If they've been sampling for more than 6 mos now, what's to stop production ramp up by the beginning of next year with product announcement at MWSF and volume shipping by Feb?
Apple already has a new DDR capable chipset, so there really should be no reason for a delay there. They don't even have to rewrite the OS, it can run in 32 bit mode with patched address space (just like now). And even if they did, they'd have had samples to work with and could have 6 mos + into it with another 3 before MWSF.
How tough could it be?
No seriously, though, what would be the delay? Why is the MWSF time frame optimistic (much less, very optimistic)? It doesn't take a year to go from samples to production (unless something went horribly wrong, and they wouldn't be announcing it at MPF if it had, I should think).</strong><hr></blockquote>
Wasn;t the G4 announced one year before it was released?
Not sure if the reason but I know the G4 to be released in Jan has been in the works for quite a while. The new mobo has had it's issues.
Wasn;t the G4 announced one year before it was released?
Not sure if the reason but I know the G4 to be released in Jan has been in the works for quite a while. The new mobo has had it's issues.</strong><hr></blockquote>
Yes it was. I remember seeing benchmarks comparing it to a Pentium III in January 99, several months before Apple announced the first machines that used it.
Wasn;t the G4 announced one year before it was released? . . .
</strong><hr></blockquote>
When chips are presented at a forum like this, some are further along than others. What happened in the past for one chip does not predict the future for another. About all we might guess is that the GPUL is far enough along that the next Microprocessor Forum would be too late to present something brand new.
When chips are presented at a forum like this, some are further along than others. What happened in the past for one chip does not predict the future for another. About all we might guess is that the GPUL is far enough along that the next Microprocessor Forum would be too late to present something brand new.</strong><hr></blockquote>
Ah, I was just trying to reason the info that I have about the GPUL being released next fall with the relationship to past history
I'm hopeful. No reason for Apple to wait so long on a chip that's been in prototypes since at least March. That's the real predictor, not the time of the MPF talk, but when prototypes began appearing.
That IBM factory went on line this summer. That's 6 months to MWSF to get production going. I don't see why it should take much longer. If it were Moto I'd understand but IBM is a pro at fabbing CPUs.
From what I can see given the above info... MOT has been slipping (tell us something we don't know). First a year then the next two revs were each year+ and said revs weren't exactly HUGE updates (just look at the G4 life line and it'll tell you that). Next we have IBM with the Sahara and in 6 months it's ready. Today we have IBM with GPUL... If you do wanna go by historic record then the GPUL will be ready 6 months from now.
<strong>If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.</strong><hr></blockquote>
Yummy in fact considering that Apple's 27th anniversary is April 1, 2003.
Motorola asserts that its partnership with two other chipmakers will probably be the first to produce a new generation of semiconductors, beating Intel by at least six months.
Earlier this week, Motorola, STMicroelectronics and Philips jointly unveiled a design for chips based on 90-nanometer circuitry, compared with the current 130-nanometer standard. Thinner circuitry makes each separate chip cheaper to produce, faster and more energy-efficient.
Chris Belden, vice president of Motorola's chip products sector, said the alliance would start production of a high-performance version of the 90-nanometer chips by the fourth quarter of this year and a lower-performing version in the third quarter of 2003.</strong><hr></blockquote>
Thank you for bringing this up. Seems a lot of people have completely given up on Motorola. I just wonder what Motorola considers high performance? High end networking a la the MPC 8560 or the MPC8540 Integrated Host processor or a as yet unannounced version for the desktop/network market??
Just a little factoid. Motorola alleges that they have been using a 0.13µ(HiP7) process on the cores of embedded processors since April 9th 2001 and that the HiP7 process includes some parts as small as 0.07µ. This was in a press release on 4/9/01.
My question is what the heck is taking Motorola sooooo loooonnnnnngggg to implement any of this for the G4????????
<strong>If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.</strong><hr></blockquote>
Another wrinkle - Apple's fiscal year starts September 1, so it's already "2003" in Apple's books. Some of the sources quoted said "2nd half of 2003" (IIRC) for when Macs might be released with this chip. Now if they were deliberately trying to misdirect us by getting us to assume Calendar 2003, when instead they meant Fiscal 2003, we get.....
April.
Hmmmmm....
I am an unflagging optimist. I'll be disappointed (but not surprised, I guess) if this chip doesn't arrive before September next year, so I'll bravely hope for this chip sooner rather than later.
Not sure of the sources of this link but they are talking planned for 2002, so maybe the announcement at the forum will be that it's pretty much ready (the chip that is, not a Mac based on it). That shouldn't take too long, as has been said before in this topic, that pre-production boxes have been out there for a while now.
How do you know they have been sampling production output? If true, you could be right about the date. More likely they had good test prototypes six months ago, and production may or may not have started.</strong><hr></blockquote>
Rumor had it that Apple had prototypes 6 mos ago. And the eWeek article claims Apple has mobos with the chip in development now. I read that to mean that the chip was sampling 6 mos ago since true prototypes are for in-house distribution during the design phase. You don't give out prototypes to your customer when what the prototype reveals may mean a redesign that coulld affect the features of the chip. (Like whether it has a VMX unit or not.)
That plus the fact that if Apple does have samples now (as the eWeek article indicates) it's likely the chip has been sampling for some time now.
. . . That plus the fact that if Apple does have samples now (as the eWeek article indicates) it's likely the chip has been sampling for some time now.
</strong><hr></blockquote>
I have been trying to understand the terminology of this industry, so I'd welcome an expert's input. I understood that sampling is where IBM makes chips in their final production facility, and samples many small runs to to see how they test out for things like yield and clock rate. It is the final step before shipping chips to be used in Apple manufacturing facility. This is what somebody told me, and if it is wrong I'd like to find out the real scoop.
So the fact that Apple has sample chips does not mean that IBM has been running final production batches of the chip. Apple would naturally have sample chips to work on just as soon as IBM has some that are representative of what they expect it to be. I would call these good prototypes, and that is what Apple has been working with, and is what the article refers to, IMO.
Now, having said that, it is still possible that they have progress to the final production facility, and are indeed sampling right now.
"Motorola to reach 0.09-microns before the rest of the industry? That may be their plan, but they're destined to fall behind schedule and fail."
Ouch. But the 'slipping' schedule timelines above show they earned that. And more.
Are we surprised 'people have given up' on Moto'?
People keep saying 'don't count them out'. They're counting themselves out. They're on the floor, losing hundreds of millions per quarter and there's no 'G5' coming from them. The last iteration of G4 show we'll be lucky to get .13/.9 from them. And if we do? I won't care. I've set my sights on the GPUL. A 1.8 gig G4 should keep the 'power'Mac going for half a year and quickly pensioned off into the consumer line.
I'm curious about this sampling issue. If prototypes have been with Apple since March/April this year...and the factory came online over the Summer...then what's stopping a ramp for San Fran'? Well, we thought the iMac would come six months sooner...at 'New York'...it didn't turn up until San Fran'...six months later. Design and flat panels seemed to be the issues there.
How long does it take from sampling to production?
Sure the G3 was announced and then delivered six months ish later. But the GPUL is going to be a 'new' chip and is certainly more complex. So a year from Summer 2002 factory online to New York annoucement 2003? Seems a fair cop.
Reports indicate IBM won't be delivering it to customers until 2nd half 2003.
Apple are a company that like to 'get things right' judging by the 'info' Ives video'. You can see why the iMac got delayed.
A new 'power'Mac case, new memory, controller, bandwidth...new cpu...new motherboard...considering how long it took them to put together the current 'bodge job' 'power'Macs then I can see why San Fran' is 'overly optimistic'.
April is my optimistic. But announcements of this weight? New York 2003.
Then...you've got the 'loose cannon' Moto-friggin'-rola. Despite trashing the development of a fairly decent cpu (y'know...boast announcements to market delivery...) when first announced... maybe...just maybe they'll (somehow...and with somebody's help, note...) 'go out' with a 'bang'. No...not .13...they'll give us .9 and a 2 gig G4/G5 thing on Rio? I'm highly skeptical. (Though I can't think why...)
Perhaps a .13 to take the G4 to 1.5 ish? Or a .9 to go to 1.8? Well to halt the arrest of 'power'Mac sales and actually get them going forwards again...well...a think something drastic is going to be needed. I don't think another die shrink G4 (even on a decent memory controller...) is the solution. It's running to stand still territory for me. Your mileage may differ.
Moki's mood music doesn't hold out any candle for a 'true' G5 from Motorola. A G4 on better throughput. And the mhz could be anything from 1.25 (heh. No. I know it's not funny...) to ? For those people who think the 'bung gaps' are 'good' then hey, these 'Rio G4s' will smoke their brains. I think I'll pass. Not a bad revision...but not startling.
What's concerning me is that we may get .13 G4s in Jan'.
.9s in the Summer.
And the GPUL gets sectioned into some weird £5000 class or uberworkstation.
More likely, the .13 G4 will be the last 'power'Mac hurrah for the G4 circa San Fran 2003 before .13 G4s and subsequent .9 G4s go into consumer lines over the next year or so.
I think the GPUL will be the G5 PowerMac and the G4 with Rio will look okay in the iMacs...especially on .13 or .9 revisions.
It's obvious one way or another that the next twelve months are going to be exciting ones for Mac users. We'll finally have our chance to crow.
The old G4/G3 tiered matrix line is looking very tired, very long in the tooth and how old is the G4 now anyhow? 3 years old?
A GPUL G5 and a Rio G4 consumer looks very much better in my estimation. Macs are overdue this move and certainly, Jaguar deserves this level of kit. People probably won't be mumbling about 'finder snap' on such machines.
Interesting forum discussion on Macrumors.com re: GPUL ie how good it will compared to Itanium 2, Clawhammer, its superscalar potential and the 64 bit running 32 bit applications issue.
If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.
</strong><hr></blockquote>
If IBM could have picked the date to present this chip, it probably would not be October 15th. That just happens to be when the forum takes place. The presentation and the development cycle are not related in any way.
With the impact this presentation will have, we can safely assume IBM picked the latest forum they could, meaning that by the next years forum, this chip will be old news.
Comments
<strong>
Eh, why?
If they've been sampling for more than 6 mos now, what's to stop production ramp up by the beginning of next year with product announcement at MWSF and volume shipping by Feb?
Apple already has a new DDR capable chipset, so there really should be no reason for a delay there. They don't even have to rewrite the OS, it can run in 32 bit mode with patched address space (just like now). And even if they did, they'd have had samples to work with and could have 6 mos + into it with another 3 before MWSF.
How tough could it be?
No seriously, though, what would be the delay? Why is the MWSF time frame optimistic (much less, very optimistic)? It doesn't take a year to go from samples to production (unless something went horribly wrong, and they wouldn't be announcing it at MPF if it had, I should think).</strong><hr></blockquote>
Wasn;t the G4 announced one year before it was released?
Not sure if the reason but I know the G4 to be released in Jan has been in the works for quite a while. The new mobo has had it's issues.
<strong>
Wasn;t the G4 announced one year before it was released?
Not sure if the reason but I know the G4 to be released in Jan has been in the works for quite a while. The new mobo has had it's issues.</strong><hr></blockquote>
Yes it was. I remember seeing benchmarks comparing it to a Pentium III in January 99, several months before Apple announced the first machines that used it.
<strong>
Wasn;t the G4 announced one year before it was released? . . .
</strong><hr></blockquote>
When chips are presented at a forum like this, some are further along than others. What happened in the past for one chip does not predict the future for another. About all we might guess is that the GPUL is far enough along that the next Microprocessor Forum would be too late to present something brand new.
<strong>
When chips are presented at a forum like this, some are further along than others. What happened in the past for one chip does not predict the future for another. About all we might guess is that the GPUL is far enough along that the next Microprocessor Forum would be too late to present something brand new.</strong><hr></blockquote>
Ah, I was just trying to reason the info that I have about the GPUL being released next fall with the relationship to past history
The G4e/V'ger/7450 was announced at 1999 MPF (16 months before release).
The Apollo/7455 was announced at 2000 MPF (16 months before release).
The Sahara/750FX was announced at 2001 MPF (6 months before release).
Barto
Because to ensure compatibility, Apple released a program "SimG4" to emulate a G4 on a G3 CPU before the Power Mac G4 was announced.
Barto
I'm hopeful. No reason for Apple to wait so long on a chip that's been in prototypes since at least March. That's the real predictor, not the time of the MPF talk, but when prototypes began appearing.
That IBM factory went on line this summer. That's 6 months to MWSF to get production going. I don't see why it should take much longer. If it were Moto I'd understand but IBM is a pro at fabbing CPUs.
<strong>The G4/Max/7400 was announced at the 1998 MPF (1 year before release).
The G4e/V'ger/7450 was announced at 1999 MPF (16 months before release).
The Apollo/7455 was announced at 2000 MPF (16 months before release).
The Sahara/750FX was announced at 2001 MPF (6 months before release).
Barto</strong><hr></blockquote>
Given this... I guess the 7460 err 7470 or whatever the current G4++ rumor is... isn't gonna happen any time soon either then.
Dave
<strong>The G4/Max/7400 was announced at the 1998 MPF (1 year before release).
The G4e/V'ger/7450 was announced at 1999 MPF (16 months before release).
The Apollo/7455 was announced at 2000 MPF (16 months before release).
The Sahara/750FX was announced at 2001 MPF (6 months before release).
Barto</strong><hr></blockquote>
Notice Motorola chips and their timeframes and then look at the IBM chip. Hmm..
--------------------------------------------
MPF 1998 = MOT = 7400. = 1 Year 0 Months
MPF 1999 = MOT = 7450. = 1 Year 6 Months
MPF 2000 = MOT = 7455. = 1 Year 6 Months
MPF 2001 = IBM = 750fx = 0 Year 6 Months
MPF 2002 = IBM = GPUL. = ? Year ? Months
--------------------------------------------
From what I can see given the above info... MOT has been slipping (tell us something we don't know). First a year then the next two revs were each year+ and said revs weren't exactly HUGE updates (just look at the G4 life line and it'll tell you that). Next we have IBM with the Sahara and in 6 months it's ready. Today we have IBM with GPUL... If you do wanna go by historic record then the GPUL will be ready 6 months from now.
Dave
[ 10-11-2002: Message edited by: DaveGee ]</p>
<strong>If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.</strong><hr></blockquote>
Yummy in fact considering that Apple's 27th anniversary is April 1, 2003.
<strong>
Motorola asserts that its partnership with two other chipmakers will probably be the first to produce a new generation of semiconductors, beating Intel by at least six months.
Earlier this week, Motorola, STMicroelectronics and Philips jointly unveiled a design for chips based on 90-nanometer circuitry, compared with the current 130-nanometer standard. Thinner circuitry makes each separate chip cheaper to produce, faster and more energy-efficient.
Chris Belden, vice president of Motorola's chip products sector, said the alliance would start production of a high-performance version of the 90-nanometer chips by the fourth quarter of this year and a lower-performing version in the third quarter of 2003.</strong><hr></blockquote>
Thank you for bringing this up. Seems a lot of people have completely given up on Motorola. I just wonder what Motorola considers high performance? High end networking a la the MPC 8560 or the MPC8540 Integrated Host processor or a as yet unannounced version for the desktop/network market??
Just a little factoid. Motorola alleges that they have been using a 0.13µ(HiP7) process on the cores of embedded processors since April 9th 2001 and that the HiP7 process includes some parts as small as 0.07µ. This was in a press release on 4/9/01.
My question is what the heck is taking Motorola sooooo loooonnnnnngggg to implement any of this for the G4????????
[ 10-11-2002: Message edited by: rickag ]</p>
<strong>If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.</strong><hr></blockquote>
Another wrinkle - Apple's fiscal year starts September 1, so it's already "2003" in Apple's books. Some of the sources quoted said "2nd half of 2003" (IIRC) for when Macs might be released with this chip. Now if they were deliberately trying to misdirect us by getting us to assume Calendar 2003, when instead they meant Fiscal 2003, we get.....
April.
Hmmmmm....
I am an unflagging optimist. I'll be disappointed (but not surprised, I guess) if this chip doesn't arrive before September next year, so I'll bravely hope for this chip sooner rather than later.
<a href="http://www.netneurotic.de/mac/powerpc.html"; target="_blank">GPUL info</a>
Not sure of the sources of this link but they are talking planned for 2002, so maybe the announcement at the forum will be that it's pretty much ready (the chip that is, not a Mac based on it). That shouldn't take too long, as has been said before in this topic, that pre-production boxes have been out there for a while now.
<strong>
How do you know they have been sampling production output? If true, you could be right about the date. More likely they had good test prototypes six months ago, and production may or may not have started.</strong><hr></blockquote>
Rumor had it that Apple had prototypes 6 mos ago. And the eWeek article claims Apple has mobos with the chip in development now. I read that to mean that the chip was sampling 6 mos ago since true prototypes are for in-house distribution during the design phase. You don't give out prototypes to your customer when what the prototype reveals may mean a redesign that coulld affect the features of the chip. (Like whether it has a VMX unit or not.)
That plus the fact that if Apple does have samples now (as the eWeek article indicates) it's likely the chip has been sampling for some time now.
<strong>
. . . That plus the fact that if Apple does have samples now (as the eWeek article indicates) it's likely the chip has been sampling for some time now.
</strong><hr></blockquote>
I have been trying to understand the terminology of this industry, so I'd welcome an expert's input. I understood that sampling is where IBM makes chips in their final production facility, and samples many small runs to to see how they test out for things like yield and clock rate. It is the final step before shipping chips to be used in Apple manufacturing facility. This is what somebody told me, and if it is wrong I'd like to find out the real scoop.
So the fact that Apple has sample chips does not mean that IBM has been running final production batches of the chip. Apple would naturally have sample chips to work on just as soon as IBM has some that are representative of what they expect it to be. I would call these good prototypes, and that is what Apple has been working with, and is what the article refers to, IMO.
Now, having said that, it is still possible that they have progress to the final production facility, and are indeed sampling right now.
Ouch. But the 'slipping' schedule timelines above show they earned that. And more.
Are we surprised 'people have given up' on Moto'?
People keep saying 'don't count them out'. They're counting themselves out. They're on the floor, losing hundreds of millions per quarter and there's no 'G5' coming from them. The last iteration of G4 show we'll be lucky to get .13/.9 from them. And if we do? I won't care. I've set my sights on the GPUL. A 1.8 gig G4 should keep the 'power'Mac going for half a year and quickly pensioned off into the consumer line.
I'm curious about this sampling issue. If prototypes have been with Apple since March/April this year...and the factory came online over the Summer...then what's stopping a ramp for San Fran'? Well, we thought the iMac would come six months sooner...at 'New York'...it didn't turn up until San Fran'...six months later. Design and flat panels seemed to be the issues there.
How long does it take from sampling to production?
Sure the G3 was announced and then delivered six months ish later. But the GPUL is going to be a 'new' chip and is certainly more complex. So a year from Summer 2002 factory online to New York annoucement 2003? Seems a fair cop.
Reports indicate IBM won't be delivering it to customers until 2nd half 2003.
Apple are a company that like to 'get things right' judging by the 'info' Ives video'. You can see why the iMac got delayed.
A new 'power'Mac case, new memory, controller, bandwidth...new cpu...new motherboard...considering how long it took them to put together the current 'bodge job' 'power'Macs then I can see why San Fran' is 'overly optimistic'.
April is my optimistic. But announcements of this weight? New York 2003.
Then...you've got the 'loose cannon' Moto-friggin'-rola. Despite trashing the development of a fairly decent cpu (y'know...boast announcements to market delivery...) when first announced... maybe...just maybe they'll (somehow...and with somebody's help, note...) 'go out' with a 'bang'. No...not .13...they'll give us .9 and a 2 gig G4/G5 thing on Rio? I'm highly skeptical. (Though I can't think why...)
Perhaps a .13 to take the G4 to 1.5 ish? Or a .9 to go to 1.8? Well to halt the arrest of 'power'Mac sales and actually get them going forwards again...well...a think something drastic is going to be needed. I don't think another die shrink G4 (even on a decent memory controller...) is the solution. It's running to stand still territory for me. Your mileage may differ.
Moki's mood music doesn't hold out any candle for a 'true' G5 from Motorola. A G4 on better throughput. And the mhz could be anything from 1.25 (heh. No. I know it's not funny...) to ? For those people who think the 'bung gaps' are 'good' then hey, these 'Rio G4s' will smoke their brains. I think I'll pass. Not a bad revision...but not startling.
What's concerning me is that we may get .13 G4s in Jan'.
.9s in the Summer.
And the GPUL gets sectioned into some weird £5000 class or uberworkstation.
More likely, the .13 G4 will be the last 'power'Mac hurrah for the G4 circa San Fran 2003 before .13 G4s and subsequent .9 G4s go into consumer lines over the next year or so.
I think the GPUL will be the G5 PowerMac and the G4 with Rio will look okay in the iMacs...especially on .13 or .9 revisions.
It's obvious one way or another that the next twelve months are going to be exciting ones for Mac users. We'll finally have our chance to crow.
The old G4/G3 tiered matrix line is looking very tired, very long in the tooth and how old is the G4 now anyhow? 3 years old?
A GPUL G5 and a Rio G4 consumer looks very much better in my estimation. Macs are overdue this move and certainly, Jaguar deserves this level of kit. People probably won't be mumbling about 'finder snap' on such machines.
Interesting forum discussion on Macrumors.com re: GPUL ie how good it will compared to Itanium 2, Clawhammer, its superscalar potential and the 64 bit running 32 bit applications issue.
Lemon Bon Bon
[ 10-11-2002: Message edited by: Lemon Bon Bon ]
[ 10-11-2002: Message edited by: Lemon Bon Bon ]</p>
<strong>
If they stick to the 6 month time frame, the GPUL could rear its beautiful head sometime in April... sounds delicious.
</strong><hr></blockquote>
If IBM could have picked the date to present this chip, it probably would not be October 15th. That just happens to be when the forum takes place. The presentation and the development cycle are not related in any way.
With the impact this presentation will have, we can safely assume IBM picked the latest forum they could, meaning that by the next years forum, this chip will be old news.
[ 10-11-2002: Message edited by: snoopy ]</p>