PDA

View Full Version : 90nm PPC 970's & Altivec 2




MacRumors
Aug 3, 2003, 06:54 AM
MacBidouille claims (http://www.hardmac.com/niouzcontenu.php?date=2003-08-03#322) tht the 90nm PowerPC 970 will start production in November, with volume production by January 2004. The site claims PowerBooks will utilize the smaller processor by Feb/March 2004.

In addition, MacBidouille reports that the specs for Altivec 2 have been completed, and should be implemented in 2005, with the PPC 990.



arn
Aug 3, 2003, 07:02 AM
Page 2'd because of the usual reasons - macbid has variable accuracy. This rumor is unclear, and difficult to verify, as with many of MacBidouille's rumors:

Did PPC 970's really start production in April, as MacBid claimed (http://www.macrumors.com/pages/2003/04/20030409083217.shtml)?

Will Apple sue Motorola? (http://www.macrumors.com/pages/2003/04/20030415134547.shtml)

What happened to the 1.4GHz PowerMac 970's which rolled off assembly lines (http://www.macrumors.com/pages/2003/05/20030502173926.shtml) in May?

dstorey
Aug 3, 2003, 07:03 AM
I wonder if Altivec 2 will be fully code compatible with Altivec, or whether programs will have to change thier code to get the advantages of it over vanilla altivec

Brother Mugga
Aug 3, 2003, 08:01 AM
As has been stated several times in previous posts, it wouldn't be that surprising if a 90nm G5 popped out in January. It all depends on what Ned Flanders (the IIIrd) from IBM meant at the WWDC when he said that engineers were already "working on the next generation of chips."

Does 'working' mean 'sampling' and does 'next generation' mean '90nm 970' or '980' (or whatever the Power5 derivative will be called).

Although I'm still getting a dual G5 (switcher - can't put up with windows for another 5 months...), the evidence that it will top out around 2.5Gig would certainly seem to necessitate a chip-change if Jobs' pledge to be at 3Gig by next summer is to be realised. (In fact, given that Stevie doesn't exactly delight in being made to look like a nobber [q.v. Motorola], I feel this prediction might, in fact, be on the conservative side, but there you go...)

Interesting rhetorical observations r.e. MacBids recent claims, by the way Arn - particularly with respect to the 1.4 970s. I know I've asked this before, but does anyone have any idea whether the power dissipation of a 1.0/1.2/1.4 970 would allow it to be shoved into a Powerbook? (I don't want one, and I doubt they'll do it before going 90nm: I'm just nosey :).)

Cheers

Brother Mugga

PS: Or, indeed, and iMac.

bennetsaysargh
Aug 3, 2003, 08:41 AM
i can't wait. i think tht those 1.4Ghz G5s will be pouut into iMacs soon.
at least h*opefully:)

abdul
Aug 3, 2003, 09:22 AM
power comsumption in 1.2 or was it 1.4 GHz g5 is ~18 watts

i dont know how many watts the current G4 produces

by knowing how many watts the g4 produces we can guess.

tom.96
Aug 3, 2003, 09:27 AM
Originally posted by bennetsaysargh
i can't wait. i think tht those 1.4Ghz G5s will be pouut into iMacs soon.
at least h*opefully:)

I agree! A 1.4 ghz G5 in an imac would be ideal, a powerful consumer machine and it would also differentiate it from the emac which could take the faster G4 which has been reported. I'm in the situation where I can't afford a G5 PM but I still want a decent machine to last 3-4 years, and a G5 imac would fit that nicely.

bennetsaysargh
Aug 3, 2003, 09:45 AM
here is what i see once the iMac goes G5

PowerMac-G5
iMac-G5
Powerbook-G5
eMac-new G4
iBook-new G4

i can't wait untill the iBooks go G4!

MrMacMan
Aug 3, 2003, 10:03 AM
Good news for all people who want the powerbook in G5...

They would have to lower the clock speed by alot...

But who knows they 'needed' 4 fans to keep the G5 cooled...

That heatsink is huugee.

akushlan
Aug 3, 2003, 10:22 AM
Originally posted by bennetsaysargh
here is what i see once the iMac goes G5

PowerMac-G5
iMac-G5
Powerbook-G5
eMac-new G4
iBook-new G4

i can't wait untill the iBooks go G4!

They aren't going to put G4s in the iBook, instead they will put the G3+

Mr. Anderson
Aug 3, 2003, 10:28 AM
I like the idea of no Moto chips in any Apple product. The sooner they get away from the G4 the better.

If this all turns out to be true it will be great for Apple....:D

D

Fender2112
Aug 3, 2003, 10:31 AM
Originally posted by Macrumors
The site claims PowerBooks will utilize the smaller processor by Feb/March 2004.

That's funny.:D

Brother Mugga
Aug 3, 2003, 11:07 AM
Originally posted by abdul
power comsumption in 1.2 or was it 1.4 GHz g5 is ~18 watts

i dont know how many watts the current G4 produces

by knowing how many watts the g4 produces we can guess.

Cheers, mate.

I reckon that's do-able.

So I guess we shall see come September?

Brother Mugga

jzieske
Aug 3, 2003, 11:34 AM
Ok maybe I am wrong but, aren't there rumors out there saying that 970's for PM G5's still have not started production at Fishkill. One would then assume that an R&D lab must be making all of the chips for them. So now I am supposed to believe that 980's will be in mass production in Jan/Feb 2004. Sorry, I don't buy it.

[edit: fixed spelling]

jaedreth
Aug 3, 2003, 12:34 PM
Now it's been said here before that G3+ is an IBM revved G3 with Altivec.

It's also been said that G4 is nothing but G3 + Altivec.

I'm not going to assume both are right, but if they are, then why can't IBM rev the G3+ into its own G4, and the iBooks and eMacs use that?

(Unless of course the performance of said chips is slower than the previous G4 chips, because it would be a downgrade for eMac since they already use G4.)

Jaedreth

RHutch
Aug 3, 2003, 12:40 PM
Originally posted by jzieske
Ok mayb I am wrong but, aren't there rumors out there saying that 970's for PM G5's still have not started production at Fishkill. One would then assume that an R&D lab maybe making all of the chips for them. So now I am supposed to believe that 980's will be in mass production in Jan/Feb 2004. Sorry, I don't buy it.

The rumor about the chips not being in production at Fishkill was based on when someone took a tour of the facility; no one knows when that tour was. So we know that the chips weren't being produced there at some time, but not how long ago that was.

Also, where did the rumor for this thread say that 980's were going to be in mass production by Jan/Feb 2004? The rumor said that 90 nm 970's would be produced at that time. Do you buy this?

daveL
Aug 3, 2003, 12:52 PM
Originally posted by jzieske
Ok mayb I am wrong but, aren't there rumors out there saying that 970's for PM G5's still have not started production at Fishkill. One would then assume that an R&D lab maybe making all of the chips for them. So now I am supposed to believe that 980's will be in mass production in Jan/Feb 2004. Sorry, I don't buy it.
They're not talking about the 980, they're talking about moving the 970 from 130 nm to 90 nm. You might consider this a 970+, but not a 980. Most consider the 980 designation to be the chip that results from the Power5 development. Anyway, before we get on a three day exchange over what model numbers mean what, let me say I don't give a rat's a$$. The important thing is that the chip will transition to the 90 nm process.

I really don't consider that assertion to be out of line with other info concerning IBM's timeline for moving to 90 nm. They are supposed to be right behind Intel.

jzieske
Aug 3, 2003, 12:52 PM
Originally posted by RHutch
Also, where did the rumor for this thread say that 980's were going to be in mass production by Jan/Feb 2004? The rumor said that 90 nm 970's would be produced at that time. Do you buy this?

90 nm chips ... I am guess I still don't buy mass production in Jan/Feb 2004. That would mean only 3-4 months with thee current chips. Unless they don't put the 90nm chips in PM's.

jaedreth
Aug 3, 2003, 01:04 PM
The 90 nm chips would be perfect for portable units and consumer products.

Aka PowerBook G5 and iMac G5.

Jaedreth

daveL
Aug 3, 2003, 02:14 PM
Originally posted by jzieske
90 nm chips ... I am guess I still don't buy mass production in Jan/Feb 2004. That would mean only 3-4 months with thee current chips. Unless they don't put the 90nm chips in PM's.
What's wrong with a PM speed bump? I'm not sure where you get 3-4 months. With G5 shipments starting in August, we'd be looking at a 6 month update. Seems reasonable, even likely, to me.

Brother Mugga
Aug 3, 2003, 02:32 PM
To be honest, a three month upgrade cycle is precisely what Apple needs to remain competitive (a point that has also been hammered to death on these forums in the past). After all, it wouldn't exactly be revolutionary, more a case of "welcome to the rest of the PC world..."


The only downside would be having to listen to eeediots whining that their "old Powerbook now looks all weedy and underpowered" on a more regular basis (because there truly is no pleasing some people...jeeeez...).

Cheers

Brother Mugga

3G4N
Aug 3, 2003, 02:43 PM
Originally posted by jaedreth
It's also been said that G4 is nothing but G3 + Altivec.

It's been said, but it isn't necessarily correct.

IIRC, the G3 was based, in part, on the ppc603,
a streamlined approach, and not SMP-capable
(no dual-processor support).

The G4 was based more off the ppc604,
which was dual-capable, and added
the vector engine, altivec.

I simplify, but to answer your question...

I bet IBM is working on a nice G3 just like
they are working on a nice G5. Patience.
I don't know if you'll ever get a G4, per se, in an iBook,
but you will get much better processors. (like I'm saying a lot. : )

chazmox
Aug 3, 2003, 05:19 PM
Originally posted by arn
Page 2'd because of the usual reasons - macbid has variable accuracy. This rumor is unclear, and difficult to verify, as with many of MacBidouille's rumors:

Did PPC 970's really start production in April, as MacBid claimed (http://www.macrumors.com/pages/2003/04/20030409083217.shtml)?

Will Apple sue Motorola? (http://www.macrumors.com/pages/2003/04/20030415134547.shtml)

What happened to the 1.4GHz PowerMac 970's which rolled off assembly lines (http://www.macrumors.com/pages/2003/05/20030502173926.shtml) in May?

Thank you Arn! These explanations have been very helpful!!!

ZildjianKX
Aug 3, 2003, 08:15 PM
I wouldn't hold my breath for a G5 PB... there isn't even a G4 upgrade out yet... how sad.

Rincewind42
Aug 3, 2003, 08:20 PM
Originally posted by 3G4N
IIRC, the G3 was based, in part, on the ppc603, a streamlined approach, and not SMP-capable
(no dual-processor support).

The G4 was based more off the ppc604, which was dual-capable, and added the vector engine, altivec.

I bet IBM is working on a nice G3 just like they are working on a nice G5. Patience. I don't know if you'll ever get a G4, per se, in an iBook, but you will get much better processors. (like I'm saying a lot. : )

Ok, this has been beaten to death, but lets do it one more time...

The G3 was based on the 603. Simple.

The G4 is a moniker for over half a dozen different chips. The MPC7400, 7410, 744x series (not used on Macs), and the 745x series (used in various Mac G4 models).

The MPC7400 & 7410 were based primarily on the G3. Added to the G3 design was a floating point unit that essentially came form the PPC604 and Altivec. Thus if you really want to split hairs, the G4 was based on the G3, the PPC603 and the PPC604 and added Altivec to the mix. But for simplicity, it can be said that the original G4s were basically G3+Altivec.

All other G4s (744x and 745x) are evolutions of the original design. The 7 stage pipeline, the enhanced altivec unit, and the on-die L2 cache are decided differences from the original G3 & G4. Therefore, it would be unfair to consider the current G4 as "just a G3+Altivec".

And yes, it is entirely possible that IBM could make a G3+ that includes Altivec. Will they? Who knows. What will Apple call it? Almost certainly they will call it a G4. To call it anything else would simply cause market confusion (which is why the PowerMacs are G5 and not any other of the dozens of suggestions people may have made). Customers don't care about model numbers, they just want a simple one line explaination as to why computer A is better than computer B. 5 > 4 > 3 keeps it simple.

DeusOmnis
Aug 3, 2003, 08:33 PM
990's in 2005?? That would be awesome, since that's right when i'd be getting a new computer. I cant wait.... G7 sounds so.... *gargling sound*

bennetsaysargh
Aug 3, 2003, 08:52 PM
well, although we have waited how many years for the G5, we will only have to wait 2 more years to get to a G7? does that mean in 2020 we will be at the G12? no way. it might take a little longer than that? right?
unless if IBM is really fast, we won't be doing that.

QCassidy352
Aug 3, 2003, 09:42 PM
the G5s aren't even shipping and we talking about G7s? :rolleyes: By the time they come out, you won't want them; you'll want G8s.

Sorry, it just seems silly to me to speculate beyond a couple of months, or at most a year (since Jobs did use that time frame himself).

nagromme
Aug 4, 2003, 01:35 AM
17' PowerBook with:
Dual 2 Ghz G5s (only 1 when on battery)
256 MB video board
2-button trackpad with iPod-style scroller

Early 2004 please!

Docrjm
Aug 4, 2003, 02:27 AM
Originally posted by QCassidy352
the G5s aren't even shipping and we talking about G7s? :rolleyes: By the time they come out, you won't want them; you'll want G8s.

Sorry, it just seems silly to me to speculate beyond a couple of months, or at most a year (since Jobs did use that time frame himself).
We already have a G8, its just not a processor chip! :p

displaced
Aug 4, 2003, 02:53 AM
One thought that's rumbling round my mind re: the G5's cooling and the heat generation of current G5s.

The case design will no doubt outlive many generations of processor. From what I've heard so far, the current G5 is very quiet in operation.

What I'm suggesting is that we can't immediately assume the G5 is a heat-producing monster just because of the robust cooling system in the current model.

The case was probably designed with the future in mind. Not to mention, in a 'fully loaded' desktop system, there's going to be many other heat sources (HDDs, optical drives, high-end graphics cards etc.)

Anyway - just some random thoughts to factor in to the rumor mill ... :)

Plutoniq
Aug 4, 2003, 05:21 AM
What blows my mind is that in reading the specs of the new G5, there is 23 (Right?) stages in the instruction pipeline. When the G4 was Apples hottest thing on the block, they used comparisons between the P4's pipeline stages (20 or 21) and the 7 that the G4 has as a means of "proving" the g4 is more efficient per clock cycle.....it seems that this argument has been thrown out the door with the G5, as it now has more pipeline stages and a lower clock speed than the P4. So whats up with that? Did they just say "screw it, we need more perceived speed" and increase the pipeline stages to reach a higher mhz rating?

Take the current Gobi (PPC 750GX) which is now sampling...... only 5 pipeline stages, yet will clock to 1.1ghz....... forgetting altivec for a moment, does it make sense that the 750GX is more efficient per clock cycle than the G5? Yes the G5 has got a hell of a fast bus between the RAM and the CPU, but the 1.1ghz Gobi has a 1mb of L2 cache running at 1.1ghz (1:1). If the 750VX ever does make it out (Which will be 750GX @90nm with SIMD), who in here would still take a lower clocked G5 (in a powerbook) over a 750VX in, say, an Ibook......i know where i'd be plunkin' down my cash....which is to say that maybe they would never put that much power in a consumer machine that could really be competitive with the G5.

Seeya

p9

ewinemiller
Aug 4, 2003, 07:10 AM
Originally posted by nagromme
17' PowerBook with:
Dual 2 Ghz G5s (only 1 when on battery)
256 MB video board
2-button trackpad with iPod-style scroller

Early 2004 please!

I believed it could all someday be true, then you had to break my suspension of disbelief by saying 2-button trackpad. Then I knew it was just a fantasy. :D One of things that keeps me from buying an Apple laptop is their insistence on relying on 1984 mouse technology. On the desktop I can simply replace the mouse, but with a laptop I'm stuck.

pellucidity
Aug 4, 2003, 08:01 AM
Originally posted by Plutoniq
What blows my mind is that in reading the specs of the new G5, there is 23 (Right?) stages in the instruction pipeline. When the G4 was Apples hottest thing on the block, they used comparisons between the P4's pipeline stages (20 or 21) and the 7 that the G4 has as a means of "proving" the g4 is more efficient per clock cycle

If you read the discussion of the G5 design by Hannibal of ArsTechnica, he referred to the G4 as "wide and shallow", the P4 as "narrow and deep", and the G5 as "wide and deep". Basically, the argument that the G4 is more efficient per clock doesn't have a lot to do with the pipleline, though the deep pipeline of the P4 combined with the "narrowness" of the design is a weakness. It's why the G5 destroyed it in the WWDC demos. P4s choke on non-optimised code. Hannibal went into detail about why the G5 is less vulnerable.

DakotaGuy
Aug 4, 2003, 09:04 AM
Originally posted by nagromme
17' PowerBook with:
Dual 2 Ghz G5s (only 1 when on battery)
256 MB video board
2-button trackpad with iPod-style scroller

Early 2004 please!

You forgot one other feature, the small tube on the side that a hair dryer can be inserted in and put on "air only" to aid in cooling the beast when you really are using it hard.lol

bennetsaysargh
Aug 4, 2003, 09:10 AM
if they could, would they lower the amount of pipeline stages in the G5? how long would it take to do that?

RBR2
Aug 4, 2003, 09:28 AM
Originally posted by daveL
What's wrong with a PM speed bump? I'm not sure where you get 3-4 months. With G5 shipments starting in August, we'd be looking at a 6 month update. Seems reasonable, even likely, to me.

What Apple needs to do is continually be at work and incrementally upgrade the components/systems of the Macs being sold within a given model. Apple attaches more significance to the changing of a case, excuse me, enclosure, than most anyone else. It is the technology inside that counts most and the functionality of the case should be a priority instead of merely "style". There are many "stylish" products that do not work well.

When components change within the industry they should simply be incorporated into the shipping product. Waiting (and waiting is what Apple has done) to incorporate them in a new "model" is counterproductive.

neilw
Aug 4, 2003, 01:29 PM
if they could, would they lower the amount of pipeline stages in the G5? how long would it take to do that?


Number of pipeline stages tends to creep up over time, not down. In general, more pipeline stages are added to enable higher clock speeds. It would be wonderful to have a 1-stage design (i.e., no pipeline at all), which would be wickedly simple. Unfortunately, it'd be essentially impossible to make it go very fast, which is why you won't find a high-performance processor designed this way.

IBM would not have added any more stages to the 970 then they found necessary to make it work at the desired speeds. After having gone to the considerable (!) effort of implementing it like that, they'd be unlikely to then make a specific effort to shorten the pipelines on the 970. With each new generation, though (980, 990, etc.) the pipeline design will get revisited, so they could conceivably have shorter pipelines.

At this point, the die is cast with the 970. Though it definitely does have very long pipelines, it also has a heck of a lot of impressive hardware in there to mitigate the effects of that. It will be interesting to see what the 980 looks like; maybe there'll be some details at MPF when IBM presents the Power5.

panphage
Aug 4, 2003, 03:03 PM
Originally posted by RBR2
When components change within the industry they should simply be incorporated into the shipping product. Waiting (and waiting is what Apple has done) to incorporate them in a new "model" is counterproductive.

As the beer commercial says, "True."

Of course, Apple includes most "components" that change on the motherboard. So most of the time they have to make a new motherboard anyway. Why not wait and do a few different components at once and release a new model? Especially when you have to add support for the new tech in the OS anyway.

I prefer it RBR2's way, but I suspect my version is how Apple operates.

jaedreth
Aug 4, 2003, 03:51 PM
It is far more expensive for Apple to make a lot of little revs to the mlbs than one huge rev to the mlb, that's the reason for distanced updates.

Jaedreth

cb911
Aug 4, 2003, 04:44 PM
sounds good. IBM's sure got a good timeline for the development of the G5 and future chips for Apple.

and altivec2 requires the use of a 65nm process!! that's freakin' awesome!! :D

ClimbingTheLog
Aug 4, 2003, 10:49 PM
This corroborates Steve's (in)famous off-stage-but-on-mic comment at WWDC that they'll be at 3GHz by December.

bennetsaysargh
Aug 5, 2003, 07:33 AM
no, he said by ths time next year.

ClimbingTheLog
Aug 5, 2003, 08:32 AM
Originally posted by bennetsaysargh
no, he said by ths time next year.

No, that's what he said on-stage. I'm talking about what he said off-stage in a private conference while they were doing the sound checks on his microphone.

bennetsaysargh
Aug 5, 2003, 08:56 AM
well then, i guess we're looking at a december updaate for the powermacs.:)*

jaedreth
Aug 5, 2003, 05:54 PM
http://www.theregister.co.uk/content/3/32162.html


Check this out.

Jaedreth

Phil Of Mac
Aug 6, 2003, 07:10 PM
I'm just doing the 90nm happy dance. If IBM can get the processes down quickly, I'll be a very happy man.

macphoria
Aug 8, 2003, 01:57 PM
Maybe AltiVec was originally designed for 32bit processor, whereas AltiVec 2 will be specifically designed for 64bit processor?

bennetsaysargh
Aug 8, 2003, 02:10 PM
Originally posted by macphoria
Maybe AltiVec was originally designed for 32bit processor, whereas AltiVec 2 will be specifically designed for 64bit processor?

or another possibility is that altivec is optimized for 32-bit and altivec 2 is optimized for 64-bit.

macphoria
Aug 8, 2003, 02:33 PM
or another possibility is that altivec is optimized for 32-bit and altivec 2 is optimized for 64-bit.
Yes, definitely.

daveL
Aug 8, 2003, 02:33 PM
Originally posted by bennetsaysargh
or another possibility is that altivec is optimized for 32-bit and altivec 2 is optimized for 64-bit.
Or, maybe you should read up on what Altivec is? It's a 128-bit SIMD unit. It has nothing to do with whether the core is 32- or 64-bit. It essentially runs as a separate unit on the chip.

Morpheus
Aug 15, 2003, 02:15 AM
Originally posted by RBR2
What Apple needs to do is continually be at work and incrementally upgrade the components/systems of the Macs being sold within a given model. Apple attaches more significance to the changing of a case, excuse me, enclosure, than most anyone else. It is the technology inside that counts most and the functionality of the case should be a priority instead of merely "style". There are many "stylish" products that do not work well.

I agree with you that function is overall more important than the style. However, it is necessary to remember that if it wasn't for the "cool" image that Apple regained with the original iMac, it could very possibly not exist today. Until the G5, Apple was behind in terms of hardware. One of the main reasons they have been successful is because people like their sylish products. Its all part of the package that makes Apples that much better than Wintels.

Phil Of Mac
Aug 15, 2003, 02:33 AM
Why do you people go around resurrecting week-old threads?