- Dual 2.0ghz G5, 512MB PC4200, 2x 80GB SATA 7200, SuperDrive, Radeon 9800 Pro 256MB or GeForceFX 5950 Ultra 256MB, $1,999
-- Dual 2.5ghz G5, 1GB PC4200, 2x 80GB SATA 10000, SuperDrive, Radeon 9800 Pro 256MB or NV6800GT 256MB, $2,499
-- Dual 3.0ghz G5, 1GB PC4200, 2x 120GB SATA 10000, SuperDrive, Radeon X800 XT 256MB or NV6800 Ultra 256MB, $2,999


Now I'm going to have to wash my shirt. You made me drool all over it. I'd be putting in an order for the dual 3GHZ machine the second it was announced. If they were able to do anything near that for $2999, that would absolutely kill any PC for value - especially with the architectural improvements of the Power5 derivative (including SMT).


Interesting trivia fact - While not sharing logic boards, the original iMac and IBook did share many components. Try taking apart a dead Bondi sometime and count how many of the components are from latops.

I didn't know that. I wonder if that is happening now? If not, it would be a good idea - especially if the eMacs and iBooks share processors, HDD, videocards, etc. If they shared logic boards then it should save money - if only by the fact that Apple only need to design one board for each.

[sjk]

Hey thatwendigo, your latest future eMac specs/price proposal sure made it less inspiring to buy the current eMac. Glad my wife likes it since it'll become her system when I upgrade (by November) and it's certainly an improvement over my pokey 600Mhz G3 iBook.

Hopefully someone who can influence Apple has been following this thread.

:)

[Dont Hurt Me]

It's interesting that you mention OLED. Have they solved the problem with the OLED dyes breaking down overtime? Most OLED's tend to lose one of their RGB components (forget which one) over time.

Still, that might not be a problem if the timeframe is long enough. Old CRT's certainly lost their color over time. I'm even writing this on a 10 year-old ViewSonic 17" that is starting to lose it's color fidelity (time to switch it to the second video card). I'm even lusting after a Sony T1 digicam which uses OLED and I'm not that worried because it's not going to be with me forever.

And who really uses a 5 year old computer these days besides cheapskates like me? I can see a $799 eMac with a 5-year-rated OLED doing very well.Blue is the one they are having a problem with.

[thatwendigo]

This is precisely what I was thinking, and Apple should be able to get a hold of a 17" LCD for under $300, likely more in the range of 200-250 if gotten in bulk for OEM purposes. You could set it on legs or you could develop a new tilt and swivel base that would give it great mobility. I'd likely place a slot-load drive on the right hand side, behind the LCD screen with USB and Firewire access located on the front of the machine. As for the e600, I believe the 25W at 2GHZ referenced referred to a dual core version of the e600, so my guess would be that a single core version for the eMac would run slightly below 20W. At $999, this would be a great AIO solution for education and for individuals. I'm not sure if they would still be able to pull of the $799 combo and $999 Superdrive, but it would be a great value with a 64MB Radeon in there as well.

The other thing is that, like PC OEMS, Apple could (and I hate to say this, but we really do need kids to see the mac if we want more home users) buy some cheaper LCDs and that weren't wide-aspect and use them specifically for this purpose. It's not like those Dells and HPs are coming with top of the line 17" OLEDs ore anything, so it shouldn't be a big deal.

Here's my idea for layout, and I'd sketch this if I had a scanner or a tablet:
You have a bezel that's about an inch out, with the same kind of frame that you'd use in the older imacs or the current eMacs, only flattened. Have the frame be hollow but strong (magnesium allow, maybe?) with a relatively thick plastic shell so that they don't scar or break easily. The board would ideally be around the size of a laptop or mini-ATX and sit up high, with a riser-style AGP slot that let the GPU exhaust directly out of the top of the machine. The RAM would have a removable (but attached!) bay door that locks with screws that aren't those tiny star-head ones, and be mounted towards the lower end of the case. In fact, the 'bay door' would swing out to reveal the HD and the RAM at the same time, along with an AirPort Extreme slot. On the bottom right edge, you'd have the slot-loading optical drive, and on the left you'd have FireWire 400 (x2), FireWire 800 (x1), USB 2.0 (x3), and analog audio in/out. The bottom edge is partially rounded off, with two legs that fold up to protect the optical drive and the cords when unlocked and closed.

I'm not willing to sacrifice the innovation in favor of marketshare, but I think there is a way of maintaining high margins while being innovative and growing the marketshare. If the quality and speed of the machine can pick up, I think OS X can win converts.

The quality is already outstanding, and the only reason that people in the mac community can point out people that they know who have the same problem is that, really, we're all a lot more likely to be able to talk about it, and to share that information. Contrary to what some would seem to like to spread around, the average PC user is not an overclocker, and the fact that you know how to fool with bios and multipliers kind of removes you from the basic group.

The beauty of the Mac experience at this point is that we have PowerPCs and OS X. The bad part is that we have PowerPCs, as well, since our supplies are far more limited than they are on the PC side. However, this could very well change with Sony, Toshiba, AMD, and others signing on to the PowerPC platform. Maybe not right now, maybe not directly for Apple, but these companies are definitely showing that it's better in some situations.

I think the biggest factor will be how much they have to pay for the CPU. If IBM and FreeScale can get CPU cost to go down (and it should as they shrink the designs) then I can see Macs getting slightly less expensive.

It should. We'll see how much of it gets passed along, though.

This is true. The benefit of buying an Apple AIO is they tend to last longer in regards to general usefulness. I know a few people running OS X on original iMacs @ 233Mhz. Things are slower, but they can still get a lot done, and there have been upgrades offered by third parties. One friend had just recently upgraded the original iMac to a G4 running above 600MHZ. Granted, it's not as fast as it could be, but it allowed him to spend less than $200 and more than double his speed (and his own tests have shown that most things do run twice as fast - my guess is Altivec has something to do with some of that).

My daily use machines are a 700mhz eMac and a 600mhz G3 snowbook, so I'm well aware of the new life that OS X keeps breating into these machines with ever revision. It's palpably better when I install, snappier and more responsive, even in real-world world applications (I xBench our machines at each revision, though I think that someone who was using the tower I keep them on tossed the results a few weeks back :rolleyes: ).


The only challenge is getting them to the Mac. I think all consumer products should be at 64MB with all Pro products starting at 128MB - and I can see that in the not-so-distant future. I will say something for Mac in the iBooks and eMacs - most similar PCs have got integrated Intel video with shared RAM. I'd much rather take the Radeon 9200 w/ 32MB, thank you.

I think we should be shooting for 128MB across the board on desktops, and and at least 64MB across the board for portables. Differentiate the models on GPU core, if you really need. ATI and nVidia both offer at least two versions of their latest GPUs, and at least three of their previous generation. Also, we need pro cards back. The Wildcat and FireGL are too important in the graphics sector to be left out of the G5, which is touted as a workhorse machine for creative professionals.

Why settle for the same as PCs, though? Lets go back to the days where we could proudly run the toasted commercials on our desktops, where we could slam Intel on TV without any chance that a lawsuit would take the advertising down.

Dual 3.5ghz 975s with x800 XTs, 10k RPM SATA performance RAID, and PC4200 RAM! Toast those bunnies! :D

Now I'm going to have to wash my shirt. You made me drool all over it. I'd be putting in an order for the dual 3GHZ machine the second it was announced. If they were able to do anything near that for $2999, that would absolutely kill any PC for value - especially with the architectural improvements of the Power5 derivative (including SMT).

This is the future, if the technology will just come together. I've looked at the availability of DDR4200 RAM, just as a starter, because it's a part they'd just need to address on the motherboard and have the memory controller feed and pull from. They're going to have to adust the ASIC anyways, so why not move it up on memory access, too?

I didn't know that. I wonder if that is happening now? If not, it would be a good idea - especially if the eMacs and iBooks share processors, HDD, videocards, etc. If they shared logic boards then it should save money - if only by the fact that Apple only need to design one board for each.

The CD drive was a laptop drive, the chip was the same, the board was almost small enough and shaped right. I've got a broken (literally) iBook Rev A and a dead iMac Rev A... Maybe I'll dig them out and compare again.

[Mr. MacPhisto]

The other thing is that, like PC OEMS, Apple could (and I hate to say this, but we really do need kids to see the mac if we want more home users) buy some cheaper LCDs and that weren't wide-aspect and use them specifically for this purpose. It's not like those Dells and HPs are coming with top of the line 17" OLEDs ore anything, so it shouldn't be a big deal.

Precisely. I think they should still keep the 17" Cinema Display hanging around. Not everyone wants to invest in a 20"+ 16:9 Cinema Display, or can afford one. It's possible now to have a decent LCD display and not break the bank - although we always do have the option of not buying Apple's display. The sacrifice there is that we lose ADC (which could use and update). Of course, for the eMac you'd have no choice.


Here's my idea for layout, and I'd sketch this if I had a scanner or a tablet:
You have a bezel that's about an inch out, with the same kind of frame that you'd use in the older imacs or the current eMacs, only flattened. Have the frame be hollow but strong (magnesium allow, maybe?) with a relatively thick plastic shell so that they don't scar or break easily. The board would ideally be around the size of a laptop or mini-ATX and sit up high, with a riser-style AGP slot that let the GPU exhaust directly out of the top of the machine. The RAM would have a removable (but attached!) bay door that locks with screws that aren't those tiny star-head ones, and be mounted towards the lower end of the case. In fact, the 'bay door' would swing out to reveal the HD and the RAM at the same time, along with an AirPort Extreme slot. On the bottom right edge, you'd have the slot-loading optical drive, and on the left you'd have FireWire 400 (x2), FireWire 800 (x1), USB 2.0 (x3), and analog audio in/out. The bottom edge is partially rounded off, with two legs that fold up to protect the optical drive and the cords when unlocked and closed.

The sounds good to me. I like the idea that there is access to the RAM and the HD. Realistically, that would be good expansion capability for a machine like that, allowing the owner to add more RAM or upgrade the HD. I'd make sure that the machine could take up to 2GB (and that's where they'd hit a snag in sharing logic boards with the laptop). The cool thing is they could make it rather inexpensively while also making it look cool (and much less bulk than the current eMac). The easy access to the FW and USB ports would also make it appealing, especially for things like digital photogrpahy and home movies for families.


The quality is already outstanding, and the only reason that people in the mac community can point out people that they know who have the same problem is that, really, we're all a lot more likely to be able to talk about it, and to share that information. Contrary to what some would seem to like to spread around, the average PC user is not an overclocker, and the fact that you know how to fool with bios and multipliers kind of removes you from the basic group.

Most PC users buy their machines, maybe upgrade the HD and RAM at some point (most likely just the RAM) and use it till it breaks. Then they toss out that machine and buy another Gateway, Dell, etc. In fact, most PC users would have to drop their computer off at CompUSA for any upgrades because they don't have the know-how to even install a DIMM (and that's sad).


The beauty of the Mac experience at this point is that we have PowerPCs and OS X. The bad part is that we have PowerPCs, as well, since our supplies are far more limited than they are on the PC side. However, this could very well change with Sony, Toshiba, AMD, and others signing on to the PowerPC platform. Maybe not right now, maybe not directly for Apple, but these companies are definitely showing that it's better in some situations.

I think PowerPC is the horse to stick with. My personal belief is that it will eventually succeed the x86 as the chip of choice because it's got a longer road in front of it. We're already seeing that the x86 architecture is being pushed to its limits - and I'm not sure how much further Intel or AMD can really push it. AMD is already onboard with IBM in moving to PowerPC anyways. Intel's solution for the future has sunk (hence why we call it Itanic - and the Itanium never impressed me much; day late and dollar short). I also think we'll see the PowerPC begin to scale faster - and it may even catch up in raw clock rate to Intel in the somewhat near future (and it could be closer than you think - I'm hoping we get shocked in June. The possibility is there to hit 3.5GHZ with the new chip and have it not run overly hot. Add the 90nm controller and you should have a cooler PowerMac, but the 3.5GHZ is not a guarantee). The PowerPC also needs something to challenge the Pentium M in the laptop market (even though it's not selling as well as Intel had hoped, a victim of their "MHZ Matters" style of advertising in the past - people don't get that a 1.6+ Pentium M is a nice chip and plenty fast) - and I think FreeScale could provide the answer there, if they deliver. A G4 with improved architecture, improved cache access, larger L2 cache, better bandwidth, etc could be a strong alternative in the Mac world.



It should. We'll see how much of it gets passed along, though.


I come from the perspective that Motorola owes Apple something for the years of failure to deliver - but corporations don't often have a strong sense of guilt or obligation, but they do have a strong sense of profit. I'm hoping the cleaner fab and the chip reduction is enough to reduce the price significantly. Of course, if they can get more embedded customers to desire those chip then the price should come down more.


My daily use machines are a 700mhz eMac and a 600mhz G3 snowbook, so I'm well aware of the new life that OS X keeps breating into these machines with ever revision. It's palpably better when I install, snappier and more responsive, even in real-world world applications (I xBench our machines at each revision, though I think that someone who was using the tower I keep them on tossed the results a few weeks back :rolleyes: ).


My primary machine is an iBook 700 and I've noticed it get speedier with each OSX update. The funny thing is that even though my PC is faster in raw speed, this G3 makes up for it in lack of down time. I don't have to spend hours trying to fix a problem, wondering why Windows isn't working right. Having an UNIX backbone is a wonderful thing. Apple did a great job of making it usuable and creating the most gorgeous OS I've ever seen. After two years, using this thing is still a joy - even though I've only got a G3 700, I can still get a lot of what I need to get done with ease (and no reboots to try to correct problems).



I think we should be shooting for 128MB across the board on desktops, and and at least 64MB across the board for portables. Differentiate the models on GPU core, if you really need. ATI and nVidia both offer at least two versions of their latest GPUs, and at least three of their previous generation. Also, we need pro cards back. The Wildcat and FireGL are too important in the graphics sector to be left out of the G5, which is touted as a workhorse machine for creative professionals.

I agree. I see no problem with the PowerBooks having the Radeon 9700 @ 64MB and the iBooks having the Radeon 9200 @ 64MB. I think the kicker here is that Apple has a problem getting the high end Radeon into the 12" PowerBook. Also agree on pro cards, especially if AutoCAD is coming to the Mac, as many rumors have stated. Even stuff done with Maya, etc needs something better than a Radeon. A bad GPU could drive a lot of people away from the PowerMac.


Why settle for the same as PCs, though? Lets go back to the days where we could proudly run the toasted commercials on our desktops, where we could slam Intel on TV without any chance that a lawsuit would take the advertising down.

Dual 3.5ghz 975s with x800 XTs, 10k RPM SATA performance RAID, and PC4200 RAM! Toast those bunnies! :D


Not too sure we'll see a RAID, though it would be a good option. As for the Dual 3.5GHZ - like I said, you may see those sooner than you think. My personal thought, based on what I know, is that we could see 4GHZ+ dual-core 65nm 975s (I'll go with the designation for now) come January, with shipment happening in the spring. That's if the transition to 65nm goes smoothly. Still, MS has announced it'll desire a dual-core 5-6GHZ chip for Longhorn with at least 2GB of RAM. I'm not sure Intel can deliver that by 2006, but I think IBM will get Apple there. I'd love it if MS had to scrap Longhorn or scratch a lot of it when they discover Intel can't scale as well as they'd hoped.


The CD drive was a laptop drive, the chip was the same, the board was almost small enough and shaped right. I've got a broken (literally) iBook Rev A and a dead iMac Rev A... Maybe I'll dig them out and compare again.

If a redesigned eMac goes with a slot-load, I can see them sharing Combo-drives across the eMac, iBook, and PowerBook lines. The same won't go for Superdrives though - not if Apple wants to keep the high speed burner now in the eMac.