970 Production info

[Transcendental Octothorpe]

Hi all,

I recieved some info from someone working on the 970 at IBM a while back. EDIT: to clarify, the info came to me a while back. The source was currently working on the 970.
Here is the jist:

-Design phase complete
-970 to start production in March
-Production to begin at 0.13um
-East Fishkill line isn't up and running well enough for it to start production there. When it is, however, the 970 will move there, and probably to 0.09nm quickly.

Obviously the key point for us is the March production date.

Just thought you'd all like to know :cool:

Can anyone else comfirm or deny this info?

[ 01-06-2003: Message edited by: Transcendental Octothorpe ]</p>

[Lemon Bon Bon]

SMOKIIIIN'!

Lemon Bon Bon :cool:

[T'hain Esh Kelch]

1st march, or 31st march?

<img src="graemlins/bugeye.gif" border="0" alt="[Skeptical]" />

[sCreeD]

March?
Hm, okay MWNY (The Final Chapter) is too far away. Ditto for WWDC.
Well what do you know!?
National Broadcasters Association in <a href="http://www.nab.org/conventions/nab2003" target="_blank">April</a>!

Screed

[Outsider]

Intoduction tomorrow with a 2 month wait. :)

[the cool gut]

Don't they have to ramp up production? Doesn't that alone take 2-3 months? That would mean they woudn't be at full capacity until June. Looks like there will on of Steve's famous "Available next month" speeches, with actuall delivery in October. <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />

[Programmer]

More likely is that Apple has to accumulate a fair stockpile of processors before beginning production of the Macs that the 970 will go into. I still say July-Sept.

[T'hain Esh Kelch]

[quote]Originally posted by Programmer:
<strong>More likely is that Apple has to accumulate a fair stockpile of processors before beginning production of the Macs that the 970 will go into. I still say July-Sept.</strong><hr></blockquote>
Agreed. These will sell like hot-cakes and Steve knows it.

[Amorph]

They did announce the original 7450 PowerMacs (the 733MHz model) almost exactly when Motorola was ramping up production. With a two month lag, of course.

[drewprops]

You know, it'll be REALLY sad if Apple DOESN'T go with the 970.....we're all salivating at the thought with no assurance that the company is headed in that direction, as obvious a direction as it may seem.

D

[BRussell]

[quote]Originally posted by Amorph:
<strong>They did announce the original 7450 PowerMacs (the 733MHz model) almost exactly when Motorola was ramping up production. With a two month lag, of course.</strong><hr></blockquote>But the 667 (also a 7450) was also announced at the same time, and was available sooner than 2 months. They also introduced the 7410 processor, which was available immediately in the low-end PowerMacs and the new Powerbook. That was the best keynote in memory.

[Crouton]

[quote]Originally posted by drewprops:
<strong>You know, it'll be REALLY sad if Apple DOESN'T go with the 970.....we're all salivating at the thought with no assurance that the company is headed in that direction, as obvious a direction as it may seem.

D</strong><hr></blockquote>

Yeah, just like when everyone thought for SURE that Apple was going to buy Be and use that OS for the basis of it's new OS. I sure remember thinking that back then. It was a shocker when they said that they had bought Next!

Who knows, maybe that will happen again and Apple will announce that they are switching over to the VIA C3 processor!
<img src="graemlins/lol.gif" border="0" alt="[Laughing]" />

[whoami]

so how crazy would it be for the 970 to be announced tomorrow? i think they understand that the 970 is expected by everyone in the know! we'll see tomorrow!

[Matsu]

March? Hmmm... Apple has had a long pre-order lead time in the past, and it may just happen again. Remember way back when (around IBM 970 microprocessor forum time) we were all trying to pin down a realistic PPC970 debut. At that time I said that if there were any plausible way that 970's would be ready within 4 months of MWSF, we'd see it demo'd and Steve would take pre-orders for it. It took Apple nearly to April to get FP iMacs into full availability (of all models) and we already know that 1.25Ghz DP G4's WILL still be available to customers after MWSF. So that's one of three models. It's not like they'll have nothing to sell pro customers as the DP1.25 migtrates to the bottom of the line-up. In the inverse of the iMac situation, you'll have to wait for the high-end machines, but Steve will show you how great they are in a bake-off. And like all bake-offs they'll be a debateable element about it -- probably he'll show it smoking a currently available top line P4, even though the 970 won't be anyone's hands untill April and by that time X86 will have something faster on offer. No matter, any pros with a significant Mac investment will have heard all they need to hear, they'll put down their money and wait, and Apple will have stemmed the tide of Pro Switchers (to the wintel camp).

It really fits Apple's pattern of selling you today what isn't quite available yet, and comparing future performance to current X86.

The wait is on and PM sales are all but dead. It isn't just the mac-web that's grown impatient, TV analysts have too -- not in the typical Apple's in trouble way, but rather more than once now I've read and seen main stream news asking very specifically about the status of the next processor and the "aging" G4. Mild G4 bumps won't bring the pro customers back for the top rung machines, a 1.25Ghz entry PM would sell like gangbusters for the 2-4 months in between (if priced like the DP867) and won't infringe on 3000+ PPC970 preorders. (two different customers in that range)

I think we may see (and order) if not touch a PPC970 sooner rather than later. Customers who would pay top line PM prices will wait if they're gtting something competitive.

We've heard that quite possibly Apple is sitting on iMacs for another couple of weeks to clear out inventory, I hope they eat it and choke on it, they have to have realized that machines only become less saleable as time passes.

[kcmac]

In this economy, wouldn't pre-orders be a good thing for Apple? They do it anyway, why not really do it this time! :)

[Transcendental Octothorpe]

[quote]Originally posted by drewprops:
<strong>You know, it'll be REALLY sad if Apple DOESN'T go with the 970.....we're all salivating at the thought with no assurance that the company is headed in that direction, as obvious a direction as it may seem.

D</strong><hr></blockquote>

Apple will use this chip.

Oh, and I don't know if it's the 1st or 31st of March.

[ 01-06-2003: Message edited by: Transcendental Octothorpe ]</p>

[Outsider]

Apple can use the HYPE of the 970 to sell G4s. Here's how: Introduce a new line up of 4 PowerMacs models, the low-end using 7457 processors in speeds of 1.2 and 1.4GHz with 200MHz buses. The other 2 high-end models are 970 machines due in March running at 1.6 and 1.8GHz with 800 and 900MHz buses respectively. They hype the 970 machines during the expo but they will have the 7457 machines in there available immediately to quench their thirst.

[Producer]

The question is how expensive will these chips be at first? Also just curious if it was your decision (whoever is reading this) and assuming that the software problems would be minumal....would you choose AMD hammer or the IBM 970 for Apple?

Also an interesting article about Sony using Rambus for there new playstion which will contain the IBM Cell chip...I know programmer is a big fan of Rambus technology...the article states that cell will scale down to consumer electronics and up to high end computing...if it scales up then why wouldn't this chip be viable for Apple? Plus Apple will be more and more into consumer electronics themselves by then and programming should be similar to powerpc...

I still say but too tired to go into the many reasons...Apple should port OS X to the Playstation III and sell a cheap computer upgrade kit for it...it's the only way Apple will be able to take over the lowend computing segment and increase market share by a significant amount...

[Transcendental Octothorpe]

I have new info that I, unfortunately can't share here, but suffice it to say:

:D

Maybe I'll be able to share a little bit more info as time goes on.

Oh, anyone know what a VSP interface is?

[Lemon Bon Bon]

[quote]I have new info that I, unfortunately can't share here, but suffice it to say:


<hr></blockquote>

I've gone weak at the knees.

HOTDOG!!!

Lemon Bon Bon <img src="graemlins/cancer.gif" border="0" alt="[cancer]" />

[KidRed]

[quote]Originally posted by Transcendental Octothorpe:
<strong>I have new info that I, unfortunately can't share here, but suffice it to say:

:D

Maybe I'll be able to share a little bit more info as time goes on.

Oh, anyone know what a VSP interface is?</strong><hr></blockquote>

Ok, you can't say but can you hint at what is new compared to your previous statement? Like you said production in march, is that still right?

[Nevyn]

[quote]Originally posted by Producer:
<strong>The question is how expensive will these chips be at first? Also just curious if it was your decision (whoever is reading this) and assuming that the software problems would be minumal....would you choose AMD hammer or the IBM 970 for Apple?</strong><hr></blockquote>

Unless I had 1) a universal translator with a 0% speed penalty, 2) very strong support from all of my major developers, 3) comittments from AMD backed by stiff 'failure to produce' clauses, I'd pick IBM.

1) Is as likely that you'll solve Fermat's last theorem on your napkin.
2) Is as likely as hurricanes in North Dakota. (Hey Quark, get a move on.)
3) AMD probably could put a compelling case forward, but who cares.

[Rhumgod]

[quote]Originally posted by Transcendental Octothorpe:
<strong>Oh, anyone know what a VSP interface is?</strong><hr></blockquote>

Sure...virtual single processor. A way of distributing an app across multiple processors so that the app thinks it's a single processor. No need for SMP coded apps, that way. <a href="http://www.windriver.com/products/vspworks/index.html" target="_blank">Wind River</a> does a bit of this, and I know IBM and Wind River are quite friendly ;)

[709]

[quote]Originally posted by Rhumgod:
<strong>

Sure...virtual single processor. A way of distributing an app across multiple processors so that the app thinks it's a single processor. No need for SMP coded apps, that way. <a href="http://www.windriver.com/products/vspworks/index.html" target="_blank">Wind River</a> does a bit of this, and I know IBM and Wind River are quite friendly ;) </strong><hr></blockquote>

Nice. Maybe Adobe asked for this feature? :)

[rickag]

[quote]Originally posted by Transcendental Octothorpe:
<strong>I have new info that I, unfortunately can't share here, but suffice it to say:

:D

Maybe I'll be able to share a little bit more info as time goes on.

Oh, anyone know what a VSP interface is?</strong><hr></blockquote>

Virtual Serial Port. Do I win a prize.

Based on your previous post I'm would expect the 970 to already be in production and the newly introduced PowerMac's with 4-6 week ship dates will actually contain the 970.
<img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" />

Cynicism aside, this would mean production begining early part of March. Hope so. :)

[serrano]

[quote]Originally posted by Rhumgod:
<strong>

Sure...virtual single processor. A way of distributing an app across multiple processors so that the app thinks it's a single processor.</strong><hr></blockquote>

sweeeeeeeeeet

[KidRed]

So which is it? Serial port or single processor?

[rickag]

[quote]Originally posted by KidRed:
<strong>So which is it? Serial port or single processor?</strong><hr></blockquote>

Both
:)

though i think virtual single processor is more current news :D

[Transcendental Octothorpe]

[quote]Originally posted by rickag:
<strong>

Both
:)

though i think virtual single processor is more current news :D </strong><hr></blockquote>

Yeah, I looked around a bit before asking, and what I was wondering about deffinately isn't a serial port.

Oh, and :D wan't really regarding time frame, although that hasn't changed really.

I don't think I should say anything more than:

"We'll all be very happy when the 970-based PM comes out."

(well, as long as it doesn't cost a fortune. I don't have any info there.)

Sorry to be such a tease, but I don't want anyone to get in trouble. I probably shouldn't have said anything, as all I've done to you folks is "wake your appitite without bedin' her back down". :(

[ 01-31-2003: Message edited by: Transcendental Octothorpe ]</p>

[sCreeD]

Hm, would multiple cores "act" much like multiple processors? Hm...

:D

Screed

[Nitzer]

Video Sound Processor

<img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />

[Rhumgod]

[quote]Originally posted by sCreeD:
<strong>Hm, would multiple cores "act" much like multiple processors?</strong><hr></blockquote>

970's aren't multi-core.

[quote]Originally posted by Octothorpe:
<strong>I don't think I should say anything more than:

"We'll all be very happy when the 970-based PM comes out."</strong><hr></blockquote>

Gee, do ya think ?

:D

[Lemon Bon Bon]

So, an interface to allow two cpus to serve as one?

Who needs hyperthreading when you've got this?

Maybe PPC could deliver on its promise to be twice as fast at half the cost?

Dual 1.8 acting as a 3.6 Gigger! Acting as a 7.2 gig G4

:eek: :eek: :eek:

Adobe no need to do multiprocessor code anymore?

Golly.

Intel 'blow away' the 970..?

Sounds like the full details (the fine print) of the Apple/IBM 970 implementation are yet to be revealed.

Ace up the sleeve?

Lemon Bon Bon <img src="graemlins/cancer.gif" border="0" alt="[cancer]" />

[rickag]

That leaves yields and speeds.

Exceeding the projected 1.8GHz.

Wish I could believe any of this. I'm still expecting the MPC7457 mid summer and the 970 very very late 2003 or early 2004.

message to IBM - suprise me, please. :)

[Rhumgod]

Just did a little searching...turns out a company called <a href="http://www.transtech-dsp.com/powerpc/vqg4.htm" target="_blank">Transtech</a> makes quad G4 boards for serious applications. Apparently, they already have a <a href="http://www.transtech-dsp.com/software/vsipl.htm" target="_blank">VSP</a> Interface or VSIPL working through VxWorks. Cool stuff - didn't really know anyone was doing this.

[cowerd]

[quote]Adobe no need to do multiprocessor code anymore?<hr></blockquote>When did they start?

[709]

[quote]Originally posted by cowerd:
<strong>When did they start?</strong><hr></blockquote>

Exactly.

[KidRed]

[quote]Originally posted by rickag:
<strong>That leaves yields and speeds.

Exceeding the projected 1.8GHz.

Wish I could believe any of this. I'm still expecting the MPC7457 mid summer and the 970 very very late 2003 or early 2004.

message to IBM - suprise me, please. :) </strong><hr></blockquote>

Well we won't get 3 updates in one year, so if we don't get the 970 this fall (which I doubt) we wouldn't get it until the end of next January.

So Transcendental Octothorpe-
You're saying without a doubt we will get the 970? So if production is in March, how long before they are shippable?

[O and A]

[quote]Originally posted by sCreeD:
<strong>March?
Hm, okay MWNY (The Final Chapter) is too far away. Ditto for WWDC.
Well what do you know!?
National Broadcasters Association in <a href="http://www.nab.org/conventions/nab2003" target="_blank">April</a>!

Screed</strong><hr></blockquote>

Your logic is beyond flawed

As said if march production is made than ur looking at mwny at earliest and thats being a touch optimisitc with a little more realism albeit.

If history is any indication ur at least 6 montsh away from the next powermac update.

[ 01-31-2003: Message edited by: O and A ]</p>

[Transcendental Octothorpe]

Well, I have no info from Apple. Just IBM and Mot.

All I know is when Apple is supposed to get chips, and how long they've had to design with them.

It is my opinion (seriously, all my info could be wrong or change) that the next PM update will include a 970, and that it will happen before fall.

Maybe my judgement is clouded, as I (like the rest of us) really want a 970 PM. See my sig. It's the one thing that will finally make me drop the couple grand.

[ 01-31-2003: Message edited by: Transcendental Octothorpe ]</p>

[O and A]

[quote]Originally posted by Outsider:
<strong>Apple can use the HYPE of the 970 to sell G4s. Here's how: Introduce a new line up of 4 PowerMacs models, the low-end using 7457 processors in speeds of 1.2 and 1.4GHz with 200MHz buses. The other 2 high-end models are 970 machines due in March running at 1.6 and 1.8GHz with 800 and 900MHz buses respectively. They hype the 970 machines during the expo but they will have the 7457 machines in there available immediately to quench their thirst.</strong><hr></blockquote>

<img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" /> <img src="graemlins/bugeye.gif" border="0" alt="[Skeptical]" /> <img src="graemlins/lol.gif" border="0" alt="[Laughing]" /> :rolleyes: <img src="graemlins/bugeye.gif" border="0" alt="[Skeptical]" />

[O and A]

[quote]Originally posted by Transcendental Octothorpe:
<strong>Well, I have not info from Apple. Just IBM and Mot.

All I know is when Apple is supposed to get chips, and how long they've had to design with them.

It is my opinion (seriously, all my info could be wrong or change) that the next PM update will include a 970, and that it will happen before fall.

Maybe my judgement is clouded, as I (like the rest of us) really want a 970 PM. See my sig. It the one thing that will finally make me drop the couple grand.</strong><hr></blockquote>

I know what you mean. The 970 will really give me a reason to get a powermac. I can't imagine what os x would be like on that computer not to mention the render times for 3d graphics.

[Algol]

Yea, I had a little inside info a while back, and I can tell you everything I heard on the subject. But you guys already know it all. The 970 is coming this year and it will kick Intel's little fat ass all the way to china town! hehe ;) Any doubters can shove it. ;)

[ 01-31-2003: Message edited by: Algol ]</p>

[O and A]

[quote]Originally posted by Algol:
<strong>Yea, I had a little inside info a while back, and I can tell you everything I heard on the subject. But you guys already know it all. The 970 is coming this year and it will kick Intel's little fat ass all the way to china town! hehe ;) Any doubters can shove it. ;)

[ 01-31-2003: Message edited by: Algol ]</strong><hr></blockquote>

Heres to hoping apple actually comes through with our expectations.

[Algol]

The minute I see PPC970 on apple's site is the minute I hit BUY NOW. I'm just wondering if I should go the the 23" one 20 inch LCD. I can't wait much longer COME ON APPLE!!!

You know it would be funny if apple just did a quite update over night and added the 970 to the Powermacs and didn't say a word. We would all be trying to figure out what was going on. lol!

[Programmer]

So your new news says to me that the 970 has successfully reached sampling and the first samples have gone into testing and everything is looking good. Really good. Remember that IBM estimated SPECmarks for the 970. They can't bump the clock rate too much without increasing the power consumption and forcing the bus speeds higher (which could limit how much Apple could get out of the higher speeds), but at the specified clock rates the performance could turn out to be higher than the predictions. The timing is about right for this kind of info to start surfacing.

March production ramp plus time to accumulate enough stock and put the new PowerMacs into production says to me that an August - September timeframe is still just about right. I'll bet that we see a developer preview at WWDC and perhaps the rollout of the 64-bit version of MacOS X.

[Algol]

I WANT ONE NOW! ;)

Well Programmer is most likely right about not seeing the 970 till september. Well my birthday is in September so that would be fine. I'll get myself a nice present ( about time someone did).

But if apple has been working closely with IBM could we not see a 970 in June? I really hope we do.

[Programmer]

[quote]Originally posted by Algol:
<strong>I WANT ONE NOW! ;)

Well Programmer is most likely right about not seeing the 970 till september. Well my birthday is in September so that would be fine. I'll get myself a nice present ( about time someone did).

But if apple has been working closely with IBM could we not see a 970 in June? I really hope we do.</strong><hr></blockquote>

You can hope for whatever you want, but that doesn't make it any more likely. :)

[Algol]

Heck we'll have to wait till september just to get our iMac update. ;)

[moki]

[quote]Originally posted by sCreeD:
<strong>Hm, would multiple cores "act" much like multiple processors? Hm...</strong><hr></blockquote>

Depends on how it is architected; the way the dual cores work in the Power4, they appear as one processor.

[moki]

[quote]Originally posted by Algol:
<strong>Well Programmer is most likely right about not seeing the 970 till september. Well my birthday is in September so that would be fine. I'll get myself a nice present ( about time someone did).

But if apple has been working closely with IBM could we not see a 970 in June? I really hope we do.</strong><hr></blockquote>

Knowing what is a lot easier than knowing when. Even engineers working on such projects don't know exactly when they'll be done.

[gar]

[quote]Originally posted by Algol:
<strong>You know it would be funny if apple just did a quite update over night and added the 970 to the Powermacs and didn't say a word. We would all be trying to figure out what was going on. lol!</strong><hr></blockquote>

there is no over night on the internet. the world is bigger than the north american continent. for instance in europe are more people connected to the internet than in america. so most people will know about this and you are the last one to know because you were sleeping. and that would be very bad

[moki]

[quote]Originally posted by gar:
<strong>

there is no over night on the internet. the world is bigger than the north american continent. for instance in europe are more people connected to the internet than in america. so most people will know about this and you are the last one to know because you were sleeping. and that would be very bad</strong><hr></blockquote>

For broadband, at least, this is not true:

<a href="http://cyberatlas.internet.com/markets/broadband/article/0,,10099_783401,00.html" target="_blank">http://cyberatlas.internet.com/markets/broadband/article/0,,10099_783401,00.html</a>

...do you have any stats on how many people are online in the EU?

[Programmer]

[quote]Originally posted by moki:
<strong>
Depends on how it is architected; the way the dual cores work in the Power4, they appear as one processor.</strong><hr></blockquote>

Do you have an technical URL references about this technology, Moki? I'm very curious at what level these "appear as one processor".


[EDIT] Nevermind I went and looked at the Wind River stuff. Careful what talking about how it makes "thing look like a single processor". According to my definition of what a single processor looks like, it does not do this. To take advantage of multiple processors the application must still be written as multiple threads or tasks. This means it doesn't look at more like a single processor than the current Apple duals, except that resource sharing between the processors might be somewhat improved. As I read it the intention of this system is to support NUMA systems and clusters, making them seem like they are traditional SMP designs. Multiple processors still can't speed up a single thread, but multiple threads can be spread across disparate processors.

[ 02-02-2003: Message edited by: Programmer ]</p>

[moki]

[quote]Originally posted by Programmer:
<strong>

Do you have an technical URL references about this technology, Moki? I'm very curious at what level these "appear as one processor".</strong><hr></blockquote>

This is good for a start, I think: <a href="http://www.digit-life.com/articles/ibmpower4/" target="_blank">http://www.digit-life.com/articles/ibmpower4/</a>

[Programmer]

[quote]Originally posted by moki:
<strong>
This is good for a start, I think: <a href="http://www.digit-life.com/articles/ibmpower4/" target="_blank">http://www.digit-life.com/articles/ibmpower4/</a></strong><hr></blockquote>

Interesting article, albeit a little old. Almost the last line, however, is quite interesting:

[quote]
Will SMT be integrated into personal POWER4 processors?
<hr></blockquote>

Yes, I've been wondering that too. :)

[ 02-02-2003: Message edited by: Programmer ]</p>

[rogue27]

On the timeline subject...

Back in August or September, moki said Apple got a shipment of 1.4Ghz CPUs from Motorola. It took 4 months for those processors to actually show up in new computers. With that in mind, I would expect it to take between 2 and 6 months after "March" for there to be PPC 970 Macs available. So between May and September sounds realistic to me. October or November are even feasible, since this is all new so it may take more time to test and refine such machines.

I wouldn't be surprised if there was a minor PowerMac bump in the spring that puts the 7457 into at least the high end power mac and drops variants of the current mid and high models down to low and mid.

When the 970 does become available, I don't expect there to be some PowerMacs using the 970 and some using the G4 at the same time. Once the 970 is in PowerMacs, I think the G4 will be used solely in laptops and consumer products like the iMac.

[rickag]

[quote]Originally posted by rogue27:
<strong>...I think the G4 will be used solely in laptops and consumer products like the iMac.</strong><hr></blockquote>

If Motorola ever produces a cpu on the 0.13µm process.

[Outsider]

[quote]Originally posted by Programmer:
<strong>So your new news says to me that the 970 has successfully reached sampling and the first samples have gone into testing and everything is looking good. Really good. Remember that IBM estimated SPECmarks for the 970. They can't bump the clock rate too much without increasing the power consumption and forcing the bus speeds higher (which could limit how much Apple could get out of the higher speeds), but at the specified clock rates the performance could turn out to be higher than the predictions. The timing is about right for this kind of info to start surfacing.

March production ramp plus time to accumulate enough stock and put the new PowerMacs into production says to me that an August - September timeframe is still just about right. I'll bet that we see a developer preview at WWDC and perhaps the rollout of the 64-bit version of MacOS X.</strong><hr></blockquote>
You don't think there is a variable that can allow for 3x and 4x ratios in bus speed? Maybe even half integer increments of 2.5 and 3.5? I fand it hard to believe they would limit the bus to a static ratio. I could be wrong. By "could be" I mean "probably am". :p

[Programmer]

[quote]Originally posted by Outsider:
<strong>You don't think there is a variable that can allow for 3x and 4x ratios in bus speed? Maybe even half integer increments of 2.5 and 3.5? I fand it hard to believe they would limit the bus to a static ratio. I could be wrong. By "could be" I mean "probably am". :p </strong><hr></blockquote>

I don't know... the IBM presentation was pretty clear that the bus speed is half the processor speed. Period. The other documents I've seen do nothing to dissuade me from this notion. Its also a significantly different bus design than MPX, and the emphasis is clearly on high bandwidth. Since there is no L3 cache support my suspicion is that they aren't interested in being able to cripple the processor's FSB.

Its worth pointing out that the FSB is just a connection from processor(s) to the companion chip, and this will likely be a tightly controlled link of minimal length. Combined with its high clock rate this would tend to make me think that the companion chip will sit on a high quality daughtercard with the processor(s). It remains to be seen where the memory sits, and how the connection to the southbridge on the motherboard is done. In this kind of a setup Apple might be able to support very high clock rates on the FSB (i.e. &gt;1 GHz).

[Transcendental Octothorpe]

[quote]Originally posted by Programmer:
<strong>

...Combined with its high clock rate this would tend to make me think that the companion chip will sit on a high quality daughtercard with the processor(s)processor's FSB....</strong><hr></blockquote>

Yes, from the looks of everything on the new companion chip (can anyone guess the name?), I would think it would have to be extreamly close.

BTW, I think we'll all be pleasently suprised when we see this new chip. :D

[Outsider]

They could position the companion chip directly beneath the CPU on the other side of the PCB. You don't get any closer than that.

[Algol]

[quote]Originally posted by Transcendental Octothorpe:
<strong>

Yes, from the looks of everything on the new companion chip (can anyone guess the name?), I would think it would have to be extreamly close.

BTW, I think we'll all be pleasently suprised when we see this new chip. :D </strong><hr></blockquote>
You know something?

[Algol]

Bump*

<a href="http://www.realworldtech.com/page.cfm?AID=RWT012603224711" target="_blank">http://www.realworldtech.com/page.cfm?AID=RWT012603224711</a>

In it, they discuss IBM's Power4 and, of course, IBM's PowerPC 970 which is coming later this year:


Even more intriguing for Apple is that the 970?s typical power consumption drops to 19 W at 1.2 GHz which makes it a natural competitor to Intel?s Banias processor for high end mobile applications and very small form factor and/or silent desktop PCs. Given the reduced design margin and greater market emphasis for clock frequency of desktop processors, it is also conceivable that IBM could turn out limited numbers of 970 MPUs that clocked at 2 GHz or higher for high end desktop Macs, an important psychological milestone for Apple?s struggle for survival in an increasingly x86 dominated PC world.
While the MHz boost is purely speculation, Apple is aware of their lagging processor speeds (as compared to Intel/AMD), and IBM's PowerPC 970 is widely expected to be incorporated into PowerMacs as early as later this year.


So just some more info on the 970. I think we already know most of what there is to really know as far as when etc. I think it's quite sure we will see one in an Apple this year. But the more info the better...

[ 02-02-2003: Message edited by: Algol ]</p>

[Tomb of the Unknown]

[quote]Originally posted by Programmer:
<strong>As I read it the intention of this system is to support NUMA systems and clusters, making them seem like they are traditional SMP designs. Multiple processors still can't speed up a single thread, but multiple threads can be spread across disparate processors.</strong><hr></blockquote>
Hmmm. Interesting. I wonder how that fits with Apple's supposed XGrid?

<a href="http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=6380998745" target="_blank">Ars reader says XGrid in place at his school.</a>

Are we likely to see the PPC970 in 1U XServes configured as an XGrid cluster this March?

[handygeek]

This would be a great sales booster. However, from a practical standpoint, I'm more enthused about HyperTransport and Infiniband integration. These technologies will open up the biggest bottleneck (other than hard drives) current machines have - the bus. If this all comes together for an announcement at NAB, I'll be on cloud nine. It would make Apple, once again, the center of attention at that show. :)

[DCQ]

Sorry to be a downer. But Steve told us when the PowerMacs would be updated to the 970.

"2003 will be the year of the portable."

He wouldn't say that if he knew that this year would bring a processor to its desktop line that would threaten Intel.

At best, the 970-based PowerMac will debut in September-October. This would give Apple its usual 9-month update cycle, and fit perfectly in with both the Christmas shopping season and corporate fiscal Q1 tech purchase-order plans (and be only a tad late for the education buying season). At absolute-mind-bogglingly-wowzers-best, we'll get another G4 speedbump at MWNY, and the 970 in Sept-Oct.

More likely, we'll get a G4 speedbump in Sept-Oct (1.58GHz, or 1.6 if they hit a 200MHz DDR motherboard), and the 970 at MWSF 04.

I'm looking forward to the 970 too. Not for my own purposes; I'll hopefully be taking the plunge into the laptop world this summer/early fall (if I can find a job by then :( ). I want the 970 to arrive because I think it will be the thing that gets Apple its competitivity back in the professional world, which will give it more viability all over, giving us more applications and hardware. With bigger marketshare, prices can begin to drop in a meaningful way (although I never expect to see a truly affordable Apple :( ). When that happens, Apple will regain "its" education market. Since I am/will be a teacher, I want to see Macs in the classroom again.

Also, a 970-based desktop will allow them to put G4s in the iBook :D . (I know, that won't happen for a while, but I can dream of a $1000 G4, can't I?)

But reality is what it is. 970 production won't begin for another 2 months. And switching to a completely new chip family involves a shitload of headaches. Motherboards will have to be redesigned, software recompiled (I would assume), developers appeased, commercials and marketing and production positioning settled (which, knowing Apple, is probably the most time-comsuming part :rolleyes: ), bugs quashed, and OS compatibilty among 3 processors maintained. And that's probably not the half of it (I'm pretty much the farthest thing from an expect as you can get).

Sure a lot of work has already been done, but the chip barely exists yet. Apple will be smart not to rush this. The last thing they need is a repeat of the original G4 introduction when they discovered the bugger wouldn't work above 500MHz after they announced it.

I know the speculation gets contagious. But--PLEASE--don't make any life-altering decisions based on the premise that the 970 will be here this summer. If you need a high-end box before January, buy a 1.42 dualie now (always best to buy one at the beginning of a product cycle). Otherwise you'll be tearing your hair out when Steve says "One more thing..." at MWNY and finally speedbumps the iMac. <img src="graemlins/lol.gif" border="0" alt="[Laughing]" />

If it gets here early, you'll be delightfully surprised. If not, you'll know that it'll be here sometime between September and January 04.

Why do I say all this? Mostly, I'm sick of jackasses threatening to switch to Wintels and/or kill Steve Jobs (just do it already... :p ) when reality doesn't match their highly overblown expectations.

"The only useful computer is the one you have in front of you."

-DCQ

[Algol]

You better believe that apple has been working with IBM on this since the day they found out. I would imagine that if they start production in march we will see a powermac 970 in another 3 months. Just in time for June.

[Nevyn]

[quote]Originally posted by DCQ:
<strong>But reality is what it is. 970 production won't begin for another 2 months. And switching to a completely new chip family involves a shitload of headaches. Motherboards will have to be redesigned, software recompiled (I would assume), developers appeased, commercials and marketing and production positioning settled (which, knowing Apple, is probably the most time-comsuming part :rolleyes: ), bugs quashed, and OS compatibilty among 3 processors maintained. And that's probably not the half of it</strong><hr></blockquote>

IBM claims 2H03 = wide availability. Yes - headaches. Yes - motherboard design. No - recompile. No - developers to appease. (And SJ pretty much said 'Bite me' to Quark) Pretty sure the pentium-on-snail commercials all done :) Bugs - HW bugs are IBM's problem mostly.

OS compatibility - getting it to _run_ shouldn't be a problem. Getting it to run well shouldn't be a problem. Getting the _full_ benefits might be an issue, but getting it running across the 604, G3, G4, and ppc970 shouldn't be an issue - the differences are primarily in scheduling/ordering instructions, which is in optimization. One guy (outside of Apple) is keeping the 604 & 604e mostly working,

[KidRed]

[quote]Originally posted by DCQ:
<strong>Sorry to be a downer. But Steve told us when the PowerMacs would be updated to the 970.

"2003 will be the year of the portable."
-DCQ</strong><hr></blockquote>

And? That simply means apple's portable will rock this year. Nothing more, nothing less. He didn't say, screw the towers, we're only doing portables this year. That's not to mention the towers were just updated and we now have a 1.4ghz tower. As for your time frame, it's the fall, that's been the consensus since day one. Of course it won't be surprising if we get it around Sept, because that's the fall. Also, the towers aren't on a 9 month update line. From the info I've had, G4s are done. So i don't think we will see anymore G4s in the towers and I don't think we will have to wait till Jan 2004 for the 970.

[ 02-03-2003: Message edited by: KidRed ]</p>

[applenut]

perhaps it will also be the year of the portable because the powerbook will get a PowerPC 970 this year.

hey, ya never know

[agent302]

[quote]Originally posted by applenut:
<strong>perhaps it will also be the year of the portable because the powerbook will get a PowerPC 970 this year.

hey, ya never know</strong><hr></blockquote>

The low voltage 1.2 Ghz PPC970 supposedly only uses 19W...

[T'hain Esh Kelch]

[quote]Originally posted by agent302:
<strong>The low voltage 1.2 Ghz PPC970 supposedly only uses 19W...</strong><hr></blockquote>
Yeah, its really nice, isnt it? ;)

[Guartho]

Does anyone know off the top of their head what wattages the current notebook processors pull? I'd look it up, but I'm lazy ;) and on my way to class.

[Amorph]

[quote]Originally posted by DCQ:
<strong>
But reality is what it is. 970 production won't begin for another 2 months. And switching to a completely new chip family involves a shitload of headaches.</strong><hr></blockquote>

The 970 is a PowerPC, and the PowerPC spec was designed for 64 bit processors from day one.

Hardware will definitely be affected. Software shouldn't even notice that it's not running on one (or two) G4s.

[Programmer]

You're reading way too much into Steve's comment about "the year of the portable". Even if you take it literally it doesn't mean that the tower's aren't going to improve... perhaps the towers will go to the 970 but that is nothing compared to what will happen to them in 2004? The thing to remember is that Steve made that comment at a show when they were introducing a whole line of new portables and he needs those portables to carry as much of Apple's sales until later in the year when the desktop sales can be boosted again. The first 9 months of this year are going to be all about the portables because we know their desktop lineup is keep, and so clearly that makes them important for most of 2003.

My prediction is still that we see the first of the new towers shipping in Aug-Sept. We might see a preview earlier.

[sCreeD]

[quote]Originally posted by Guartho:
<strong>Does anyone know off the top of their head what wattages the current notebook processors pull? I'd look it up, but I'm lazy ;) and on my way to class.</strong><hr></blockquote>

In the neighborhood of 11W to 13W (guesstimate).

So 19 would be a little high, but who knows...

Powerbooks and Powermacs with 970s and the consumer lines all with G4s by the end of the year or MWSF '04 would be lovely.

Screed

[ 02-03-2003: Message edited by: sCreeD ]</p>

[smalM]

[quote]Originally posted by Programmer:
<strong>I don't know... the IBM presentation was pretty clear that the bus speed is half the processor speed. Period.</strong><hr></blockquote>

The presentation said 900 MHz <strong>bit rate</strong>. Are you shure the bus isn't double pumped? This would make a 450 MHz cycle and a bus ratio of 4. I can't remember to have seen that the ratio is fix. Question mark.

[engpjp]

[quote]Originally posted by KidRed:
<strong>

Well we won't get 3 updates in one year, so if we don't get the 970 this fall (which I doubt) we wouldn't get it until the end of next January.
</strong><hr></blockquote>

For all practical purposes, the introduction of the updated G4 and of PPC970 will split the present quad-tiered desktop program into something more confusing... much like the i-/PowerBook has been.

And NOT that early...

engpjp

[Vvmp]

Since the LCD monitor lineup was just updated, we should assume the new 970 Pwrmacs will use the current industrial design or at least the same colors/materials so as to make the hardware match no?

[KidRed]

[quote]Originally posted by engpjp:
<strong>

For all practical purposes, the introduction of the updated G4 and of PPC970 will split the present quad-tiered desktop program into something more confusing... much like the i-/PowerBook has been.

And NOT that early...

engpjp</strong><hr></blockquote>

Not sure about the confusion, but I was told FALL, that's not early to me, it's my INFO.

[neutrino23]

[quote]Originally posted by Guartho:
<strong>Does anyone know off the top of their head what wattages the current notebook processors pull? I'd look it up, but I'm lazy ;) and on my way to class.</strong><hr></blockquote>

According to IBM's spec sheet the G3 750FX uses 3.6 watts @800MHz. That's why the iBooks have great battery life.

[Algol]

redkid is the source that told you fall reliable. Would you bet good money on it? I'm just trying to get a feel as to how sure you are about this fall thing. And since the 970 at 1.2ghz is rather cool running chip would we not see 970 Powerbooks at or around the same time. I don't believe we are going to have a 2 year wait to get a 970 powerbook like we did the G4 powerbook.

[rickag]

[quote]Originally posted by neutrino23:
<strong>

According to IBM's spec sheet the G3 750FX uses 3.6 watts @800MHz. That's why the iBooks have great battery life.</strong><hr></blockquote>

methinks Motorola should contract w/ IBM to produce the MPC7455 on their 0.13µm process...end of rant :(


edit oops, there is no MPC7555, <img src="graemlins/lol.gif" border="0" alt="[Laughing]" />

[ 02-05-2003: Message edited by: rickag ]</p>

[mmicist]

[quote]Originally posted by Programmer:
<strong>

I don't know... the IBM presentation was pretty clear that the bus speed is half the processor speed. Period. The other documents I've seen do nothing to dissuade me from this notion. Its also a significantly different bus design than MPX, and the emphasis is clearly on high bandwidth. Since there is no L3 cache support my suspicion is that they aren't interested in being able to cripple the processor's FSB.

Its worth pointing out that the FSB is just a connection from processor(s) to the companion chip, and this will likely be a tightly controlled link of minimal length. Combined with its high clock rate this would tend to make me think that the companion chip will sit on a high quality daughtercard with the processor(s). It remains to be seen where the memory sits, and how the connection to the southbridge on the motherboard is done. In this kind of a setup Apple might be able to support very high clock rates on the FSB (i.e. &gt;1 GHz).</strong><hr></blockquote>

The front side bus works at integer dividers of the clock frequency, and is double pumped, the 1.8GHz has a 4:1 divider, giving 900MHz effective data rate.

see <a href="http://www.realworldtech.com/page.cfm?AID=RWT101502203725&p=2" target="_blank">David Wang's article</a>

The bus is also wave-pipelined, which means that more than one bit of data can be on the line at once, so that line delays can be more than one cycle long. The bus is also unidirectional, which means you can put the next request on the bus without having to wait for the bus to turn around and send the requested data back. The bus is actually well suited to long distances between the CPU and the companion chip. That being said, laying out the board would be easier, and latencies would be reduced, if the two were close.

michael

[Programmer]

Err, yeah I was ignoring the double-pumped nature of the FSB just for simplicity and calling it 900 MHz as seems to be standard practice these days. So yes, the clock rate multiplier is actually 4:1.

I'd been refering to a 3rd party article about the 970 so I went back to the MDF presentation and checked... it doesn't talk about mulitpliers but it does say "Up to 900MHz bitrate". Hmmm, I guess it does support slower buses. I really hope this is not an option Apple chooses to use.

My supposition about a small daughtercard arrangement is just speculation based on the idea that PCBs for 900 MHz buses are expensive, and this is the portion that will differ between single, dual, and quad processor machines. If all that will live on these cards is the processor(s) and companion chip, then they can be very small to minimize cost and latency. The connection to the main motherboard will be interesting as it will probably have to carry both the southbridge connection and the memory bus... unless the memory lives on the daughtercard.

I can't wait to see these new machines, it'll be the first time in a while that Apple really has had a chance to innovate on the motherboard. Hopefully they do it well.

[BRussell]

[quote]Originally posted by Programmer:
<strong> it doesn't talk about mulitpliers but it does say "Up to 900MHz bitrate". Hmmm, I guess it does support slower buses. I really hope this is not an option Apple chooses to use.</strong><hr></blockquote>They're probably just referring to the slower FSB speeds for the slower processors (&lt;1.8Ghz).

[Programmer]

[quote]Originally posted by BRussell:
<strong>They're probably just referring to the slower FSB speeds for the slower processors (&lt;1.8Ghz).</strong><hr></blockquote>

That had been my original thought, but the implication from the people quoting Mr Wang's article is otherwise. Unfortunately I can't get to that article at the moment for some reason.

[Outsider]

I was curious if they could implement a 5:1 ratio. This would allow a 2.25GHz Machine to still retain it's 900MHz bus while being more flexible with configurations. While 1GHz buses are not out of the question, it may have obstacles we are not away of.

[rickag]

[quote]Originally posted by Programmer:
<strong>....If all that will live on these cards is the processor(s) and companion chip, then they can be very small to minimize cost and latency.....</strong><hr></blockquote>

After reading the Ars Technica article and some of the information available on the web, I'm still confused over what exactly the "companion chip" does. I would have thought it was some kind of controller chip?? Or is it just some kind of interface to be connected to the controller chips, if so what other possible uses beyond just communication could/would be included on the companion chip(re: odd name)

[ 02-04-2003: Message edited by: rickag ]</p>

[Programmer]

[quote]Originally posted by rickag:
<strong>After reading the Ars Technica article and some of the information available on the web, I'm still confused over what exactly the "companion chip" does. I would have thought it was some kind of controller chip?? Or is it just some kind of interface to be connected to the controller chips, if so what other possible uses beyond just communication could/would be included on the companion chip(re: some odd name)</strong><hr></blockquote>

The companion chip is just the thing on the opposite end of the FSB. Given Apple's typical tightly integrated designs I would guess that we'll see their companion chip having 2-4 FSB ports, a memory controller, and some kind of a connection to the southbridge on the motherboard. This means "companion chip == northbridge".

[KidRed]

[quote]Originally posted by Algol:
<strong>redkid is the source that told you fall reliable. Would you bet good money on it? I'm just trying to get a feel as to how sure you are about this fall thing. And since the 970 at 1.2ghz is rather cool running chip would we not see 970 Powerbooks at or around the same time. I don't believe we are going to have a 2 year wait to get a 970 powerbook like we did the G4 powerbook.</strong><hr></blockquote>

Hard to say exactly. I feel good about it. I was told about an IBM chip replacing moto released in the fall of 2003 last year before IBM announced anything. So the source was right about a lot of things and when the press release said 2H of 2003 and other sources have said the fall, I just feel strongly about it based on circumstances not more evidence.

[rickag]

Programmer

Thanks, I just started wondering because I had never heard the name companion chip before, causing wild unfounded speculation racing through my head concerning other possible functions that may be rolled into the companion chip.

[Transcendental Octothorpe]

[quote]Originally posted by Programmer:
<strong>
This means "companion chip == northbridge".</strong><hr></blockquote>

Then Uninorth, then U2, and then with the 970...

Oh, and other than a few supprises, Programmer has the content of the chip pretty close.

[rickag]

[quote]Originally posted by Transcendental Octothorpe:
<strong>

Then Uninorth, then U2, and then with the 970...</strong><hr></blockquote>

I must be dense, I do not understand your post. Could you explain it for me? Bear in mind I have very very little technical knowledge.

[Ompus]

Slow-tech folks, like myself, might find <a href="http://www.cpuplanet.com/features/article.php/1490831" target="_blank">this article</a> about basic motherboard layout informative.

[ 02-04-2003: Message edited by: Ompus ]</p>

[rickag]

[quote]Originally posted by Ompus:
<strong>Slow-tech folks, like myself, might find <a href="http://www.cpuplanet.com/features/article.php/1490831" target="_blank">this article</a> about basic motherboard layout informative.</strong><hr></blockquote>

Okey Dokey. I thought I had a rudimentary understanding of the cpu - northbridge - southbridge layout, but what the heck. I read the article you linked to, twice, very nice article. :) Thank you.

But regretably no mention of Uninorth nor U2(re: nice band, used to listen to them in the 80's) I believe Apple computers use a bridge chip called Uninorth, right. But what's U2?? A chip for USB 2.0???

[Programmer]

[quote]Originally posted by Transcendental Octothorpe:
<strong>

Then Uninorth, then U2, and then with the 970...

Oh, and other than a few supprises, Programmer has the content of the chip pretty close.</strong><hr></blockquote>

It'll also have a DMA engine and perhaps the DSP-like features that were talked about last year. We can hope for a direct AGP 8x port as well. Is this what you are refering to TO?

[RBR]

It'll also have a DMA engine and perhaps the DSP-like features that were talked about last year. We can hope for a direct AGP 8x port as well. Is this what you are refering to TO?[/QB][/QUOTE]
___________________________________________

Did you know that the AGP 8X spec provides for a *second* AGP 8X slot? Interesting possibilities.

[Nordstrodamus]

Forgive me if this has been asked and answered, but what is the shortest time between upgrades that Apple has ever done?

Apple just updated the powermacs, but I can't imagine that they wouldn't use the 970 as soon as it was available. At least by putting it in the high end only.

[Leonis]

[quote]Originally posted by Nordstrodamus:
<strong>Forgive me if this has been asked and answered, but what is the shortest time between upgrades that Apple has ever done?

Apple just updated the powermacs, but I can't imagine that they wouldn't use the 970 as soon as it was available. At least by putting it in the high end only.</strong><hr></blockquote>


Upgrade cycle on PM usually is around 6-8 months.

[mooseman]

...seems iteresting that while the 970 is garnering the lion's share of attention, the fact that a 1.42GHz G4 came out on a 180nm process has escaped mention.

Additonally, just last summer, Philips, Motorola, and STMicroelecronics announced they were preparing a facility capable of churning out 90 nanometer microprocessors on 300mm silicon....by the end of 2002. So if preproduction silicon is coming out by the end of 2002, whats keeping Moto from introducing a 2+GHz G4 on a 90nm process 2Q '03?

Just some thoughts. Any opinions on it?

[Transcendental Octothorpe]

[quote]Originally posted by Nordstrodamus:
[QB]Forgive me if this has been asked and answered, but what is the shortest time between upgrades that Apple has ever done?
[QB]<hr></blockquote>

I think that the PM G4 came only three months or so of the BW PM G3. <a href="http://www.apple-history.com" target="_blank">www.apple-history.com</a> is slower than dirt right now, or I'd check for sure.

[quote] It'll also have a DMA engine and perhaps the DSP-like features that were talked about last year. We can hope for a direct AGP 8x port as well. Is this what you are refering to TO? <hr></blockquote>

Well, those wouldn't be suprises, would they? Well, maybe the DSP.

[Amorph]

The Wallstreet PDQ series was rushed out within three months of the Wallstreet. I think that's a record. It wasn't an Earth-shattering upgrade, but it was needed.

[quote]Originally posted by mooseman:
<strong>...seems iteresting that while the 970 is garnering the lion's share of attention, the fact that a 1.42GHz G4 came out on a 180nm process has escaped mention.

Additonally, just last summer, Philips, Motorola, and STMicroelecronics announced they were preparing a facility capable of churning out 90 nanometer microprocessors on 300mm silicon....by the end of 2002. So if preproduction silicon is coming out by the end of 2002, whats keeping Moto from introducing a 2+GHz G4 on a 90nm process 2Q '03?

Just some thoughts. Any opinions on it?</strong><hr></blockquote>

I've been talking about that for a while now. ;) The STM facility in Grenoble is a potential ace in the hole for Mot. They've said that since the issues involved in designing for .13&mu; and .09&mu; are essentially the same, they might as well leap all the way down to .09&mu;. Since it's not their fab, it shouldn't have the problems that their own .13&mu; facility has been having due to draconian cost-cutting measures.

If Mot actually pulls this off this year, they'll suddenly have an attractive little chip. Perfect for iBooks. ;) And if they feel like bringing back the ballyhooed Eleven on that process, it might suddenly run cool enough not to explode in Apple's test mules. I'm not expecting an answer to the 970 out of Mot any time soon, though, but I wouldn't turn down a tiny little G4 on a really fast MaxBus.

But this line of discussion is more appropriate for the 7447/7457 thread next door.

[ 02-04-2003: Message edited by: Amorph ]</p>

[Bigc]

Well, with IBM getting into the desktop market MOT may be changing their tune with respect to processor design. They may be worried that IBM will make them look bad in the embedded market. Lets hope for some good-ol' Intel/AMD battling on the CPU front. Bragging rights sometimes drive markets ;)

Apple (and, therefore, we) have nothing to lose.

[ 02-04-2003: Message edited by: Bigc ]</p>

[boots]

[quote]Originally posted by mooseman:
<strong>...seems iteresting that while the 970 is garnering the lion's share of attention, the fact that a 1.42GHz G4 came out on a 180nm process has escaped mention.

Additonally, just last summer, Philips, Motorola, and STMicroelecronics announced they were preparing a facility capable of churning out 90 nanometer microprocessors on 300mm silicon....by the end of 2002. So if preproduction silicon is coming out by the end of 2002, whats keeping Moto from introducing a 2+GHz G4 on a 90nm process 2Q '03?

Just some thoughts. Any opinions on it?</strong><hr></blockquote>

Make it 1.6, 1.8, 2.0, 500 GHz...

if they don't change the FSB, who cares.

[ 02-04-2003: Message edited by: boots ]</p>

[Programmer]

[quote]Originally posted by boots:
<strong>

Make it 1.6, 1.8, 2.0, 500 GHz...

if they don't change the FSB, who cares.

</strong><hr></blockquote>


You're saying you wouldn't take a 5 watt 1.5 GHz G4+ w/ 512K L2 cache? The FSB is important folks, but its not everything. And the 7457-RM is always a possibility for early 2004 according to Moto's roadmap.

[DrBoar]

Faster G4s are kind of good as it means faster Macs and faster upgrades. But as far as desktop CPUs the G4 is dead end and speed boosts only extend the lenght of the dead end. Sure a 2 GHz G4 in a powerbook would be fantastic but given Motorolas track record the last 10 years, is it likely that they will do that in a resonable time frame?

[hardcore]

[quote]Originally posted by Programmer:
<strong>
You're saying you wouldn't take a 5 watt 1.5 GHz G4+ w/ 512K L2 cache? The FSB is important folks, but its not everything. And the 7457-RM is always a possibility for early 2004 according to Moto's roadmap.</strong><hr></blockquote>

I don't know about boots, but I wouldn't. Really, the CPU has been fast enough for me for a while now. What I can't stand is the overall system architecture. slow disks, slow FSB, slow memory, slow PCI, slow AGP, mediocre built-in sound.

I'm a big fan of fast and wide.

having said that I don't think my B&W can hold out much longer. As soon as the 9700 Pro becomes available I'll pick up a new machine and then sell it when something I like comes out.

[rickag]

[quote]Originally posted by mooseman:
...Additonally, just last summer, Philips, Motorola, and STMicroelecronics announced they were preparing a facility capable of churning out 90 nanometer microprocessors on 300mm silicon....by the end of 2002. So if preproduction silicon is coming out by the end of 2002, whats keeping Moto from introducing a 2+GHz G4 on a 90nm process 2Q '03?

...Any opinions on it?<hr></blockquote>

I remember those press releases. I personally got excited. They mentioned something like high end processors ?? manufactured by the end of 2002.

My opinion now, press release bologna. Suprise me Motorola, let's have another press release announcing ANYTHING is being manufactured using a 0.09µm process - ANYTHING AT ALL.


edit: originally wrote 0.9µm process above. Ha, I actually showed Motorola going backwards.

[ 02-07-2003: Message edited by: rickag ]</p>

[Addison]

It's clear that next year the whole range will be unrecognisable in terms of performance, and these boards will go quite quiet. However in the mean time...

I think that we are not seeing the real performane of the current processors because of the bottlekneck caused by the FSB. If we can get a decent FSB then the performance on current machines and processors will leap.

I doubt that Moto would make the current processors on the 0.09 process, I am sure they are fully aware of the shortcommings too. Any move to 0.09 would be accompanied with a revised FSB.

[Bancho]

[quote]Originally posted by Addison:
<strong>It's clear that next year the whole range will be unrecognisable in terms of performance, and these boards will go quite quiet. However in the mean time...

</strong><hr></blockquote>

Are you saying that we won't find anything to bitch about? Is that some sort of challenge?

:D

[Addison]

[quote]Originally posted by Bancho:
<strong>

Are you saying that we won't find anything to bitch about? Is that some sort of challenge?

:D </strong><hr></blockquote>

Well what's wrong with being an optimist.

[Bancho]

<img src="graemlins/lol.gif" border="0" alt="[Laughing]" /> Watch what happens if Apple releases these 970 mahines and it turns out they are several orders of magnitude faster than todays machines. There will be a banshee wail of screaming/bitching/moaning that Apple has screwed us all again by making our current machines look pathetic in comparison. <img src="graemlins/lol.gif" border="0" alt="[Laughing]" />

I for one am quite eager to see what pops up next but in the meantime I may just have to get the wifey one of those new 17" iMacs to replace her old reliable Rev. A iMac.

[DrBoar]

The aluring siren songs about the bus speed.
That mantra about that slow bus speed throttling the G4 is everywere but were is the evidence for this?

Can anyone show that a dual 1.2 GHz G4 on a G4/400 (with a 100 MHz bus) is much slower CPU wise than a dual 1.25 with a halfway DDR bus at 167 MHz?

According to xlr8yourmac test of dual 1GHz CPUs with 133 and DDR167 the differecen between them is 10% or less even if the DDR is at least 25% faster in the bus speed.

I would guess that a dual 1.42 GHz G4 could saturate the 30 MHz bus on a PM 6100. However, I have yet to see any evidence that the main problem of the G4 is slow bus speed. The fact that that it is a common opinion does not make it true <img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" />

[Nevyn]

[quote]Originally posted by DrBoar:
<strong>The aluring siren songs about the bus speed.
That mantra about that slow bus speed throttling the G4 is everywere but were is the evidence for this?....The fact that that it is a common opinion does not make it true <img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" /> </strong><hr></blockquote>

There's a variety of computationally-bound programs designed to take full advantage of multiprocessing that see a miserable speedup going from a single 1GHz G4 to a dual 1GHz G4. That in itself doesn't prove anything - but when the same program sees almost the complete doubling when going from a 500MHz G4 to a dual 500MHz dual.... Strongly indicates that the second G4 isn't doing so much _in_some_cases.

A second important note is that the two G4's share a single bus -&gt; there's conflict there. And it makes a Quad insane. There wouldn't be any screaming about price/performance if there was a real bus + quads available (on the high end, there'd still be millions screaming for a $500 box).

You can calculate the 'max' data consumption rate, it is something like 10x the FSB max throughput, way out of whack. It isn't crippling... in an _single_ CPU situation.

[Bigc]

<a href="http://www.barefeats.com/pmddr.html" target="_blank">FWIW</a>

[pey/coy-ote]

[quote]Originally posted by Nevyn:
<strong>


A second important note is that the two G4's share a single bus -&gt; there's conflict there.</strong><hr></blockquote>

Please forgive my ignorance, but this brings up a question for me. How are the duals connected to the FSB? As I understand it, they are connected in such a way that the FSB is part of Motos fab, and Apple cannot change the bus. This led me to assume the FSB was part of the G4 cpu, and that Duals would have two busses? And, if the FSB is not on the cpu but rather on the chipset, what about the design has made, altering it too expensive?

[muah]

Ok, I assume that the 970 is going to be the next processor for Apple. But here is the rumor I don't get:

"They are working on software so that 2 processors will work seamlessly like 1." If that is true, then why wouldn't they have just asked IBM to make an actual (dual core) Power4 with the Altivec unit???

What is the probability that they will even debut with ANY dualies? I realize their are some gains by using multiple processors, but I would think prices will be fairly high in the beginning when yields of this new processor are low.

Those 2 issues alone make me doubt a lot of the talk that is going on about the 970. Especially when people start spouting off saying "I know a guy who knows a guy that was working on x, y, and z for the dual 970 ..."

I would love something that resembles a fact about the whole situation, but until I see it otherwise I will assume any 970 based mac debuts as a single processor. And I would think that is a tricky situation when you debut a "best" system with a single 1.8 Ghz 970 system next to a dual 1.42 Ghz G4 at "better". At least for the marketing department.

[AirSluf]

[Eskimo]

[quote]Originally posted by mooseman:
[QB
Additonally, just last summer, Philips, Motorola, and STMicroelecronics announced they were preparing a facility capable of churning out 90 nanometer microprocessors on 300mm silicon....by the end of 2002. So if preproduction silicon is coming out by the end of 2002, whats keeping Moto from introducing a 2+GHz G4 on a 90nm process 2Q '03?

Just some thoughts. Any opinions on it?[/QB]<hr></blockquote>

Yeah, here's my thoughts, no way. First off Crolles, the facility you speak of is a R&D center first and foremost. It has limited production capabilities and is more of a pilot line for development work on 90nm and sub 90nm processes.

As for your time frame there is little chance that anyone will beat Intel, IBM, or AMD to 90nm for high performance complex CMOS production. Even they are shooting for 90nm production in Q4 '03 at best. There have been a number of issues that have arisen trying to get 90nm going and now a good segment of the industry has pushed out adoption to late '04 or early '05. It is becoming paramount for designers to take into account the production challenges present at 130nm and below. This means simple scaling is no longer as feasible for a design intended for an older technology like .18um.

[Programmer]

[quote]Originally posted by pey/coy-ote:
<strong>Please forgive my ignorance, but this brings up a question for me. How are the duals connected to the FSB? As I understand it, they are connected in such a way that the FSB is part of Motos fab, and Apple cannot change the bus. This led me to assume the FSB was part of the G4 cpu, and that Duals would have two busses? And, if the FSB is not on the cpu but rather on the chipset, what about the design has made, altering it too expensive?</strong><hr></blockquote>

Each chip (the processors and the chipset) have a "bus interface" to which the bus connects. MPX is a shared bus, so in Apple's dual processor machines there are 3 devices all connected by the same bus (a bus is basically a set of motherboard traces which are organized by some protocol). Only one device can use the bus at a time, so they need to negotiate for who is going to use it next (they can all listen at once to make sure they know what is going on).

[Programmer]

[quote]Originally posted by Eskimo:
<strong>Yeah, here's my thoughts, no way. First off Crolles, the facility you speak of is a R&D center first and foremost. It has limited production capabilities and is more of a pilot line for development work on 90nm and sub 90nm processes.

As for your time frame there is little chance that anyone will beat Intel, IBM, or AMD to 90nm for high performance complex CMOS production. Even they are shooting for 90nm production in Q4 '03 at best. There have been a number of issues that have arisen trying to get 90nm going and now a good segment of the industry has pushed out adoption to late '04 or early '05. It is becoming paramount for designers to take into account the production challenges present at 130nm and below. This means simple scaling is no longer as feasible for a design intended for an older technology like .18um.</strong><hr></blockquote>

Yeah, I wouldn't hold my breathe for an early adoption of anything by Motorola. <img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" /> I wonder if IBM's "rapid migration" plan for the 970 on 0.09 takes into account the issues that "have arisen". I believe IBM has been experimenting with 0.09 for a while now, and their design techniques may allow them to move the 970 to the new process more rapidly than other companies can manage to move their designs.

[pey/coy-ote]

[quote]Originally posted by Programmer:
<strong>

Each chip (the processors and the chipset) have a "bus interface" to which the bus connects. MPX is a shared bus, so in Apple's dual processor machines there are 3 devices all connected by the same bus (a bus is basically a set of motherboard traces which are organized by some protocol). Only one device can use the bus at a time, so they need to negotiate for who is going to use it next (they can all listen at once to make sure they know what is going on).</strong><hr></blockquote>

Thanks very much for the information Programmer.

Back to the 970...
According to
<a href="http://www.realworldtech.com/page.cfm?AID=RWT101502203725" target="_blank">David Wang's article on the 970</a> (Oct,2002). The 970's companion chip will be difficult to engineer. [quote]<strong>... the specification of the ~900 MHz operation on the system board would require considerable investment into the system support chip. Moreover, the nature of the point to point interconnect means that to support a dual CPU system, the companion chip must be designed with the dual CPU SMP in mind, with dedicated channels devoted to each CPU. Furthermore, to support the high bandwidth available on the system interconnect, a dual channel PC2700 DDR SDRAM memory system would appear to be a minimum requirement to support a single CPU... </strong><hr></blockquote> The article goes on to say this will add to the 970's expense and restrict it to the PowerMacs. Guess we won't see the 970 in a Mac untill Apple has it's support chip ready. No reason to expect Apple hasn't been planning for this, or is behind schedule. But there is at least one other (rather complicated) factor involved in the timing of the 970's appearance. So even if IBM has chips in quantity early... Or maybe for Apple it'll all be as easy as pie. ...And speaking of Apple Pi could that be one of the suprises Trancendental Octothorpe was referring to?

[Outsider]

Don't forget the possibility of Apple using IBM's companion chip (which will be released at the same time as the 970) and that chip in turn connects to an Apple peripheral controller which includes PCI, AGP, Firewire, Ethernet, USB, ATA, etc. (basically, everything except for the memory controller).

[Ompus]

[quote]Originally posted by muah:
<strong> ...
I would think that is a tricky situation when you debut a "best" system with a single 1.8 Ghz 970 system next to a dual 1.42 Ghz G4 at "better". </strong><hr></blockquote>

Avoiding this marketing difficulty is one reason why we'll see multi-processor 970s at the outset.

A second reason is to capture mindshare. Which would get more media play: A single 970 that puts the Mac platform on par with the p4, or a multi-processor 970 that smokes 'em.

A third reason is profit margins. It's almost axiomatic that margins improve as you move up the ladder. Better to sell one Porsche than 2 Miatas.

A fourth reason is internal morale. Companies are made up of people with egos. I'm sure their engineers are frustrated and a bit embarassed by shootouts between p4s and g4s. Building the baddest and fastest machine around will pump THEM up.

A fifth reason is because they can.

None of this is to say that the 970 will NEVER be released in a SP config. Just that I think it will DEBUT as a screamer.

[ 02-06-2003: Message edited by: Ompus ]</p>

[Transcendental Octothorpe]

The 970 and the 'companion chip' should appear at the same time. (barring unforseen production problems, of course.)

Apple won't have to scramble to design a chip once they recieve production 970s. They'll already have them.

[Mark- Card Carrying FanaticRealist]

[quote]Originally posted by Ompus:
<strong>

Avoiding this marketing difficulty is one reason why we'll see multi-processor 970s at the outset.

A second reason is to capture mindshare. Which would get more media play: A single 970 that puts the Mac platform on par with the p4, or a multi-processor 970 that smokes 'em.

A third reason is profit margins. It's almost axiomatic that margins improve as you move up the ladder. Better to sell one Porsche than 2 Miatas.

A fourth reason is internal morale. Companies are made up of people with egos. I'm sure their engineers are frustrated and a bit embarassed by shootouts between p4s and g4s. Building the baddest and fastest machine around will pump THEM up.

A fifth reason is because the can.

None of this is to say that the 970 will NEVER be released in a SP config. Just that I think it will DEBUT as a screamer.</strong><hr></blockquote>

If it was me, the machine I'd debut would be a dual 1.6 and I'd keep it there for around 2 months.

Why?

1) A dual 1.6 970 will still blow the pants of any G4 system.
2) By releasing a dual system only, you keep the initial cost high, which keeps demand down whilst you pick up yield and production capacity.
3) A 1.6 is below the initial proposed maximum of 1.8 GHz, which means you still have some headroom to bring out a 1.8 uniprocessor once you have enough chips in the store cupboard and a massive latent demand.

[Programmer]

[quote]Originally posted by Mark- Card Carrying FanaticRealist:
<strong>

If it was me, the machine I'd debut would be a dual 1.6 and I'd keep it there for around 2 months.

Why?

1) A dual 1.6 970 will still blow the pants of any G4 system.
2) By releasing a dual system only, you keep the initial cost high, which keeps demand down whilst you pick up yield and production capacity.
3) A 1.6 is below the initial proposed maximum of 1.8 GHz, which means you still have some headroom to bring out a 1.8 uniprocessor once you have enough chips in the store cupboard and a massive latent demand.</strong><hr></blockquote>

I think the intro-level performance should be a factor of how many fast chips they are getting in the fabs. If the yields of 1.8 GHz are terrific then ship a dual 1.8 GHz machine, if they are louzy then ship a 1.6 GHz machine (or whatever speed the yields are decent for). If the yields are really good at the 1.8 level then you can expect it'll ramp well and quickly beyond the initial 1.8, if they're not this gives Apple a bit of headroom once the 1.8's are plentiful. As a result I think speculating about the details and trying to figure out Apple's thought process behind the introduction is pointless -- their decision will be governed primarily by their supply of components as has always been the case.

[jwdawso]

As far as the PPC970 memory accessing needs: Apple's architecture is already there and Apple is selling it now. We look at the system controller as a way to get around the G4 memory speed - read SLOW - limitations. BUT - it also will get around the the PPC970 memory FAST SPEED needs. Apple's custom - read EXPENSIVE - solution was not aimed at just adding DDR to a G4, it's aimed at high speed access, so that the memory subsystem can be designed independtly of the Processor capability. In other words, for the PPC970, the System Controller will take care of the relatively slow memory, whereas on the G4 it takes care of the relatively fast memory.

[pey/coy-ote]

[quote]Originally posted by Transcendental Octothorpe:
<strong>The 970 and the 'companion chip' should appear at the same time. (barring unforseen production problems, of course.)

Apple won't have to scramble to design a chip once they recieve production 970s. They'll already have them.</strong><hr></blockquote>
Yeah, I just saw an IBM .pdf simplified diagram that included the companion chip, and it hit me that they'd have to have one for their own use. Apple could either use IBM's or design their own depending on which suited them best. I still wonder if Apple pi will show up?

[Nevyn]

[quote]Originally posted by muah:
<strong>"They are working on software so that 2 processors will work seamlessly like 1." If that is true, then why wouldn't they have just asked IBM to make an actual (dual core) Power4 with the Altivec unit???</strong><hr></blockquote>

I don't have the actual manufacturing yeilds or # of chips per wafer, but here's how it goes:

I make a 'wafer' which is as many 'chips' as possible wedged onto it. Say that's 20 single core chips.

There are ten defects -&gt; _10_ survived. I can make 5 "Dual CPU" macs out of it.

A dual core is roughly double in size to a single core CPU. Yes, 2 singles should operate _very_ similar to one dual core. But. They're twice as large in chip area.

So that same wafer could only have 10 dual-core chips. But there's still going to be _10_ defects. If the defects were evenly distributed there'd be _zero_ usable dual core chips that survive. That really sucks.

My numbers are completely fake, but you can see that at least sometimes '2 single core CPUs' are prefered to aiming for dual core CPUs. When successrates are very high (meaning defects are very low) it starts getting useful to consider dual cores... or more. For the Power4, which is normally a dual-core chip, IBM does make boxes out of 'defective' ones, where one core tests out fine, and the other core is dead. A dual core chip saves quite a bit on complexity/wiring/motherboard issues, but it also concentrates the heat problems.

So there is a balancing act between 'which is better'.

[Vvmp]

Here's a bit of a side track, but I ran into an interesting story titled SERVER START-UP GETS FUNDS.

<a href="http://www.bayarea.com/mld/mercurynews/business/technology/5107216.htm" target="_blank">http://www.bayarea.com/mld/mercurynews/business/technology/5107216.htm</a>

Basically....Key Research got funding to make 64-bit servers running the Linux OS (inexpensive). They don't reveal what processor will be implemented, but will be released in the 2nd 1/2 of the yr. Perhaps they will be another 970 customer??

"Rosenthal declined to say what chips the company would use, but he said the company will rely on low-cost alternatives to the Itanium. Rosenthal said the company would reveal more about its machines later and will launch them in the second half of 2003."

[Gilsch]

Come on, we all know Apple will offer Quad 970 XtremeMacs(fastest) with processors running at 1.8 Ghz. If you can't afford that then it's DP 970 XtremeMacs at 1.6Ghz(faster) or single 970 1.6Ghz XtremeMacs. :D

[AirSluf]

[Transcendental Octothorpe]

[quote]Originally posted by AirSluf:
<strong>

Not a problem for IBM, they already deal quite effectively with it. They manufacture both dual and single core versions of the Power4, the single core versions are just a dual core that tests as a single core failure and is then laser isolated to generate a single core sellable chip.</strong><hr></blockquote>

Uh, did you not read his post?

[quote] For the Power4, which is normally a dual-core chip, IBM does make boxes out of 'defective' ones, where one core tests out fine, and the other core is dead. <hr></blockquote>

[rickag]

[quote]Originally posted by Eskimo:
.....First off Crolles, the facility you speak of is a R&D center first and foremost.....<hr></blockquote>

I don't know how the Crolles facility got brought into this? If I remember correctly, the Motorola press release concerning high end processors ,manufactured at the end of 2002 using a 0.09µm or 0.10µm process, mentioned an Asian or Southeast Asian facility.

Then again, I could have dreamt the whole thing, since I can no longer find the press release.(Now I'm even getting confused as to whether they mentioned 0.090 or 0.10µm, they release so much #@$^ and never meet their dates, it's vexing.)

0.09µm, 0.10µm phooeey they forgot 0.15µm ... oh great googly moogly, I forgot they changed the size of their rulers/micrometers, whatever, so actually the 0.18µm was really actually 0.15µm, when in fact I thought it was truely 0.22µm. Anyway, was Motorola involved the Mars explorer that missed because different units of measure were co-mingled?

Bottom line, Motorola ain't shipped 0.13µm let alone 0.10µm or 0.090µm parts and don't expect them soon. @#!^$*&%$#* :mad:

damn, now what did I do with that bottle of lithium tablets.....???? Oh well, found my prozac.

[ 02-06-2003: Message edited by: rickag ] :)

[ 02-06-2003: Message edited by: rickag ]

I just noticed in the above rant, I continually wrote 0.9µm instead of 0.09µm, please forgive this adled brain of mine. Kind of funny though, having Motorola going backwards by a factor of 9X's in die size.

[ 02-07-2003: Message edited by: rickag ]</p>

[Mark- Card Carrying FanaticRealist]

[quote]Originally posted by Programmer:
<strong>

I think the intro-level performance should be a factor of how many fast chips they are getting in the fabs. If the yields of 1.8 GHz are terrific then ship a dual 1.8 GHz machine, if they are louzy then ship a 1.6 GHz machine (or whatever speed the yields are decent for). If the yields are really good at the 1.8 level then you can expect it'll ramp well and quickly beyond the initial 1.8, if they're not this gives Apple a bit of headroom once the 1.8's are plentiful. As a result I think speculating about the details and trying to figure out Apple's thought process behind the introduction is pointless -- their decision will be governed primarily by their supply of components as has always been the case.</strong><hr></blockquote>

You and I are probably not that far apart on this issue, the only fundamental difference being that my position pre-emptively seeks to make a 1.6 (or something similar that isn't quite the bleeding edge) the engine for a dual-processor system rather than just saying "let's wait and see whether we can get out enough 1.8s (or whatever defines the edge of the envelope) and then define the product/marketing development accordingly".

That said, if the process starts to reliably churn out 1.8s and you knew that was happening early enough - say June - then you could take a more relaxed view with the aim of pushing out a launch for some initial low-end units sometime in October, shipping for Thanksgiving, followed by another "buckle up" display of the "The Fastest (two-way, 64-bit) Machine You Can Buy For $3000" at MWSF '04 and - with any luck - an innvovative rack-mounted 4-way 1.6 at NAB 04.

All of this should just about keep the favourable column inches rolling for about six-eight months which will just about cover the fact that there may not be a summer show in '04.

[Pilmour Boy]

From my understanding of the situation, it was Apple's job to build the companion chip. As of the end of 2001, they were behind.

[Programmer]

[quote]Originally posted by Mark- Card Carrying FanaticRealist:
<strong>

You and I are probably not that far apart on this issue, the only fundamental difference being that my position pre-emptively seeks to make a 1.6 (or something similar that isn't quite the bleeding edge) the engine for a dual-processor system rather than just saying "let's wait and see whether we can get out enough 1.8s (or whatever defines the edge of the envelope) and then define the product/marketing development accordingly".

That said, if the process starts to reliably churn out 1.8s and you knew that was happening early enough - say June - then you could take a more relaxed view with the aim of pushing out a launch for some initial low-end units sometime in October, shipping for Thanksgiving, followed by another "buckle up" display of the "The Fastest (two-way, 64-bit) Machine You Can Buy For $3000" at MWSF '04 and - with any luck - an innvovative rack-mounted 4-way 1.6 at NAB 04.

All of this should just about keep the favourable column inches rolling for about six-eight months which will just about cover the fact that there may not be a summer show in '04.</strong><hr></blockquote>

Its been standard practice for Apple to decide on the exact processor speeds at a very late date before introduction of the machines. They've had to do this a lot with Motorola's track record.

[Mark- Card Carrying FanaticRealist]

[quote]Originally posted by Programmer:
<strong>

Its been standard practice for Apple to decide on the exact processor speeds at a very late date before introduction of the machines. They've had to do this a lot with Motorola's track record.</strong><hr></blockquote>

No argument on this, but in reality that's why you would pre-emptively aim to deliver behind the bleeding edge - it's a psychology thing, I promise myself and aim to deliver something I know is within my grasp. Then, if my supplier turns out to be capable of walking the walk, I know I have somewhere to go within three months which keeps the product "hot" in the marketplace.

Genuinely, Programmer, I'm not looking for an argument here; I'm just putting forward a different way of bring 970 to the marketplace (under promise, then over deliver at regular intervals) so that we can spend the next three to five years continually moving the bar on the Wintel community. Let Dell and HP eat some crow for a while, it's only fair.

[Programmer]

[quote]Originally posted by Mark- Card Carrying FanaticRealist:
<strong>Genuinely, Programmer, I'm not looking for an argument here; I'm just putting forward a different way of bring 970 to the marketplace (under promise, then over deliver at regular intervals) so that we can spend the next three to five years continually moving the bar on the Wintel community. Let Dell and HP eat some crow for a while, it's only fair.</strong><hr></blockquote>

I have a sneaking suspicion that IBM is under-promising on the 970, and Apple hasn't promised anything. I know what you're saying about not delivering the bleeding edge, my point is simply that 1.8 GHz may very well not be the bleeding edge of the 970.

[MacJedai]

[quote]Originally posted by pey/coy-ote:
<strong>
Yeah, I just saw an IBM .pdf simplified diagram that included the companion chip, and it hit me that they'd have to have one for their own use. Apple could either use IBM's or design their own depending on which suited them best. I still wonder if Apple pi will show up?</strong><hr></blockquote>

Yup

[AirSluf]

[Programmer]

[quote]Originally posted by Transcendental Octothorpe:
<strong>Oh, anyone know what a VSP interface is?</strong><hr></blockquote>

I just realized that we were completely barking up the wrong tree with the VSP thing. In this context it does not mean "Virtual Single Processor" at all... it means "Vector Signal Processor". Big difference.

For those that remember this is probably those DSP-like features in the chipset that Moki was talking about last summer. Four vector units much like you find in GPUs, game consoles, or other dedicated vector processors. Sitting in the chipset they would have direct high speed access to the system memory and the AGP bus. Audio, Quartz rendering, OpenGL, QuickTime, and custom signal processing algorithms should all be possible on these VSPs.

Apple has tried this kind of thing before way back in the 68040 days with the DSPs in the 660av and 840av. Unfortunately back then they weren't very powerful and media processing was much less common. Things are quite different now. The main thing is can Apple commit to keeping these things in their machines going forward? The '040 AV machines were like orphaned children. Integrating it into the chipset keeps the component count (and cost) down, and lets Apple tailor them specifically for Apple's needs. There is also a lot of technology floating around these days for keeping the code running on these things independent of the hardware details (i.e. OpenGL's shader technologies). The integrated nature of the beast might allow it to show up in the low-end as well as the high-end, allowing substantially improved performance from G4-based computers.

Could be very cool things in store...

[KidRed]

[quote]Originally posted by Pilmour Boy:
<strong>From my understanding of the situation, it was Apple's job to build the companion chip. As of the end of 2001, they were behind.</strong><hr></blockquote>

Well, that was a year ago.

[JCG]

[quote]Originally posted by Programmer:
<strong>

I have a sneaking suspicion that IBM is under-promising on the 970, and Apple hasn't promised anything. I know what you're saying about not delivering the bleeding edge, my point is simply that 1.8 GHz may very well not be the bleeding edge of the 970.</strong><hr></blockquote>

If I recall correctly, when IBM anounced the Gecko chip with Nintendo its speed was 400 mhz, but the shipping system have a 485 mhz chip in them acording to Nintendo. This is a precident for IBM delivering more than promised. Also IBM said that the 970 will scale rapidly to 2 Ghz with a die shrink, and actual real world tests of the chips might be able to reliably handle better than what they have anounced, just as the G4 reliably handled less speed than was promised. I'm sure that IBM and Apple do not want to repeat Moto's mistake there.

[Brian Green]

Just a quick question here to clear something up for me. How do these MHz / GHz ratings come about in the first place? Do these fabs have a computer without a CPU standing by that runs the equivalent of iBench with these processors, or do they have a seperate machine they drop the CPU into that says, "Yup, it's a 1.6." How do they know what a chip runs at and how much can they get away with lying about? Like a chip in their machine is actually a 1.4, but they say, "Screw it, let's call it a 1.6", just like hard drive manufacturers can get away with calling an 18.62 GB hard drive a 20 Gig (like the one in my iBook). I'm just curious how these numbers actually come about. Is there someone out there that can explain this stuff without needing an advanced degree in quantum mechanics? Thanks. :)

[Transcendental Octothorpe]

[quote]Originally posted by Brian Green:
<strong>...just like hard drive manufacturers can get away with calling an 18.62 GB hard drive a 20 Gig (like the one in my iBook). I'm just curious how these numbers actually come about. Is there someone out there that can explain this stuff without needing an advanced degree in quantum mechanics? Thanks. :) </strong><hr></blockquote>

Hmm, one at a time, I guess:

HD size: This one really is a hoodwink. HD manufactures do (sort of) fib about the size of their units ;) . They quote GB as "billions of bytes" (so 1GB = 1,000,000,000 bytes) instead of 2^30 (1,073,741,824 bytes), which is what a GB means to any programmer or engineer. Its just like 1K in computer world really means 1024 (2^10), not 1000. That is the standard, and has been for decades.

See <a href="http://www.itd.umich.edu/~doc/Digest/1195/feat03side01.html" target="_blank">this</a> for more info.

[ 02-12-2003: Message edited by: Transcendental Octothorpe ]</p>

[Transcendental Octothorpe]

[quote]Originally posted by Brian Green:
<strong>Just a quick question here to clear something up for me. How do these MHz / GHz ratings come about in the first place? Do these fabs have a computer without a CPU standing by that runs the equivalent of iBench with these processors, or do they have a seperate machine they drop the CPU into that says, "Yup, it's a 1.6." How do they know what a chip runs at and how much can they get away with lying about? Like a chip in their machine is actually a 1.4, but they say, "Screw it, let's call it a 1.6"...</strong><hr></blockquote>

Now, about CPU clock speed: it depends what you are talking about. PowerPCs (and Intel) use an actual clock speed that is measured with automated testing equipment. The parts have sets of data run through them at varying speeds, and some parts (within the same production batch) will run those data sets correctly at faster speeds than other chips. This is called "speed binning" and is done with memory as well.

If you are talking about SPEC numbers, that is a test done in software measuring how fast that CPU can perform certain calculations.

If you are talking about AMD's numbers, that's another <a href="http://archive.infoworld.com/articles/hn/xml/01/10/15/011015hnathlonxp.xml" target="_blank">story</a>.

[ 02-12-2003: Message edited by: Transcendental Octothorpe ]</p>

[Transcendental Octothorpe]

(please forgive any forthcoming condescending tone)

Shoot, well while we're at it, a few notes about semantics:

A CPU is not "a 1.4 GHz". Nor does "this chip have 1.6GHz".

The chip runs at 1.6GHz.

Also:

A chip does not "have 42 Watts". It uses energy, which can be quantified in Watts.

This chip uses or draws or dissapates 42 Watts.

I've heard people here and elsewhere comment "not only does this chip run cooler, it also draws less battery, making it perfect for powerbooks!" While true, it is pointing out the obvious. Energy usage (for CPUs, at least) is always the same as the dissapated heat. Conservation of energy and all that. A lower wattage device will always produce less heat and draw less energy than a higher wattage part. They aren't two seperate goals, both of course both are good :) .

Correct as necessary...

[ 02-12-2003: Message edited by: Transcendental Octothorpe ]</p>

[Junkyard Dawg]

A car is never a "V8", as in, "did you get the Camaro V8?"

A car is powered by a V8 engine.

So why don't you go tell the millions of car-freaks out there that their grammar is all wrong?

[Rhumgod]

[quote]Originally posted by Junkyard Dawg:
<strong>So why don't you go tell the millions of car-freaks out there that their grammar is all wrong?</strong><hr></blockquote>

Really, who gives a sh!t. Not I. As long as it does what I want it to, and doesn't slow me down in the process.

Nice to see you back, JYD? On vacation in sunny Aruba or something? ;)

[Transcendental Octothorpe]

[quote]Originally posted by Rhumgod:
<strong>

Really, who gives a sh!t. Not I. As long as it does what I want it to, and doesn't slow me down in the process.

Nice to see you back, JYD? On vacation in sunny Aruba or something? ;) </strong><hr></blockquote>

Hey, sorry. I was just in informative mode.

I'll try to never be informative again. :D

[Rhumgod]

[quote]Originally posted by Transcendental Octothorpe:
<strong>

Hey, sorry. I was just in informative mode.

I'll try to never be informative again. :D </strong><hr></blockquote>

I wasn't trying to be abrasive, just pointing out I think a lot of computer terminology is for a small portion of the computer buying public. As long as it's fast, yada, yada. Sorry if I appeared disgruntled ;)

[Vvmp]

Just another story to confirm the arrival date of the 970. A NEWS article on new Blade Servers coming from IBM using the PPC 970.

<a href="http://news.com.com/2100-1001-984353.html?tag=fd_top" target="_blank">http://news.com.com/2100-1001-984353.html?tag=fd_top</a>

[Telomar]

[quote]Originally posted by Transcendental Octothorpe:
<strong>

Hmm, one at a time, I guess:

HD size: This one really is a hoodwink. HD manufactures do (sort of) fib about the size of their units ;) . They quote GB as "billions of bytes" (so 1GB = 1,000,000,000 bytes) instead of 2^30 (1,073,741,824 bytes), which is what a GB means to any programmer or engineer. Its just like 1K in computer world really means 1024 (2^10), not 1000. That is the standard, and has been for decades.</strong><hr></blockquote>

Close but very wrong. HD manufacturers and most ISPs actually have it correct as 1GB is 1 000 000 000 bytes. It relates to the prefix Giga, which is formally defined as 10^9. That really is the standard as defined by the SI system (don't they teach people that in high school? :confused: ).

The problem arises that in the early days of computers there were no prefixes for binary numbers so computers just stole the nearest SI unit and said, "That's close enough." Problem is as the orders of magnitude increase the errors get substantially larger and it no longer is close enough.

So where is this leading? Well there is in fact another set of prefixes for binary numbers that you can read all about over <a href="http://www.romulus2.com/articles/guides/misc/bitsbytes.shtml" target="_blank">here.</a> Problem is the system makers seem to refuse to use them but at the end of the day the HD manufacturers have it right so 3 cheers for the HD manufacturers :D

Edit: <a href="http://physics.nist.gov/cuu/Units/binary.html" target="_blank">Or another link.</a>

[ 02-12-2003: Message edited by: Telomar ]</p>

[Transcendental Octothorpe]

[quote]Originally posted by Telomar:
<strong>...</strong><hr></blockquote>

<img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" />

I wasn't wrong. It is a standard in any computer related field to use powers of 2 in increments of 10.

Of course you are right the the real Scientific standard is that mega = 10^6 etc., but we are in base two here, not base 10, and the convention as I have stated it has been in place since at least the 60's.

[Edit, I just read your second link. Interesting, but it'll be a while before they change the accepted jargon of the industry. I still mantain that the HD manufature's notation is intentionally misleading, and not some noble attempt to be faithful to SI units.]


But, as others have said: who cares? We all just want a kick ass comp


(and I believe that we will be getting one from Apple before fall)

[ Edit: DOH! you'd think I'd know mega from giga. <img src="embarrassed.gif" border="0"> ]

[ 02-12-2003: Message edited by: Transcendental Octothorpe ]</p>

[TJM]

[quote]Originally posted by Transcendental Octothorpe:
[

Of course you are right the the real Scientific standard is that mega = 10^9 etc., <hr></blockquote>

Oh, Boy! Now it's my turn to be hyper-picky! ;)

kilo = 10^3
mega = 10^6
giga = 10^9
tera = 10^12

:D :p

[BRussell]

[quote]Originally posted by Vvmp:
<strong>Just another story to confirm the arrival date of the 970. A NEWS article on new Blade Servers coming from IBM using the PPC 970.

<a href="http://news.com.com/2100-1001-984353.html?tag=fd_top" target="_blank">http://news.com.com/2100-1001-984353.html?tag=fd_top</a></strong><hr></blockquote>

From that article: [quote]In the longer term, though, IBM has even grander ambitions. It's possible that a special link could join two four-processor blade servers into an eight-processor system. The approach would use a variant of IBM's EXA "Summit" chipset, which can use high-speed cables to link four-processor groups into eight-, 12- and 16-processor x440 servers, Benck said in an interview.

"We have the technology to create blades that plug in to create larger symmetrical multiprocessor servers," he said.<hr></blockquote>moki and others have talked about IBM may be sandbagging a bit, and others have talked about technologies to treat multiple chips as a single chip. I wonder if this is important and whether it fits into Apple's plans.

[KidRed]

[quote]Originally posted by Vvmp:
<strong>Just another story to confirm the arrival date of the 970. A NEWS article on new Blade Servers coming from IBM using the PPC 970.

<a href="http://news.com.com/2100-1001-984353.html?tag=fd_top" target="_blank">http://news.com.com/2100-1001-984353.html?tag=fd_top</a></strong><hr></blockquote>


I like this line-
[quote] ...will use the PowerPC 970 processor. The 1.8GHz processor is expected to arrive later this year, IBM said. <hr></blockquote>

So i think they bodes well for getting the 970 this year and not next :)

[ 02-12-2003: Message edited by: KidRed ]</p>

[Tom West]

The potential problem with Mac's bearing 970's is not with IBM getting the chip out, it's whether Apple has the design competence to create the companion chip and motherboard that the 970 needs.

It's not a trivial undertaking and there are real questions about whether Apple will be capable of creating the technology to use the 970 effectively.

[Hoping for the October 2003, expecting March 2004]

[ 02-13-2003: Message edited by: Tom West ]</p>

[Barto]

If Apple was so technologically inept, you wouldn't be talking about them. That's because Apple would be irrelevent.

Apple r0x0r, they have been let down by Motorola. Sure, it was partly their fault not to forsee the G4 fiasco, but they probably had other things on their mind in 97/98. Like how to stay alive.

Trust us. There is no way Apple will not be 100% ready for the next Power Mac. This is most probably the #1 priority at Apple. The company's fate depends on it.

I'm serious. If Apple keeps using embedded chips, their market share will continue to erode in all but the home user market.

And when there is only the home user left, the application base will dry up. And when that happens, they will lose home user market, and Apple will be nothing more than QuickTime and investment.

-----------------

First, Apple is convincing their customers that they don't have to buy Wintel. They can buy Apple, because Apple gets rid of all the associated costs and professional setup. This is happening.

But that's only half the story.

Second, Apple needs kick-arse hardware to be better than Wintel. It doesn't matter whether it is a big mofo machine, clustering, or somewhere in between. As long as it is kick-ass for the creative market.

That is why Apple cares so much about the 970.

That is why Apple will get everything right this time.

And that is why the next Power Mac will be the turn of the tide.

Barto

[ 02-13-2003: Message edited by: Barto ]</p>

[Transcendental Octothorpe]

[quote]Originally posted by Tom West:
[QB]The potential problem with Mac's bearing 970's is not with IBM getting the chip out, it's whether Apple has the design competence to create the companion chip and motherboard that the 970 needs.
[QB]<hr></blockquote>

You are right, it is complicated.

But that doesn't change the fact that the chipset is already done. As in finished. NOT in design stage. Those complications have been dealt with. (to be clear I have no idea how far Apple is with a MB, but I would think it would track with the chipset design.)

Why do people keep saying these things over and over. I still hear people saying how Apple can't use the 970 because its 64bit and every piece of existing SW will break.

:rolleyes:

Get over it. Apple will use the chip. I won't be 2004 (unless IBM's fabs burn to the ground in the next couple months). 32bit software will run. Apple won't recieve a pallet of chips one day and comment "oh crap, we better get started designing a MB and a chipset".

[ 02-13-2003: Message edited by: Transcendental Octothorpe ]</p>

[kroehl]

[quote]Originally posted by Transcendental Octothorpe:
[qb]

You are right, it is complicated.

But that doesn't change the fact that the chipset is already done. As in finished. NOT in design stage. Those complications have been dealt with. (to be clear I have no idea how far Apple is with a MB, but I would think it would track with the chipset design.)

Why do people keep saying these things over and over. I still hear people saying how Apple can't use the 970 because its 64bit and every piece of existing SW will break.

:rolleyes:

<hr></blockquote>

I agree. Of course Apple is ready and capable. Think ApplePI, FireWire, MoBo chipsets etc. etc. It's not like they are still going to the Radioshack and sticking things together in a wooden box in Scully's garage (or was that Woz's).

Apple knows the specs, the interface, the bus' and everything else to build the boxes and have them ready by the time UPS pulls in with a delivery of 970s. I'm sure Ives is working on the box as we speak. <img src="graemlins/cancer.gif" border="0" alt="[cancer]" />

kroehl

[ 02-13-2003: Message edited by: kroehl ]</p>

[Ensoniq]

[quote]Originally posted by kroehl:
<strong>

Apple knows the specs, the interface, the bus' and everything else to build the boxes and have them ready by the time UPS pulls in with a delivery of 970s. I'm sure Ives is working on the box as we speak. <img src="graemlins/cancer.gif" border="0" alt="[cancer]" />
</strong><hr></blockquote>

EXACTLY! It amazes me how many people here think that Apple is out of the loop on this one, and can't even begin to work on the new machines until IBM publicly announces the 970 is shipping in quantity.

Apple and IBM are partners in this chip, whether that is ever admitted publicly or not. They are so intimate with each other that I hope they are both using birth control. :)

There is no doubt that Apple has working 970 prototypes in their labs right now. They have everything they need to build the next kick-ass Apple system, and you can be certain that as soon as IBM is ready to publicly announce that the 970 is shpping in quantity, Apple will be there with the new machine. To think otherwise clearly indicates that you just don't "get" Apple at all.

-- Ensoniq

[Tom West]

[quote]But that doesn't change the fact that the chipset is already done. As in finished. NOT in design stage. Those complications have been dealt with. (to be clear I have no idea how far Apple is with a MB, but I would think it would track with the chipset design.)
<hr></blockquote>

Wow. Did you actually get that from an Apple contact? My contact only broadly hinted about chip troubles (not wanting to lose his job). But if you have (relatively) hard info, that's *great* news.

[quote] Why do people keep saying these things over and over.<hr></blockquote>

Um, 'cause that's what was hinted at from our friends at Apple? That and we don't want another round of "The World is Coming to an End" when expectations are dashed yet again :-).

[quote] I still hear people saying how Apple can't use the 970 because its 64bit and every piece of existing SW will break.<hr></blockquote>

I'm not worried about software, only hardware. I figure Apple will punt on 64 bits for a few years, given how little gain there is for them to invest the time to do it right.

[JBL]

[quote]Originally posted by Ensoniq:
<strong>

They are so intimate with each other that I hope they are both using birth control. :)
</strong><hr></blockquote>

I hope not! I am waiting impatiently to adopt one of their baby PowerMac 970's!!

[Amorph]

[quote]Originally posted by Tom West:
<strong>
I figure Apple will punt on 64 bits for a few years, given how little gain there is for them to invest the time to do it right.</strong><hr></blockquote>

I'm almost positive that they'll roll out a 64-bit OS with the rollout of the 970, or shortly thereafter. Oracle and Sybase are either coming to OS X or already here. IBM is asking its customers if they want DB/2 on OS X. Apple is wooing UNIX workstation users, who are long since used to 64 bit platforms, and it has a strong presence in biotech, at NASA, and in Los Alamos.

Most of the traditional Apple markets might be a bit slow to move to 64 bit, since the number of educational CD-ROMs that need to access terabytes of virtual memory are few and far between. But don't forget the markets that Apple is moving into. Scientists, high-end video and 3D artists, and enterprise customers will cheerfully consume all the power that Apple can throw at them. I'm sure that some Photoshop users wouldn't mind having essentially unlimited VM, and more than 2GB real RAM, to play around in either. (Yes, I know that the current G4 can address more than 2GB RAM...).

[ 02-13-2003: Message edited by: Amorph ]</p>

[Nevyn]

[quote]Originally posted by Amorph:
<strong>

I'm almost positive that they'll roll out a 64-bit OS with the rollout of the 970, or shortly thereafter. </strong><hr></blockquote>

...and at the very least, they'll want to _DEMO_ something that has OOOOOhhhh factor in it. I'm not (and I don't think Amorph is either) talking about a full rewrite where maximum OS goodness/optimizations are done to see speed improvements due to 64-bitness. Just that the OS will _work_, and allow an app to both access &gt;4GB of RAM directly, _and_ allow integer math etc on 64 bit values.

That doesn't require a complete rewrite of Quicktime, or most anything really. The virtual memory pager and the hardware drivers (kexts) for the actual chips that are present would seem to be the big issues - and new kexts are written for every major new hardware anyway. (The 12" has a new 'Intrepid' chip in it.-&gt; new kext needed.)

(It can't be said enough that) 64 bits don't help most consumer apps. However, Apple can certainly find things where having a 64-bit integer unit is (easily) a major enhancement. Add that to the extra FPU -&gt; a dual 970 with a _minimally_ enhanced OS -&gt; major improvements for _certain_ things. I think Apple can find those things :) If we don't get a Bake Off, I'll be shocked.

[Amorph]

[quote]Originally posted by Nevyn:
<strong>

...and at the very least, they'll want to _DEMO_ something that has OOOOOhhhh factor in it. I'm not (and I don't think Amorph is either) talking about a full rewrite where maximum OS goodness/optimizations are done to see speed improvements due to 64-bitness. Just that the OS will _work_, and allow an app to both access &gt;4GB of RAM directly, _and_ allow integer math etc on 64 bit values.
</strong><hr></blockquote>

I don't think that will ever be done, actually. For all that people talk about how you need a G4 to run OS X, the overwhelming bulk of OS X is straight C, C++, and/or Objective-C. Depending on how long they've prepared for the transition to 64 bit, they could get OS X to 64 bits - and optimized, to the extent of the compiler's abilities, for the 970 - simply by recompiling with a few #defines set appropriately. Most UNIX apps won't need more than a re-Configure/make install at the utmost - not to run, because they should run fine compiled for 32 bit PPC; but to support the 970's 64 bit capabilities.

Not all of OS X has to move to 64 bit, either. QuickTime could (and probably will) remain 32 bit for a while, not least because of its hoary, cross-platform codebase. Since the 970 implements the PPC spec fully, allowing for seamless compatibility with 32-bit code, that's fine. All that needs to be 64 bit to make the jump worthwhile is the core of the OS, including Core Foundation.

The PowerPC is not x86. There is no crisis here, no big move full of uncertainties. The PowerPC was ready for 64 bit from its launch in the early '90s. This transition should be not only less painful than the IA-32 -&gt; IA-64 transition, but less painful than the 68k -&gt; PPC transition, because the ISA is unchanged. Apple is "moving" from PPC to PPC.

[ 02-13-2003: Message edited by: Amorph ]</p>

[Jared]

[quote]Originally posted by kroehl:
<strong>

I agree. Of course Apple is ready and capable. Think ApplePI, FireWire, MoBo chipsets etc. etc. It's not like they are still going to the Radioshack and sticking things together in a wooden box in Scully's garage (or was that Woz's).

[ 02-13-2003: Message edited by: kroehl ]</strong><hr></blockquote>

Whoa now...read up on your Apple history... that is Steve Job's garage ;)

Anyway...this news all sounds very good...May will be here before we know it...I hope something, anything (even if it is only talk about developing 64 bit applications) comes up at WWDC...

[tink]

I have a question about OS X going 64 bit. :confused:

I understand the new hardware (PowerPC 970..), will have minimal issues running the OS be it updated for 64-bit or if it remains the current 32-bit. :cool:

However if/when the OS is updated and compiled to run 64-bit, is there then an issue with older 32-bit legacy systems like my beige, Blue & White and Ti's ability to run this 64-bit OS? <img src="graemlins/surprised.gif" border="0" alt="[surprised]" />

Is Apple going to need to provide two OS update packages, one 32-bit for backwards compatibility with older hardware and one 64-bit that can take full advantage of 64-bit CPUs? <img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" />


-tink

[Amorph]

[quote]Originally posted by tink:
<strong>
However if/when the OS is updated and compiled to run 64-bit, is there then an issue with older 32-bit legacy systems like my beige, Blue & White and Ti's ability to run this 64-bit OS? <img src="graemlins/surprised.gif" border="0" alt="[surprised]" />
</strong><hr></blockquote>

I sincerely doubt that.

At a bare minimum, the system installer can figure out what hardware it's installing onto, and choose the bits that it installs accordingly.

Remember that OS X is based on OpenStep, which ran on all kinds of completely different systems. The people in charge have years of experience making this kind of thing work.

The 970 itself will present less of a challenge, from a systems programming point of view, than the board it runs on, especially in MP setups, but applications and the end user should be blissfully ignorant of the complexities involved.

[ 02-13-2003: Message edited by: Amorph ]</p>

[Transcendental Octothorpe]

[quote]Originally posted by tink:
<strong>
However if/when the OS is updated and compiled to run 64-bit, is there then an issue with older 32-bit legacy systems like my beige, Blue & White and Ti's ability to run this 64-bit OS? <img src="graemlins/surprised.gif" border="0" alt="[surprised]" />

Is Apple going to need to provide two OS update packages, one 32-bit for backwards compatibility with older hardware and one 64-bit that can take full advantage of 64-bit CPUs? <img src="graemlins/oyvey.gif" border="0" alt="[oyvey]" />

-tink</strong><hr></blockquote>

Well, Sun's been distributing Solaris install CDs that swtich between 32 and 64 bit-ness for years now.

It's no big deal.

[Interface Widget]

I'm guessing Apple has been working on next generation systems built around the 970 processor for quite some time. Probably for much longer than most people outside Apple might expect...

Looking at the company's software strategy, all the indications are that Apple is looking well ahead of the current marketplace, with a clear plan and timeframe for bringing new products and technology to market.

Rendezvous, Bluetooth, NG Firewire, iLife, Keynote, Safari and other software innovations... these things should instill confidence that Apple is doing things right. I'm 100% certain that further productivity software is also on the way, so expect more than one application to accompany Keynote before the year is out.

Apple are clearly not shying away from confronting the problems in their complex relationship with Microsoft head on. Consider the rumors of x86 OS X and Marklar..... This is not a company sitting back, they're making some very bold and aggressive moves.

Apple must have done a lot of work with Motorolla on the G5. How much of this can be carried forward with the 970 is anybody's guess, but I don't think Apple has much breathing space. Still, it pays to be cautious when playing guessing games over delivery dates of unannounced products. Nvidia and the current Geforce FX fiasco show how finely balanced issues with new hardware can throw a project into serious trouble. I expect the motherboard and accompanying components of 970-based systems to be the biggest challenge facing Apple this year, not the delivery timeframe or yields of 970 silicon from IBM.

I would feel fairly confident in saying that significant work must be going on within Apple to bring a 64-bit ready version of Mac OS X to market. FreeBSD is already 64-bit clean, look at the ports to SPARC, Alpha and IA64 for evidence of this.... It would garner serious industry attention if Apple were able to bring a combined 64-bit hardware and software solution to market before the end of the year.

With AMD's 64-bit plans drifting back and Microsoft apparently in no particular rush to bring a 64-bit Windows to market, there exists an opportunity too good to miss.

[ 02-13-2003: Message edited by: Interface Widget ]

[ 02-13-2003: Message edited by: Interface Widget ]</p>

[Nebrie]

[quote]Originally posted by DCQ:
<strong>Sorry to be a downer. But Steve told us when the PowerMacs would be updated to the 970.

"2003 will be the year of the portable."

He wouldn't say that if he knew that this year would bring a processor to its desktop line that would threaten Intel.
</strong><hr></blockquote>

He also said Apple has a few tricks up it's sleeve when asked about desktops.

[Tom West]

[quote]"2003 will be the year of the portable."<hr></blockquote>

[quote]He wouldn't say that if he knew that this year would bring a processor to its desktop line that would threaten Intel.<hr></blockquote>

Well, absolutely every clue I can scrounge from Apple and elsewhere points to a 970 machine in the Jan - March 2004 timeframe, although it's all deduction, whisper and employee vague hints.

However, given the forcefulness of Transcendental Octothorpe statements that the 970 is coming, and soon, (which I'm willing to accept at face value, if he'll admit to having an Apple source), he'd better have a source inside Apple.

It's that or he's a MS double agent, sent here to utterly alienate the last of the Mac fanatics by raising hopes into the stratosphere (hopes that Apple has done their best to deny) and then watching all that enthusiasm become hatred and disdain for Apple when they fail to get a machine out for a deadline they never promised :-)

In about 3 months, we'll see which it is.

[ 02-13-2003: Message edited by: Tom West ]</p>

[Ed M.]

As mentioned in some of my other posts.. Apple is in a much better position to bring 64-bit computing to the mainstream and consumer desktop environment. There doesn't seem to be any commitment or focus I the Wintel world regarding 64-bit Winders....

- Which architecture will they support? AMD (X86-64) or Itanic (IA-64)?

This leads to other questions and uncertainties....

- There certainly won't be two versions because that would mean that software developers will be asked to code for two different architectures and I don't see that happening, so which (if any) will they choose?

- Drivers ... Another issue. Which architecture to support?

- IA-64 (Itanic) not meant for the desktop, so why bother writing common commercial apps for it?

- If Microsoft chooses to go X86-64 for Winders then I'd say Intel is behind the 8-ball, because they seem hell-bent on getting Itanic to work out at any cost.

- If Microsoft chooses IA-64, where does that leave AMD? Dead. But maybe not because IA-64 really isn't designed for the desktop environment, which will likely continue with 32-bit Winders...

- the old 32-bit codebase from legacy apps ... They will run like crap providing a crappy experience, not to mention that both AMD and Intel will be fighting the MHz myth themselves ;-)

- What about all the people who are committed to the 32-bit machines that they just bought ? ;-)

Oh and the BIG question...

How much *more* consumer-confusion is this going to generate in the Wintelon world?

The questions keep mounting.. too much uncertainty... It isn't like that with Apple and 64-bit. It's just a matter of "when"

--
Ed M.

[DCQ]

[quote]Originally posted by Tom West:
<strong>

Well, absolutely every clue I can scrounge from Apple and elsewhere points to a 970 machine in the Jan - March 2004 timeframe, although it's all deduction, whisper and employee vague hints.

However, given the forcefulness of Transcendental Octothorpe statements that the 970 is coming, and soon, (which I'm willing to accept at face value, if he'll admit to having an Apple source), he'd better have a source inside Apple.</strong><hr></blockquote>

I've been a rumor-watcher for a while. And an Apple watcher since the Mac Plus days. Rumor's have a way of getting repeated and reiterated on various boards and sites, each referring to the other. Think about the recent utterly confident rumors about a tablet coming out at this MWSF. Result? Nothing whatsoever. Crack smoke from someone somewhere that got repeated, cross referenced, and accepted as holy writ.

Nothing against TransOct, but we have no way of verifying anything anyone says on these boards. Hell, the people who run these types of rumor sites are almost laughingstocks (some more than others) among most of us--and it's their *job* to dig up this stuff. Board posters have no accountability whatsoever. (For the record, the only "rumor" guy who I trust implicitly is Matthew Rothenburg, the original--the one and only--Mac the Knife.)

Powermac 970 at MWSF '04. If it comes out in the fall, we can all smile big, broad smiles. It won't come out this summer.

As for the "tricks up our sleeve" comment, think of it this way: the biggest trick of all would be managing to get another 10.5 months of life out of the G4! <img src="graemlins/lol.gif" border="0" alt="[Laughing]" /> :D

Next year, he'll be able to point to how successful Apple's portables have been.* 75% of their units or whatever it will be by that time. Then he can repeat how he was right about "the year of the portable" thing. Then he can say but this year will be different: "the year Intel bites the dust (TM)" or however they decide to market it. ;) Then he rolls out a single 1.8 GHz 970 tower, matches it to a top-of-the-line 3.6GHz P4, and the 970 beats it handily. Then he intros that as the...Low-End PowerMac 970. He then rolls out a Dual 2.2 GHz 970 beast. But instead of doing a bake-off, Steve's cell-phone rings, he answers it, hangs up, and announces that he has just been informed that Intel has preemptively filed for bankruptcy protection. :cool:

As a gesture of magnanimity, Steve will personally offer to upgrade at no charge the approximately 43 individuals who have actually bought a PowerMac in the previous 6 months. :p

-DCQ

*BTW, was anybody else coughing nervously when Steve raved about how its portables were an unusually large percent of its total units shipped? Ummm, Steve?...It's cause your towers blow!

[Amorph]

[quote]Originally posted by DCQ:
<strong>

I've been a rumor-watcher for a while. And an Apple watcher since the Mac Plus days. Rumor's have a way of getting repeated and reiterated on various boards and sites, each referring to the other. Think about the recent utterly confident rumors about a tablet coming out at this MWSF. Result? Nothing whatsoever. Crack smoke from someone somewhere that got repeated, cross referenced, and accepted as holy writ.</strong><hr></blockquote>

Absolutely true, and worth remembering.

[quote]<strong>*BTW, was anybody else coughing nervously when Steve raved about how its portables were an unusually large percent of its total units shipped? Ummm, Steve?...It's cause your towers blow!</strong><hr></blockquote>

PC Towers must really be sucking, then, because notebooks make up a <a href="http://news.com.com/2100-1040-898370.html" target="_blank">large and steadily increasing chunk</a> of PC sales, too. ;)

(To be fair, Apple's desktop/laptop split is more even, but still.)

[KidRed]

[quote]Originally posted by Tom West:
<strong>

Well, absolutely every clue I can scrounge from Apple and elsewhere points to a 970 machine in the Jan - March 2004 timeframe, although it's all deduction, whisper and employee vague hints.

However, given the forcefulness of Transcendental Octothorpe statements that the 970 is coming, and soon, (which I'm willing to accept at face value, if he'll admit to having an Apple source), he'd better have a source inside Apple.

It's that or he's a MS double agent, sent here to utterly alienate the last of the Mac fanatics by raising hopes into the stratosphere (hopes that Apple has done their best to deny) and then watching all that enthusiasm become hatred and disdain for Apple when they fail to get a machine out for a deadline they never promised :-)

In about 3 months, we'll see which it is.

[ 02-13-2003: Message edited by: Tom West ]</strong><hr></blockquote>

I side with Transcendental Octothorpe as the info I've had since last summer said this fall for the 970.

[Who]

Is there an Apple event the end of May or in June? Just wondering...

[Outsider]

[quote]Originally posted by Who:
<strong>Is there an Apple event the end of May or in June? Just wondering...</strong><hr></blockquote>
There's <a href="http://macworldexpo.com/macworld2003/V33/index.cvn" target="_blank">one</a> in mid July. :D

[Programmer]

[quote]Originally posted by Who:
<strong>Is there an Apple event the end of May or in June? Just wondering...</strong><hr></blockquote>

And one in <a href="http://developer.apple.com/wwdc/" target="_blank">May</a>.

[Programmer]

[quote]Originally posted by KidRed:
<strong>I side with Transcendental Octothorpe as the info I've had since last summer said this fall for the 970.</strong><hr></blockquote>

Engineering projects are notorious for slippage, but all the noises I've heard point to this fall as well. IBM's projected introduction of the 970 is 2H '03 which fits that exactly. Barring unexpected problems, that is.

[Algol]

I side with those predicting the fall/summer release date of the 970. I agree with that time frame for two reasons:
First, IBM said that they will be releasing Blade servers based on the 970 this year. Second, the 970 will go into production in March allowing a lot of time for apple to get a PowerMac out this year. March, April, May, June, JULY. Three and a half months between production and Powermac. If not in July, then between July and September.

I think many of us find it hard to believe that apple will every pull through the sink hole they are currently in. However, we must realize that Steve knows the performance problems are hurting sales and probable has been working on a solution for a number of years now. The PPC970 is the absolute perfect fit for apple. We won't have to wait much longer. I believe...

[mavster]

We know that for the most part, Apple has been shying away from new product introductions at Macworlds (aside from the new powerbooks), but wouldn't MWNY in july (assuming they attend) be the best and biggest stage with which to unveil their new 64 bit OS and hardware?!?

Fits well with the projected IBM 2H03 release date and a nice six month refresh.. and I'm sure that when they do move the towers and powerbooks over to the 970, the imacs and ibooks would have a jump start on being scaled as we know they've got a 1.42 ghz part out now, with the 7457 looming to scale a bit..

Wouldn't it be nice to see a lineup of 1.8 ghz (970 tower), 1.2 ghz (970-low power) pbook, 1.42 ghz (g4) imac, and a 1 ghz (g4) ibook???

[Nevyn]

<a href="http://macworldexpo.com/macworld2003/V33/index.cvn" target="_blank">Macworld Expo New York</a>, mid July 03.

<a href="http://developer.apple.com/wwdc/" target="_blank">Apple's Worldwide Developer Conference</a> May 19-23, 03.

<a href="http://www.seyboldseminars.com/sf2003/" target="_blank">Seybold Publishing Conference</a> Sept 22-25 03. The G4's were introduced at Seybold 99.

<a href="http://www.namm.com/tradeshow/" target="_blank">NAMM, Music trade show</a> July 18-20, 03.

I've forgotten the name of the education conference, but they're unlikely to introduce the 970 there.

[costique]

[quote]Originally posted by DCQ:
<strong>Nothing against TransOct, but we have no way of verifying anything anyone says on these boards.</strong><hr></blockquote>

We don't need to verify anything, we need to know what to expect. These days many good people suffer for breaking their NDA and I don't think the end quite justifies the means. On the other hand, we are able to deduce some more or less reliable picture from official information. And this time everything points to 970 coming to Macs this fall. Did you notice nobody questions the idea of 970 in Macs, despite I can't remember any spokesman announcing that? Well, this might be an obvious example, but anyway...

[boy_analog]

Of course, autumn in the Southern Hemisphere starts in March. Looks like you Northerners will have to wait an extra 6 months or so for your 970s.

:p

[ 02-14-2003: Message edited by: boy_analog ]</p>

[MacRonin]

I would love to see Apple announce, demo AND sell the PowerMac 970 Workstation at this venue...

<a href="http://www.siggraph.org/s2003/" target="_blank">http://www.siggraph.org/s2003/</a>

Since Apple is not listed as an exhibitor, I doubt it would happen...

And the date of the show (July 27 - 31) is a bit early for the 'projected' 970 intro anyway...

Maybe the Alias|wavefront booth will have sealed units demoing Maya for OS X, with the ASP locked out...

;^p

[gar]

[quote]Originally posted by Who:
<strong>Is there an Apple event the end of May or in June? Just wondering...</strong><hr></blockquote>

yes, the introduction of the new ibook. :D

(or was it in april...?)

[Programmer]

You all do realize that Apple doesn't need some event to roll out new product, don't you? The original iMac, after all, was introduced at an Apple "Special Event" at the end of April. WWDC is interesting because that is usually where they preview upcoming system software technologies (like a 64-bit OS, for example) and often the associated hardware. Other than that the hardware comes when the hardware is ready.

[os10geek]

Apple, without question, will use the 970. Moto has treated the PPC as an integrated solution not a workstation solutio. Motorola thinks "Who brags about the speed of their VCR's clock speed?", and therefor won't advance the PPC as a computer chip. IBM, on the other hand, is focused on high-end server performance. :)

[Lemon Bon Bon]

[quote] With AMD's 64-bit plans drifting back and Microsoft apparently in no particular rush to bring a 64-bit Windows to market, there exists an opportunity too good to miss.

<hr></blockquote>

Lemon Bon Bon :cool:

[Algol]

Thats right, Windows, Intel, AMD, Dell, HP, Sony, Silicon vally in general, here we COME! Royal beatings! Yum...
:D <img src="graemlins/lol.gif" border="0" alt="[Laughing]" /> ;)

[os10geek]

Apple will still have the challenge of proving the "Megahertz Myth" even though the 970 will whip both the Xeon and the 'Hammer's butts. But this time through, it won't be as hard, because Apple can say "...the PowerPC 970 is a proven server workhorse...". Having a server processor in a desktop is certainly something that gives you bragging rights. <img src="graemlins/lol.gif" border="0" alt="[Laughing]" />

[hotboxd]

hush...

[Amorph]

Just testing, since I'm seeing a post by hotboxd in the forums index, but I can't see it in the thread proper.

[edit by Amorph: Well, hot damn, there it is. I'm going to close this thread since it seems to be getting flaky. Darned UBB. Please continue in the similarly-named thread that I'll start up]

[ 02-17-2003: Message edited by: Amorph ]</p>