I'm not sure if this has been mentioned or not, but someone here asked what would be compelling enough, beyond the current feature set, to make people buy a G4 based Powerbook. I think the FIRST DUAL CPU notebook would be a big draw. What Greg said today doesn't in anyway rule that out.
If newer G4's are a lower power/heat configuration then it would stand to reason that Dual 1.42+ AiBooks would be a huge advance for the portable pro.
It's true that config wise there isn't much more other than larger memory capacity to add to the existing powerbooks, it seems to me that cramming a huge 970 CPU into laptop with lower FSB and 4200/5400RPM hard drives doesn't make much real world sense....Hopefully the architecture of a dual cpu unit would be bottleneck minimal...Not having a ton of tech experience I don't know for sure...
Don't remember where I read it but some folks I believe over in Ars forums are convinced that the G5 case is totally over developed cooling wise at the moment. These cases are made for the whole generation, not just the first salvo. So what I'm saying is, I don't think its right to say that these G5's NEEDED 9 fans. Part of all those fans undoubtedly have to do with the whole architecture working together to make a lower db rating. I mean they got a lot of bad press with the MDD G4's sounding like windtunnels. I don't think it will take a year to see the G5's move down...
Amorph, I really though that the Powermacs had the best margins. I don't see how there could be more profit in a 12"PB than a 1.6Ghz G5 and certainly not any of the previous gen G4's. LCD and battery and all sorts of miniaturization penalties to pay in a book. I think that the 15" superdrive and 17" probably do pull in a nice margin though.
I'm not going to rule out the possibility that things have changed, but the simple fact is that notebooks aren't all that expensive to make: They're basically iMacs, only without the costly swivel arm, and with slower drives. The boards have fewer expensive RAM slots and fewer ASICs, requiring fewer expensive traces to connect them all. BTO options are limited, simplifying manufacturing and logistics. The titanium shell was cut from sheets and stamped rather than poured and cast, which got around the main reason why titanium is a costly metal to work with. It wouldn't surprise me to hear that the PowerMac's case cost Apple more than the PowerBook's. Certainly, the XPC7455 in the PowerMac costs more than the MPC7455 (and certainly the L3-cache-less MPC7445) in the PowerBook.
If the PowerBook goes to a 7457 while the PowerMac goes to a 970, this will only become more true.
Quote:
Judging by edu discounts, I wouldn't be sure. None of the PM's nor the 15" or 17" PB's have a really great percentage discounted. The 12 OTOH, does, so who knows? I may be completely off, or Apple is just unwilling to make too much of an edu discount for students who they'd rather have buy consumer stuff? The 15" must be on the outs though, becuase the recent edu prices have been pretty nice.
Can't wait to see what the Canadian G5 edu prices are like.
Apple's edu discounts for individual purchase have hovered around nominal since they brought their prices down out of the stratosphere 6 years ago. Before then a student could save a good chunk of money on a Mac, but only because consumers were paying an average 40% margin on their kit. You win some, you lose some.
"IBM is a MUCH, MUCH more driven company than Mot ever seems to have been, at least in the computing area. "
That's only been true for last couple of years. Motorola WAS an incredible company. They lost interest in the chip and didn't want to invest billions once the economy tanked.
MOTO has been a company in decline for more than a few years now. They are a much, much smaller company than they were just a few years ago and are more concerned with their survival that with whether they provide Apple with better CPUs. Their fabs are not up to date either for the most part.
I'm not sure if this has been mentioned or not, but someone here asked what would be compelling enough, beyond the current feature set, to make people buy a G4 based Powerbook. I think the FIRST DUAL CPU notebook would be a big draw. What Greg said today doesn't in anyway rule that out.
The engineers' first task would be to eke out the room for a second CPU on PowerBook's crowded board. Assuming an MPC7457, cooling issues are manageable, but cost could be a problem; and, of course, you'd have bottlenecks galore. It would be an open question, I think, as to whether it would be easier to engineer a saleable dual-G4 PowerBook or a single-970 PowerBook.
I'm sure there have been dual-CPU notebook board prototypes on Apple test benches for years. It seems like a logical next step, and a worthy engineering challenge. But it might not be worth taking it from a way to hone your chops to a retail product.
Quote:
It's true that config wise there isn't much more other than larger memory capacity to add to the existing powerbooks, it seems to me that cramming a huge 970 CPU into laptop with lower FSB and 4200/5400RPM hard drives doesn't make much real world sense....Hopefully the architecture of a dual cpu unit would be bottleneck minimal...Not having a ton of tech experience I don't know for sure...
The 970's FSB is fixed relative to the CPU clock speed, so the "slow FSB" is not a problem in practice. Even if Apple used the current RAM technology, there would be an immediate doubling of real-world memory bandwidth to the CPU, just because the CPU bus would no longer be a bottleneck.
The answer for slow hard drives, on all platforms, is lots of RAM. In fact, as Steve pointed out in the WWDC keynote, this is true for PowerMacs as well. Hard drives are slow, period, relative to RAM.
1) For heat and power reasons, 1 GHz might be the fastest Apple would want to push the 7455 in a portable device.
2) Apple is not likely to issue an update of the 15" that doesn't include a speed boost. I can't think of a time when Apple issued a major form-factor change without a speed increase at the same time.
3) The 7457 would provide a way to get a speed boost, while perhaps even improving battery life at the same time, as a simple drop-in replacement for the existing 7455.
Sounds about right. Of course IBM could have slower G5 processors at say 1.0 and 1.25 Ghz that run a lot cooler than their big brothers in the new PowerMacs. Still, considering Greg Joswiak's comments and the upcoming release of 7457 G4 chips taylor made for PowerBooks I think we are stuck with G4s for the PowerBooks until next year. Here is what I think we will see announced within the next month:
PowerBook 12" 1.0 Ghz G4 (can we have a backlit keyboard please?)
PowerBook 15" 1.0 Ghz / 1.3 Ghz G4 models (this will be the best selling line)
PowerBook 17" 1.3 Ghz G4
I would love to see at least one of the PowerBooks go dual but it's doubtful. Hell, I would love to see a dual 1 Ghz iMac while we're at it. Why not?
I don't think we'll see Apple stick with the G4 in the PowerBooks for any more than a year. They'll want to keep the separation in the pro and consumer lines, so we'll see iBook G4, eMac G4 and iMac G4.
I'd say they'll revise the 15" model quite soon..perhaps at Macworld, or possibly in September. Then, we'll see a speed bump in January and a a G5 redesign next summer, with the entire PB line going to G5. I don't believe the Dual PB specualtion AT ALL. The power consumption and physical space limitations (not to mention heat) are all major factors.
Motorola appears to have brought forward volume production of the next major G4-class PowerPC processor, the MPC7457, from Q4 to Q3. The 7457 remains the leading contender for upcoming Apple PowerBooks. While some Mac users hoped IBM's 64-bit PowerPC 970, which Apple is calling the G5, would make it into PowerBooks, the chip's significant heat dissipation prevents it from taking on such a role.
If the story is correct and DUAL boards are being evaluated...I wonder if Apple will keep a low-priced G4 7457 tower in the lineup for a while, until G5 prices fall.
My guess: we're waiting for the 15 because Apple is waiting for the 7457 from Moto, and that chip will also go into the next revision of the 12 and 17.
I WAS wondering WHAT was keeping the Powerbook 15 incher.
Now I know. NO WAY is the 970 going into the 15 incher anytime soon. Looks like its a low power 1.3 G4 with the latest Ati Mobility Radeon. Should make for a nice machine.
Dual 1.3 G4? Hmmm. In the 17 incher? More room to get away with that...but would that put out more heat than a 1.2 970?
Dual Powerbook. That would certainly turn a few heads.
Keep the Wintel Centris at bar for now. Could always turn one of the cpus off in powersaving 'non mains' use...?
Lemon Bon Bon
G5 Powerbook. Not gonna happen until late 1st half 04.
Here is what I think we will see announced within the next month:
PowerBook 12" 1.0 Ghz G4 (can we have a backlit keyboard please?)
PowerBook 15" 1.0 Ghz / 1.3 Ghz G4 models (this will be the best selling line)
PowerBook 17" 1.3 Ghz G4
I agree with everyting here except for the within the next month part. Why? Because these updates hinge on Motorola getting the 7457 out the door. When that happens, the upgrades will follow, but don't expect you're 15" aluminum powerbook before that.
I'm just hoping that they get that 15" Al out the door as soon as possible. I was hoping for a G5 Powerbook too, but I guess I'll just have to deal with 15" Al. I hope they don't wait until fall to udate the thing.... that would be the timespan of a YEAR to update something and that usually doesn't happen with Apple. My guess is that they have their processor plan all ready and are just waiting for compal to finish up production... They could be waiting for some new "special" feature to help combat centrino and keep up pace with the new G5's??
Maybe IBM or Motorola has a processor specificaly made for laptops that they haven't been discussing with anyone. I think we'll see an announcement in July and they'll be shipping immediatly...
I'm not sure in what sense you mean , but at the risk of taking your wink the wrong way, I'd like to say that the 17" PowerBook was more than just a "form factor" change from the previous PowerBooks, with the most salient feature being the extra screen real estate.
What, exactly, is Q3...July, August and September, right?
If those specs/claims are accurate, there's NOTHING wrong with having a new aluminum 15.4" PowerBook based on that!
I'm guessing that's what the hold up has been, then? Waiting for this chip?
This might work out REALLY sweet, as I'll be putting the final chunk of my PowerBook fund in place in mid August, so...
Question: does that business about the 2MB L3 cache mean that's what that chip automatically has or does (I know nothing about this sort of stuff) and, by that I mean: if Apple uses this, does that mean the revved PowerBooks using this chip will automatically, no-questions-asked have this 2MB of L3 cache? Or, in typical fashion, do you suppose Apple might somehow downclock or cripple it (who knows why, they just will).
A 1.3GHz with all those specs mentioned would be GREAT.
What, exactly, is Q3...July, August and September, right?
Q3 is... now. And the next couple of months, too.
And it's sounding like Mot does in fact have the surprise up their sleeve that I've speculated they were trying to have - all this talk about a performance oriented "PowerPC" in 2004, built in Crolles on 90nm with all the goodies.
If true, this is all great news for us. The more healthy PowerPC partners Apple has, and the more great products they make, the better off we all are.
Cool, that's what I was wanting: some sort of comment or opinion by someone here who has a bit of a clue (the influx of teenaged chuckleheads spouting off "dual G5 17" PowerBooks!" stuff of late has REALLY been working on my last nerve).
This news just spurs me to keep on saving, keep on doing a little overtime here and there, snag the occasional freelance job, cook at home more, etc. for the next few months in anticipation of a very sweet reward, come late summer/early autumn.
Amorph, we know Moto has (or had, hi Centrino!) some good people working on their stuff, production and an unwillingness to invest in it always seemed to be the problem. Now we hear of strategic partnerships, and a .09u capability lurking in the confines of SMC/moto. Frankly, for even a PB, I'm more interested in that kinda product. If the .13u numbers are correct, a surprise .09u product should absolutely sip power. iBook or PB, I'm very interested in that...
I was thinking about that over that a little while ago: maybe, for whatever reason, Motorola got their stuff together a bit more and, since we have this new G5 for the pro tower crowd, maybe the other things get snappier G4s for the remainder of 2003 and early 2004...including...the...dare I say it...iBook?
From Dete at Ars about the power saving features of the ppc970
(presumably info gathered at WWDC):
Quote:
Normally, the machines are running at about 2/3 their total clock speed (for 2GHz machines, this is 1.4GHz), this jumps up to the full speed whenever it's required. The ramp time up or down is ~1ms, but the CPU is running normally during this time, so there is no performance "hiccup". This results in about 60% power/heat savings, which jumps up to about 85% savings if the machine is idle and they "turn on other power saving features".
This is a good indication that IBM allready has in place the power saving features needed for laptop use at the core of the G5. The bigger energy culprit seems to lie in those high throughput point to point buses and the areas of silicon devoted to them in the processor and companion chip. It seems reasonable that the missing pieces are an integrated memory controller and a process shrink.
Comments
If newer G4's are a lower power/heat configuration then it would stand to reason that Dual 1.42+ AiBooks would be a huge advance for the portable pro.
It's true that config wise there isn't much more other than larger memory capacity to add to the existing powerbooks, it seems to me that cramming a huge 970 CPU into laptop with lower FSB and 4200/5400RPM hard drives doesn't make much real world sense....Hopefully the architecture of a dual cpu unit would be bottleneck minimal...Not having a ton of tech experience I don't know for sure...
Any thougts?
Originally posted by Matsu
Amorph, I really though that the Powermacs had the best margins. I don't see how there could be more profit in a 12"PB than a 1.6Ghz G5 and certainly not any of the previous gen G4's. LCD and battery and all sorts of miniaturization penalties to pay in a book. I think that the 15" superdrive and 17" probably do pull in a nice margin though.
I'm not going to rule out the possibility that things have changed, but the simple fact is that notebooks aren't all that expensive to make: They're basically iMacs, only without the costly swivel arm, and with slower drives. The boards have fewer expensive RAM slots and fewer ASICs, requiring fewer expensive traces to connect them all. BTO options are limited, simplifying manufacturing and logistics. The titanium shell was cut from sheets and stamped rather than poured and cast, which got around the main reason why titanium is a costly metal to work with. It wouldn't surprise me to hear that the PowerMac's case cost Apple more than the PowerBook's. Certainly, the XPC7455 in the PowerMac costs more than the MPC7455 (and certainly the L3-cache-less MPC7445) in the PowerBook.
If the PowerBook goes to a 7457 while the PowerMac goes to a 970, this will only become more true.
Judging by edu discounts, I wouldn't be sure. None of the PM's nor the 15" or 17" PB's have a really great percentage discounted. The 12 OTOH, does, so who knows? I may be completely off, or Apple is just unwilling to make too much of an edu discount for students who they'd rather have buy consumer stuff? The 15" must be on the outs though, becuase the recent edu prices have been pretty nice.
Can't wait to see what the Canadian G5 edu prices are like.
Apple's edu discounts for individual purchase have hovered around nominal since they brought their prices down out of the stratosphere 6 years ago. Before then a student could save a good chunk of money on a Mac, but only because consumers were paying an average 40% margin on their kit. You win some, you lose some.
Originally posted by SDW2001
"IBM is a MUCH, MUCH more driven company than Mot ever seems to have been, at least in the computing area. "
That's only been true for last couple of years. Motorola WAS an incredible company. They lost interest in the chip and didn't want to invest billions once the economy tanked.
MOTO has been a company in decline for more than a few years now. They are a much, much smaller company than they were just a few years ago and are more concerned with their survival that with whether they provide Apple with better CPUs. Their fabs are not up to date either for the most part.
Originally posted by gcarswell
I'm not sure if this has been mentioned or not, but someone here asked what would be compelling enough, beyond the current feature set, to make people buy a G4 based Powerbook. I think the FIRST DUAL CPU notebook would be a big draw. What Greg said today doesn't in anyway rule that out.
The engineers' first task would be to eke out the room for a second CPU on PowerBook's crowded board. Assuming an MPC7457, cooling issues are manageable, but cost could be a problem; and, of course, you'd have bottlenecks galore. It would be an open question, I think, as to whether it would be easier to engineer a saleable dual-G4 PowerBook or a single-970 PowerBook.
I'm sure there have been dual-CPU notebook board prototypes on Apple test benches for years. It seems like a logical next step, and a worthy engineering challenge. But it might not be worth taking it from a way to hone your chops to a retail product.
It's true that config wise there isn't much more other than larger memory capacity to add to the existing powerbooks, it seems to me that cramming a huge 970 CPU into laptop with lower FSB and 4200/5400RPM hard drives doesn't make much real world sense....Hopefully the architecture of a dual cpu unit would be bottleneck minimal...Not having a ton of tech experience I don't know for sure...
The 970's FSB is fixed relative to the CPU clock speed, so the "slow FSB" is not a problem in practice. Even if Apple used the current RAM technology, there would be an immediate doubling of real-world memory bandwidth to the CPU, just because the CPU bus would no longer be a bottleneck.
The answer for slow hard drives, on all platforms, is lots of RAM.
Consider the following:
1) For heat and power reasons, 1 GHz might be the fastest Apple would want to push the 7455 in a portable device.
2) Apple is not likely to issue an update of the 15" that doesn't include a speed boost. I can't think of a time when Apple issued a major form-factor change without a speed increase at the same time.
3) The 7457 would provide a way to get a speed boost, while perhaps even improving battery life at the same time, as a simple drop-in replacement for the existing 7455.
Sounds about right. Of course IBM could have slower G5 processors at say 1.0 and 1.25 Ghz that run a lot cooler than their big brothers in the new PowerMacs. Still, considering Greg Joswiak's comments and the upcoming release of 7457 G4 chips taylor made for PowerBooks I think we are stuck with G4s for the PowerBooks until next year. Here is what I think we will see announced within the next month:
PowerBook 12" 1.0 Ghz G4 (can we have a backlit keyboard please?)
PowerBook 15" 1.0 Ghz / 1.3 Ghz G4 models (this will be the best selling line)
PowerBook 17" 1.3 Ghz G4
I would love to see at least one of the PowerBooks go dual but it's doubtful. Hell, I would love to see a dual 1 Ghz iMac while we're at it. Why not?
I'd say they'll revise the 15" model quite soon..perhaps at Macworld, or possibly in September. Then, we'll see a speed bump in January and a a G5 redesign next summer, with the entire PB line going to G5. I don't believe the Dual PB specualtion AT ALL. The power consumption and physical space limitations (not to mention heat) are all major factors.
From The Register, today. See here.
If the story is correct and DUAL boards are being evaluated...I wonder if Apple will keep a low-priced G4 7457 tower in the lineup for a while, until G5 prices fall.
My guess: we're waiting for the 15 because Apple is waiting for the 7457 from Moto, and that chip will also go into the next revision of the 12 and 17.
I WAS wondering WHAT was keeping the Powerbook 15 incher.
Now I know. NO WAY is the 970 going into the 15 incher anytime soon. Looks like its a low power 1.3 G4 with the latest Ati Mobility Radeon. Should make for a nice machine.
Dual 1.3 G4? Hmmm. In the 17 incher? More room to get away with that...but would that put out more heat than a 1.2 970?
Dual Powerbook. That would certainly turn a few heads.
Keep the Wintel Centris at bar for now. Could always turn one of the cpus off in powersaving 'non mains' use...?
Lemon Bon Bon
G5 Powerbook. Not gonna happen until late 1st half 04.
Originally posted by TWinbrook46636
Here is what I think we will see announced within the next month:
PowerBook 12" 1.0 Ghz G4 (can we have a backlit keyboard please?)
PowerBook 15" 1.0 Ghz / 1.3 Ghz G4 models (this will be the best selling line)
PowerBook 17" 1.3 Ghz G4
I agree with everyting here except for the within the next month part. Why? Because these updates hinge on Motorola getting the 7457 out the door. When that happens, the upgrades will follow, but don't expect you're 15" aluminum powerbook before that.
Maybe IBM or Motorola has a processor specificaly made for laptops that they haven't been discussing with anyone. I think we'll see an announcement in July and they'll be shipping immediatly...
Originally posted by fred_lj
eh, heard of the 17" LapZilla?
I'm not sure in what sense you mean
http://www.theregister.co.uk/content/39/31424.html
What, exactly, is Q3...July, August and September, right?
If those specs/claims are accurate, there's NOTHING wrong with having a new aluminum 15.4" PowerBook based on that!
I'm guessing that's what the hold up has been, then? Waiting for this chip?
This might work out REALLY sweet, as I'll be putting the final chunk of my PowerBook fund in place in mid August, so...
Question: does that business about the 2MB L3 cache mean that's what that chip automatically has or does (I know nothing about this sort of stuff) and, by that I mean: if Apple uses this, does that mean the revved PowerBooks using this chip will automatically, no-questions-asked have this 2MB of L3 cache? Or, in typical fashion, do you suppose Apple might somehow downclock or cripple it (who knows why, they just will).
A 1.3GHz with all those specs mentioned would be GREAT.
Originally posted by pscates
Well this came out today. Means good stuff, right? I think so!
http://www.theregister.co.uk/content/39/31424.html
What, exactly, is Q3...July, August and September, right?
Q3 is... now. And the next couple of months, too.
And it's sounding like Mot does in fact have the surprise up their sleeve that I've speculated they were trying to have - all this talk about a performance oriented "PowerPC" in 2004, built in Crolles on 90nm with all the goodies.
If true, this is all great news for us. The more healthy PowerPC partners Apple has, and the more great products they make, the better off we all are.
This news just spurs me to keep on saving, keep on doing a little overtime here and there, snag the occasional freelance job, cook at home more, etc. for the next few months in anticipation of a very sweet reward, come late summer/early autumn.
Hell I don't know...
(presumably info gathered at WWDC):
Normally, the machines are running at about 2/3 their total clock speed (for 2GHz machines, this is 1.4GHz), this jumps up to the full speed whenever it's required. The ramp time up or down is ~1ms, but the CPU is running normally during this time, so there is no performance "hiccup". This results in about 60% power/heat savings, which jumps up to about 85% savings if the machine is idle and they "turn on other power saving features".
This is a good indication that IBM allready has in place the power saving features needed for laptop use at the core of the G5. The bigger energy culprit seems to lie in those high throughput point to point buses and the areas of silicon devoted to them in the processor and companion chip. It seems reasonable that the missing pieces are an integrated memory controller and a process shrink.