The 3-level cache/memory is a nice feature for embedded applications.
The 745x series has had that capability all along with a 2 meg limit and the ability to divide it between local memory and cache as you see fit. The 7457 just adds a single address line to double the capacity to 4 megs, but they didn't bother to add the cache tags for the new memory so its only useful to embedded users. The 7457 is an absolutely minimal design change in order to get the processor to 0.13 micron, and they are still really really late. Moto probably has one junior design engineer (or co-op student) sitting in a corner somewhere with a notepad and paper, trying to figure out how to get the G4 onto the 0.13 micron process. We can only hope that the rest of the team is still at Motorola and is working on something more worthy of our attention.
Moto probably has one junior design engineer (or co-op student) sitting in a corner somewhere with a notepad and paper, trying to figure out how to get the G4 onto the 0.13 micron process. We can only hope that the rest of the team is still at Motorola and is working on something more worthy of our attention.
Wasn't the 7457 supposed to replace Maxbus with RapidIO? I suppose that feature has been moved further into vapor. Is it even slated for inclusion in any chips on Motorola's roadmap?
Wasn't the 7457 supposed to replace Maxbus with RapidIO? I suppose that feature has been moved further into vapor. Is it even slated for inclusion in any chips on Motorola's roadmap?
The 7457 RM was supposed to include these new features (at least a DDR memory controller). However the rumors said that this chip is cancelled, and it seems this chip has disappeared from Mot's roadmap. At the place they schedule a dual core G4 chip. I bet this chip will have RIO or the equivalent. In the contrary if this chip is stuck with maxbus, they should better put this project in a garbage.
The 7457 is an old design on a new fabbing process.
This chip will be used by Apple during the one year transition. New powerbook and i mac reaching 1,3 ghz will be showed at Paris. It seems that Mot is dead in the high end PPC market.
I realize this is nothing but wild guessing, but what does all this mean, in your opinion, re: G5 PowerBooks? "One year transition" meaning one year for the consumer lines to move to the 970, or one year for the portables? Or is the rumoured 1.3GHz G4 PowerBook to be the last of its line, replaced in January by a nice G5 part?
WHAT makes anyone think that we're going to get a new G4 PowerBook soon (by end of September, say) and THEN a whole new, redesigned G5 PowerBook four months later?
I mean, honestly.
Has Apple EVER done this? Look, I want to be wrong on this, but those of you putting all your eggs in the "G5 PowerBook at MWSF 2004" basket may be setting yourselves up for some major letdown. Look at your history. Read the specs. Look at the timetables.
If Apple updated ANY of their stuff on 4-month cycles, I'd fall over. It's more like 8-10 months...at BEST. And in some sad cases, it's been a full year!
Four months? I really, really doubt it. Again, hope I'm wrong. But it'll be a major departure from Apple. I still say summer or so of 2004 for any G5 portables. And that's being optimistic and giving IBM more confidence than Motorola (if it was Motorola, I'd say "G5 PowerBooks by autumn 2006...MAYBE...").
WHAT makes anyone think that we're going to get a new G4 PowerBook soon (by end of September, say) and THEN a whole new, redesigned G5 PowerBook four months later?
I mean, honestly.
Has Apple EVER done this?
Well, the PowerMac got one last refresh as a G4 platform only a few months before ceding the stage to the G5. The Smurf G3 had a similarly short life before the G4 rolled in. In other words, it seems to me that Apple will make exceptions to their normal update schedule for major technological shifts.
The 7457 can just be dropped in, so Apple could use it as a "refresh" rather than building a whole new machine around it. In that case, I could easily see them using the '57 to goose PB sales up for a little while and hit that over 50% portable sales goal before unleashing a totally new PowerBook. All Apple would have to do, pretty much, is stop ordering '55s and start ordering '57s, and maybe improve a few other drop-in options like hard drives. Nothing serious or disruptive, though.
The problem here is that even now we know so little about the 970 and its kin that it's impossible to say what Apple's next step is.
Quote:
Four months? I really, really doubt it. Again, hope I'm wrong. But it'll be a major departure from Apple. I still say summer or so of 2004 for any G5 portables. And that's being optimistic and giving IBM more confidence than Motorola (if it was Motorola, I'd say "G5 PowerBooks by autumn 2006...MAYBE...").
Four months would be pushing it, but I think summer '04 is pushing it the other way. A lot of it depends on things that we just don't know (how far along Apple's motherboard team is, whether a version of the 130nm 970 can work, when the switch to 90nm is happening), and that leaves a lot up in the air.
I will say that if everyone resigns themselves to a summer '04 update they're likely to be pleasantly surprised, and that's a good thing.
And at roughly the same clock rate it appears they have roughly the same power consumption (20W @ ~1.3 GHz).
970: 30W typical n.n. maximum
7457: 19W typical 26W maximum
I wouldn't call this roughly the same. IBM & Apple didn't publish maximum power consumtion for the 970 but for the 1.3 GHz part expect it to be near 50W. This chip is not only fast and it's hungry too!
Quote:
Apparently its because the G4 is an "all static design", whatever that means.
Throw in more transisors for the same functionality - get less power consumption - think of it as in static and dynamic RAM.
The issue is Maxbus maxxes at 200mhz IIRC. So all PowerBooks should move to 167mhz bus at minimum, 200mhz preferably, if it is possible. I mean I bought a 12" with DDR memory, yet it totally doesn't take advantage of it. Derrr...
the only question that matters is whether whatever [technical jargon goes here] that keeps the FSB so squeezed has been addressed.
has it?
Nope, nor will it be.
I still maintain that there is no real technical reason Apple can't put the current 970 in Powerbooks right now @ 1.0 and 1.2Ghz. The heat and power consumption characteristics would be somewhat less than optimal, but not unfeasible. Power and heat will gradually decrease in magnitude as the efficiency of the production line ramps up and more of the processes switch to 90 nm. So 970 Powerbooks by next spring, guaranteed
I wouldn't call this roughly the same. IBM & Apple didn't publish maximum power consumtion for the 970 but for the 1.3 GHz part expect it to be near 50W. This chip is not only fast and it's hungry too!
Where are you getting those numbers? I remember reading that a 1.2GHz 970 only used 19 W. So I highly doubt that a 1.3 GHz part would use 50W.
And if you consider the fact that the 970's power consumption at 1.2GHz is a mere 19W, it's almost certain that we'll see a future notebook from Apple based on the new chip.
Note in the chart right above, that a 1.8 GHz part dissipates 68.4W, so I think your numbers are just a tad off...
Amorph, I think summer 04 is just about right on the PB G5 front, assuming .09u 970's or some domesticated .13u schema.
Would a 1.2Ghz part have the same power consumption as a 1.8 running at 2/3rds? Or would it be higher since a lower spec part would need a higher voltage to switch than a "low power" 970? I'm sure this is muddled but you might just be able to decipher it. What I mean to say is that to get a good reliable 1.2Ghz part for a notebook, you might just need a 1.8 level part anyway (with the same costs). From what I've understood it isn't a simple matter of using lower spec parts when it comes to the heat and volatge requirements of a laptop.
That means that they have to wait on the G5 a bit, the pessimist in me says Q3-4 '04, but summer '04 seems right about on as far as a G5 PB goes, mebbe it will even feature a .13u G5 (a more expensive, and not particularly high clock one if it comes to that)
Of course .09u would be better, and Apple needs some time to work out speed/heat management on a new CPU. I don't think they'll ever make Alienware type "laptops" -- nor should they.
And, though it seems illogical, I'm pulling for Moto to get their stuff together by the time they move to .09u. IBM may be the bees knees right now, but a second option is needed, if only as a back-up.
IBM has spent a lot at fishkill, but they're own statements reveal that the money has been spent in FAB and DESIGN expertise, not neccessarily PPC fab and design, but for a facility that can do work for other chip makers. A model that becomes increasingly likely as fabrication costs escalate with each new generation. The future may have only a few "fabs" doing most of the production for a number of designers. IBM wants to be one of those. Intel and SMC will be others.
That's both good and bad for Apple. Bad, because despite the expenditure in Fishkill, IBM is NOT tied to the PPC. I've argued this before, they can switch at any time, or redirect the PPC to areas not of use to Apple, ie the POWER end of the equation. I don't think that will happen, but IBm has been disinterested in the past, and they could become so again
But it's good, because the future of chip production means that I can show up to IBM's door with a design, plunk down some change, work it out a bit, and get my chips fabbed. It might not be out of the question to see IBM making original Motorola PPC's for Apple's use in 3-4 years time. Along with their own chips, natch.
Comments
Originally posted by blabla
The 3-level cache/memory is a nice feature for embedded applications.
The 745x series has had that capability all along with a 2 meg limit and the ability to divide it between local memory and cache as you see fit. The 7457 just adds a single address line to double the capacity to 4 megs, but they didn't bother to add the cache tags for the new memory so its only useful to embedded users. The 7457 is an absolutely minimal design change in order to get the processor to 0.13 micron, and they are still really really late. Moto probably has one junior design engineer (or co-op student) sitting in a corner somewhere with a notepad and paper, trying to figure out how to get the G4 onto the 0.13 micron process. We can only hope that the rest of the team is still at Motorola and is working on something more worthy of our attention.
Originally posted by Programmer
Moto probably has one junior design engineer (or co-op student) sitting in a corner somewhere with a notepad and paper, trying to figure out how to get the G4 onto the 0.13 micron process. We can only hope that the rest of the team is still at Motorola and is working on something more worthy of our attention.
We can?
Originally posted by Ensign Pulver
We can?
They work at a dual core version of this chip.
Originally posted by Big Mac
Wasn't the 7457 supposed to replace Maxbus with RapidIO? I suppose that feature has been moved further into vapor. Is it even slated for inclusion in any chips on Motorola's roadmap?
The 7457 RM was supposed to include these new features (at least a DDR memory controller). However the rumors said that this chip is cancelled, and it seems this chip has disappeared from Mot's roadmap. At the place they schedule a dual core G4 chip. I bet this chip will have RIO or the equivalent. In the contrary if this chip is stuck with maxbus, they should better put this project in a garbage.
Originally posted by Powerdoc
(snip)
The 7457 is an old design on a new fabbing process.
This chip will be used by Apple during the one year transition. New powerbook and i mac reaching 1,3 ghz will be showed at Paris. It seems that Mot is dead in the high end PPC market.
I realize this is nothing but wild guessing, but what does all this mean, in your opinion, re: G5 PowerBooks? "One year transition" meaning one year for the consumer lines to move to the 970, or one year for the portables? Or is the rumoured 1.3GHz G4 PowerBook to be the last of its line, replaced in January by a nice G5 part?
WHAT makes anyone think that we're going to get a new G4 PowerBook soon (by end of September, say) and THEN a whole new, redesigned G5 PowerBook four months later?
I mean, honestly.
Has Apple EVER done this? Look, I want to be wrong on this, but those of you putting all your eggs in the "G5 PowerBook at MWSF 2004" basket may be setting yourselves up for some major letdown. Look at your history. Read the specs. Look at the timetables.
If Apple updated ANY of their stuff on 4-month cycles, I'd fall over. It's more like 8-10 months...at BEST. And in some sad cases, it's been a full year!
Four months? I really, really doubt it. Again, hope I'm wrong. But it'll be a major departure from Apple. I still say summer or so of 2004 for any G5 portables. And that's being optimistic and giving IBM more confidence than Motorola (if it was Motorola, I'd say "G5 PowerBooks by autumn 2006...MAYBE...").
Originally posted by pscates
Argghhhh!
WHAT makes anyone think that we're going to get a new G4 PowerBook soon (by end of September, say) and THEN a whole new, redesigned G5 PowerBook four months later?
I mean, honestly.
Has Apple EVER done this?
Well, the PowerMac got one last refresh as a G4 platform only a few months before ceding the stage to the G5. The Smurf G3 had a similarly short life before the G4 rolled in. In other words, it seems to me that Apple will make exceptions to their normal update schedule for major technological shifts.
The 7457 can just be dropped in, so Apple could use it as a "refresh" rather than building a whole new machine around it. In that case, I could easily see them using the '57 to goose PB sales up for a little while and hit that over 50% portable sales goal before unleashing a totally new PowerBook. All Apple would have to do, pretty much, is stop ordering '55s and start ordering '57s, and maybe improve a few other drop-in options like hard drives. Nothing serious or disruptive, though.
The problem here is that even now we know so little about the 970 and its kin that it's impossible to say what Apple's next step is.
Four months? I really, really doubt it. Again, hope I'm wrong. But it'll be a major departure from Apple. I still say summer or so of 2004 for any G5 portables. And that's being optimistic and giving IBM more confidence than Motorola (if it was Motorola, I'd say "G5 PowerBooks by autumn 2006...MAYBE...").
Four months would be pushing it, but I think summer '04 is pushing it the other way. A lot of it depends on things that we just don't know (how far along Apple's motherboard team is, whether a version of the 130nm 970 can work, when the switch to 90nm is happening), and that leaves a lot up in the air.
I will say that if everyone resigns themselves to a summer '04 update they're likely to be pleasantly surprised, and that's a good thing.
has it?
Originally posted by Programmer
And at roughly the same clock rate it appears they have roughly the same power consumption (20W @ ~1.3 GHz).
970: 30W typical n.n. maximum
7457: 19W typical 26W maximum
I wouldn't call this roughly the same. IBM & Apple didn't publish maximum power consumtion for the 970 but for the 1.3 GHz part expect it to be near 50W. This chip is not only fast and it's hungry too!
Apparently its because the G4 is an "all static design", whatever that means.
Throw in more transisors for the same functionality - get less power consumption - think of it as in static and dynamic RAM.
the only question that matters is whether whatever [technical jargon goes here] that keeps the FSB so squeezed has been addressed.
has it?
Nope, nor will it be.
I still maintain that there is no real technical reason Apple can't put the current 970 in Powerbooks right now @ 1.0 and 1.2Ghz. The heat and power consumption characteristics would be somewhat less than optimal, but not unfeasible. Power and heat will gradually decrease in magnitude as the efficiency of the production line ramps up and more of the processes switch to 90 nm. So 970 Powerbooks by next spring, guaranteed
Originally posted by smalM
970: 30W typical n.n. maximum
7457: 19W typical 26W maximum
I wouldn't call this roughly the same. IBM & Apple didn't publish maximum power consumtion for the 970 but for the 1.3 GHz part expect it to be near 50W. This chip is not only fast and it's hungry too!
Where are you getting those numbers? I remember reading that a 1.2GHz 970 only used 19 W. So I highly doubt that a 1.3 GHz part would use 50W.
Ok, found it:
http://www.arstechnica.com/cpu/02q2/.../ppc970-1.html
Second to last paragraph:
And if you consider the fact that the 970's power consumption at 1.2GHz is a mere 19W, it's almost certain that we'll see a future notebook from Apple based on the new chip.
Note in the chart right above, that a 1.8 GHz part dissipates 68.4W, so I think your numbers are just a tad off...
Originally posted by kupan787
Note in the chart right above, that a 1.8 GHz part dissipates 68.4W, so I think your numbers are just a tad off...
Check again, the article you linked in your previous post claims that a 2.8 GHZ Pentium dissipates 68.4W.
19 Watt @ 1.2GHz
42 Watt @ 1.8GHz
68 Watt @ 2.5GHz (from MacBidouille, pre-WWDC news)
Originally posted by kupan787
I remember reading that a 1.2GHz 970 only used 19 W. So I highly doubt that a 1.3 GHz part would use 50W.
1.2 GHz 970 19W @ 1.1V typical
1.3 GHz 970 30W @ 1.3V typical
TYPICAL
The thermal design of a powerbook will be guided by the maximum power consumption of the cpu not the typical.
BTW Hanibal used the 1.6V version of the 7455 in his table. The 1.3V version is listed with 15W typical and 22W maximum.
Originally posted by smalM
1.2 GHz 970 19W @ 1.1V typical
1.3 GHz 970 30W @ 1.3V typical
TYPICAL
The thermal design of a powerbook will be guided by the maximum power consumption of the cpu not the typical.
BTW Hanibal used the 1.6V version of the 7455 in his table. The 1.3V version is listed with 15W typical and 22W maximum.
19 Watt vs. 15 Watt
What a tremendous difference
Would a 1.2Ghz part have the same power consumption as a 1.8 running at 2/3rds? Or would it be higher since a lower spec part would need a higher voltage to switch than a "low power" 970? I'm sure this is muddled but you might just be able to decipher it. What I mean to say is that to get a good reliable 1.2Ghz part for a notebook, you might just need a 1.8 level part anyway (with the same costs). From what I've understood it isn't a simple matter of using lower spec parts when it comes to the heat and volatge requirements of a laptop.
That means that they have to wait on the G5 a bit, the pessimist in me says Q3-4 '04, but summer '04 seems right about on as far as a G5 PB goes, mebbe it will even feature a .13u G5 (a more expensive, and not particularly high clock one if it comes to that)
Of course .09u would be better, and Apple needs some time to work out speed/heat management on a new CPU. I don't think they'll ever make Alienware type "laptops" -- nor should they.
And, though it seems illogical, I'm pulling for Moto to get their stuff together by the time they move to .09u. IBM may be the bees knees right now, but a second option is needed, if only as a back-up.
IBM has spent a lot at fishkill, but they're own statements reveal that the money has been spent in FAB and DESIGN expertise, not neccessarily PPC fab and design, but for a facility that can do work for other chip makers. A model that becomes increasingly likely as fabrication costs escalate with each new generation. The future may have only a few "fabs" doing most of the production for a number of designers. IBM wants to be one of those. Intel and SMC will be others.
That's both good and bad for Apple. Bad, because despite the expenditure in Fishkill, IBM is NOT tied to the PPC. I've argued this before, they can switch at any time, or redirect the PPC to areas not of use to Apple, ie the POWER end of the equation. I don't think that will happen, but IBm has been disinterested in the past, and they could become so again
But it's good, because the future of chip production means that I can show up to IBM's door with a design, plunk down some change, work it out a bit, and get my chips fabbed. It might not be out of the question to see IBM making original Motorola PPC's for Apple's use in 3-4 years time. Along with their own chips, natch.