Besides what if there is a problem? What if Apple had a roadmap that had a G5 available in late 2002? That was the plan, right? Then motorola messed it up. You can't have a chipset roadmap if you are not making the chips.
I think Steve learned the lesson of Osborn Computers. Never discuss the "next big model" until it's in the stores.
While I agree on the principle I would say that Osborn Computer was in a diffferent situation than Apple is now. They practically made the next 1000 computers from the money they earned from the previous 1000. Had no reserve at all. Different with Apple (unless of course they also bought vivendi at the time they preannounced the 970)
Whenever Fred Anderson or Jobs is questioned about poor Pro line sales and the performance issue, they almost invariably respond in the following manner:
[list=1][*]We're aware that sales are not at levels we would like. We attribute this to soft sales in the graphics market due to a lack of software.
[*]We are innovating our way to success.
[*]Megahertz isn't all that important.
[*]We're working on it. (Sometimes with hints that a change will come this year.)[/list=1]
Essentially, this all adds up to a kind of corporate softshoe around the issue that Pro sales are in the toilet because of performance issues and they can't do anything about it right now. To admit this would be bad publicity, so they will continue with this line of non-answer until they have an answer.
I don't think that we will hear anything else from Apple until such time as they can do something about it, and that's when Apple will be able to "discuss their relationship with Motorola". But not until then.
Damn this economy... If things were only a little better I'd GLADLY "settle" for Dual 1.25 PowerMac... and, I'm a casual user in that I normally use Appleworks, iMovie (occasionally), and iTunes... At least with the PPC I could play a few of the better modern games (specifically Freedom Force).
I've considered building a cheap Win-junk to play games with, but I really don't want to stray back into owning another M$ box. I switched four years ago and I don't want to go back just to waste time that I could be using to write compelling fiction (always the self promoter).
I want a 970 when they come out, but the speedbump I'd receive uptrading my G3 350 Bondi iMac to a dual 1.25 with a good (expandable) graphics card would give me an easy 4 years of computing happiness.
I've considered buying a second hand one, or one from e-bay but in the end I want my money to go to Apple (unless the price is so reasonable it's almost unreasonable).
Well, first as a person who is responsible for buying PCs for a business, I have to say, I don't look at roadmaps. I look at what I can currently get for my business' money. 90% of business' don't care about roadmaps and what's coming. They look at what needs upgrading in their company and what they can upgrade it with. "Hey, I've got this PIII 866 and it's running as slow as a dog... let's see, I can get a PIV 3 Ghz PC for X dollars to replace the 866."
As a computer aware, gaming and Photoshop using Mac home user, I'd like to wait until something good for my money comes along.
As far as Apple publishing a roadmap, why would they. They aren't making the chip. IBM and Moto make the chip and publish roadmaps of the chip... although how accurate those roadmaps are, I don't know... so what use are the roadmaps. And don't say Apple should make roadmaps for their computers, that's idiotic. Noone makes roadmaps for their computers. Computer features can change in a matter of months. New technologies and directions pop up faster than you can say "Apple should have...".
If you announce that the 2004 Corvette will have 5 more horsepower than the 2003, you won't lose too many sales of current 'Vettes.
On the other hand, if you announce that the 2004 will have 400 more horsepower, a new chassis and better gas mileage, you'd kill current sales.
Intel and AMD are like the first example... no massive leaps to give customers pause.
Apple, with the 970, finds itself in the second scenario. When Apple announces the 970...it's going to market it like its the second coming. And if it doesn't have the product...well nobody is boing to be buying g4s.
USUALLY, you want a roadmap to convince business customers of your longterm viability. But SOMETIMES, like with the 970, a roadmap would kill you. It's an ugly situation...but Apple has to hide or pretend the 970 doesn't exist until they're actually rolling out the door.
I don't know if Apple ever published a "roadmap", but they used to be more public with their product plans 3-6 months out.
The old IT Manager tabloid MacWeek would tell you exactly which CPUs were coming, how many slots, the expected price range, and even a little drawing of what the case looked like. All of this info was unofficial, as "Apple Computer declined to comment" -- even though the specs were far too detailed to not come right from Apple PR.
However, I don't think this information would make it out to the general public. The consumer rags like MacWorld would wait until the Macs were released before crowing.
This was going on as late as the G3 intro, as the comment about the 9700 shows. (Motorola and Power were also spreading info about their planned G3s.)
In my view, it would help Apple more than hurt them if they gave unoffiical plans for Pro products. Joe iMac doesn't care, but pros have real world budgets and purchasing constraints. How many sales have they lost because some people have been waiting 2 years for the G5? How are you supposed to plan an OS X rollout if you don't know if you are going to replace your hardware in 3 months? The upcoming WWDC has all but announced the 970, but Apple is still tightlipped. Why?
IIRC, Apple was pretty public on where it was going around OS 8.5 or 8.6, saying that it's eventual goal was to line the Mac OS up as much as it could with the future, Mac OS X. I can remember vaguely during the introduction of OS 9, Steve Jobs talking about features OS 9 had that would make the transition to OS X easier from a user standpoint. Little did we know then what Apple had in store for us, with a new GUI, UNIX, layered architecture, etc.
The question was on hardware speed concerns. Yet, Steve inched into his statement on the matter that "there will be a time" when Apple will discuss its relationship with Motorola. I found that interesting and telling. No one really asked about Apple's relationship with Motorola, but it was mentioned nonetheless.
The wording and timing of Steve's comments suggested that this relationship status report may come soon.
You talk about this new 970 as a quantum leap in performance, so how much should we expect the 970 compared with the present fastest g3 and g4 for example a photoshop command. or compare it to dialup at 28kps vs cable at 1.5mps these are things i can relate to, how much faster in real terms over a pent 4 2.5 or higher. what are the expected speeds of the 970 and how would multiple processors kick intels butt.
Give us some relative benchmarks to make us drool, or is the 970 just another speed bump give me a reason to drool and wait. how would it compare in the notebook line.
You talk about this new 970 as a quantum leap in performance, so how much should we expect the 970 compared with the present fastest g3 and g4 for example a photoshop command. or compare it to dialup at 28kps vs cable at 1.5mps these are things i can relate to, how much faster in real terms over a pent 4 2.5 or higher. what are the expected speeds of the 970 and how would multiple processors kick intels butt.
Give us some relative benchmarks to make us drool, or is the 970 just another speed bump give me a reason to drool and wait. how would it compare in the notebook line.
Well, over on Ars they're discussing some of this now. The 970 could show a 5x-6x increase in floating point performance over the G4. Integer and Altivec should be faster, but probably not that much - maybe 3x.
These are very significant numbers - on the order of what 68k->PPC promised. Keep in mind, however, that the 970 will bring it into real competition with the best from Intel and AMD, but it won't be substantially faster, and in some cases it'll still be a bit slower.
Isn't the 970 twice as fast as the G4 at the same clock cycle?
I heard something like that.
So a 1GHz 970 = a 2Ghz G4. But since the 970 is going to be above 1GHz (anywhere from 1.4-2.3GHz), the new 970s will be a lot better than the G4.
I wish I had those spec scores....I think the 970 was on par with the AMD Opteron (sp?), had faster fp than a 3GHz P4, but slower integer than the P4. The G4 was around a magnitude of 3-4x slower than any of these.
While I agree on the principle I would say that Osborn Computer was in a diffferent situation than Apple is now. They practically made the next 1000 computers from the money they earned from the previous 1000. Had no reserve at all. Different with Apple (unless of course they also bought vivendi at the time they preannounced the 970)
You will be amazed at how fast you can burn thru 4.5 billion - without buying any companies.
Regarding the 970 and pentium comparisment: That benchmark doesn´t use altivec does it? What abou the pentium? Does that have anything that isn´t mesaured in the test?
Those 'calculated' spec scores (thanks for the home work Dr. Boar...) highlight the real neglect of the G4.
Even at 3 gig, it aint competing on Spec. However, I'm sure in real world performance it would hold its own and outside of FP, the performance difference would be negligable.
However, that's 'maybe'.
What is clear is that the 970 will hammer the G4 in FP performance. Four times the FP at a given clock? So at 1.8 gig? Ouch time. Dual that and you have a multimedia powerhouse!
I'm not sure that the Spec scores give the real picture for the 970's real world performance. Cue Altivec to cut loose and I'm sure we're looking at a remarkable next gen' machine that will make make 'snap' yesterday's topic in both Web page rendering and 'X' finder performance outside of any code improvements.
Now, Lightwave is heavily optimised by Newtek for the Pentium 4. (Athlon's best doesn't come close...) So, given the raw FP performance of the 970, altivec and bandwidth enhancements and ANYTHING Newtek do...plus multiprocessing...we're probably looking at least a four fold improvement in rendering times. Shaving hours and hours off LW rendering times. I can't wait.
The implications for Photoshop performance? The 970 will be THE PS machine.
Given a 10% margin of error on IBM's conservative Spec estimate...you'd hardly notice the difference in Hammer/970 scores. But I sense a 'sucker punch' from IBM here. Factor in Altivec on THAT bus and whew, we could have a big can of whoopass on that Prescott chip. We'll have to see.
As for the thread title. Jobs hint at Motorola relationship. Sounds ominous. This has been a long, and of late, arduous, relationship. I'm note sure this relationship will recover from the G4 debacle. Cancelling the G5 at the last minute isn't the kind of thing I'd imagine Steve Jobs would forgive you for and sought to serve Apple with a warning. You can't just depend on one supplier (especially if they're incompetant...) If Apple wan't market share, I think they've got no choice but to go multi-platform. Maybe not this year or next but relatively soon. They're still losing market share. A sever or two on Intel hardware/Hammer line of PCs etc. As long as the ever ingenious Apple can maintain control of motherboard and find a way to make software performance acceptable via emulation (and the whispers say they can!) then it doesn't matter. A sale is a sale is a sale.
Motorola, an embedded cpu company will find themselves run down by the IBM juggernaut. Moto may become a 3rd tier phone cpu supplier with some games services. They may do well in that space. Either way, I think they're increasingly irrelevant to Apple.
Apple updated almost all their line earlier this year.
That sent a clear signal, re-hashing the same G4s...that Apple will do the same later this year or early next...and Apple will heave-ho Moto's ass out of their computers. The way they've drilled Apple's ass on this 'exclusive' contract...the way they've hurt Apple's 'power'Mac sales and marketshare...I don't see anyway back. A lawsuit? Most probable.
I see a significant transition within a year. Consumer line G3++. 2 gig. Rio. 200 bus minimum. From ibooks, to emacs, to low end iMacs.
I see 970s: Towers, Powerbooks and highend iMacs.
It's plain to me that Apple have took it up the ass and gathered evidence on Moto's behaviour. If I was Steve, I'd be after at least half a billion of cash to add to the Apple stock pile. That's alot of 'X' ad' revenue...
One years R&D...
I could be wrong. But this is the way I'm thinking at the mo'...
Comments
Members of cults usually know a bit of what is going on.
Originally posted by Jared
And have sales of the PowerMac grind to a stand still until whatever it is they are working on comes out?
I would rather Apple not say a word and their sales inch up slowly then fall down at light speeds.
I think Steve learned the lesson of Osborn Computers. Never discuss the "next big model" until it's in the stores.
Originally posted by wwwork
Besides what if there is a problem? What if Apple had a roadmap that had a G5 available in late 2002? That was the plan, right? Then motorola messed it up. You can't have a chipset roadmap if you are not making the chips.
That would've been hell.
Originally posted by BigMcLargehuge
I think Steve learned the lesson of Osborn Computers. Never discuss the "next big model" until it's in the stores.
While I agree on the principle I would say that Osborn Computer was in a diffferent situation than Apple is now. They practically made the next 1000 computers from the money they earned from the previous 1000. Had no reserve at all. Different with Apple (unless of course they also bought vivendi at the time they preannounced the 970)
Whenever Fred Anderson or Jobs is questioned about poor Pro line sales and the performance issue, they almost invariably respond in the following manner:
[list=1][*]We're aware that sales are not at levels we would like. We attribute this to soft sales in the graphics market due to a lack of software.
[*]We are innovating our way to success.
[*]Megahertz isn't all that important.
[*]We're working on it. (Sometimes with hints that a change will come this year.)[/list=1]
Essentially, this all adds up to a kind of corporate softshoe around the issue that Pro sales are in the toilet because of performance issues and they can't do anything about it right now. To admit this would be bad publicity, so they will continue with this line of non-answer until they have an answer.
I don't think that we will hear anything else from Apple until such time as they can do something about it, and that's when Apple will be able to "discuss their relationship with Motorola". But not until then.
I've considered building a cheap Win-junk to play games with, but I really don't want to stray back into owning another M$ box. I switched four years ago and I don't want to go back just to waste time that I could be using to write compelling fiction (always the self promoter).
I want a 970 when they come out, but the speedbump I'd receive uptrading my G3 350 Bondi iMac to a dual 1.25 with a good (expandable) graphics card would give me an easy 4 years of computing happiness.
I've considered buying a second hand one, or one from e-bay but in the end I want my money to go to Apple (unless the price is so reasonable it's almost unreasonable).
As a computer aware, gaming and Photoshop using Mac home user, I'd like to wait until something good for my money comes along.
As far as Apple publishing a roadmap, why would they. They aren't making the chip. IBM and Moto make the chip and publish roadmaps of the chip... although how accurate those roadmaps are, I don't know... so what use are the roadmaps. And don't say Apple should make roadmaps for their computers, that's idiotic. Noone makes roadmaps for their computers. Computer features can change in a matter of months. New technologies and directions pop up faster than you can say "Apple should have...".
On the other hand, if you announce that the 2004 will have 400 more horsepower, a new chassis and better gas mileage, you'd kill current sales.
Intel and AMD are like the first example... no massive leaps to give customers pause.
Apple, with the 970, finds itself in the second scenario. When Apple announces the 970...it's going to market it like its the second coming. And if it doesn't have the product...well nobody is boing to be buying g4s.
USUALLY, you want a roadmap to convince business customers of your longterm viability. But SOMETIMES, like with the 970, a roadmap would kill you. It's an ugly situation...but Apple has to hide or pretend the 970 doesn't exist until they're actually rolling out the door.
The old IT Manager tabloid MacWeek would tell you exactly which CPUs were coming, how many slots, the expected price range, and even a little drawing of what the case looked like. All of this info was unofficial, as "Apple Computer declined to comment" -- even though the specs were far too detailed to not come right from Apple PR.
However, I don't think this information would make it out to the general public. The consumer rags like MacWorld would wait until the Macs were released before crowing.
This was going on as late as the G3 intro, as the comment about the 9700 shows. (Motorola and Power were also spreading info about their planned G3s.)
In my view, it would help Apple more than hurt them if they gave unoffiical plans for Pro products. Joe iMac doesn't care, but pros have real world budgets and purchasing constraints. How many sales have they lost because some people have been waiting 2 years for the G5? How are you supposed to plan an OS X rollout if you don't know if you are going to replace your hardware in 3 months? The upcoming WWDC has all but announced the 970, but Apple is still tightlipped. Why?
The question was on hardware speed concerns. Yet, Steve inched into his statement on the matter that "there will be a time" when Apple will discuss its relationship with Motorola. I found that interesting and telling. No one really asked about Apple's relationship with Motorola, but it was mentioned nonetheless.
The wording and timing of Steve's comments suggested that this relationship status report may come soon.
Give us some relative benchmarks to make us drool, or is the 970 just another speed bump give me a reason to drool and wait. how would it compare in the notebook line.
Originally posted by NOFEER
You talk about this new 970 as a quantum leap in performance, so how much should we expect the 970 compared with the present fastest g3 and g4 for example a photoshop command. or compare it to dialup at 28kps vs cable at 1.5mps these are things i can relate to, how much faster in real terms over a pent 4 2.5 or higher. what are the expected speeds of the 970 and how would multiple processors kick intels butt.
Give us some relative benchmarks to make us drool, or is the 970 just another speed bump give me a reason to drool and wait. how would it compare in the notebook line.
Well, over on Ars they're discussing some of this now. The 970 could show a 5x-6x increase in floating point performance over the G4. Integer and Altivec should be faster, but probably not that much - maybe 3x.
These are very significant numbers - on the order of what 68k->PPC promised. Keep in mind, however, that the 970 will bring it into real competition with the best from Intel and AMD, but it won't be substantially faster, and in some cases it'll still be a bit slower.
Originally posted by DHagan4755
Isn't the 970 twice as fast as the G4 at the same clock cycle?
I heard something like that.
So a 1GHz 970 = a 2Ghz G4. But since the 970 is going to be above 1GHz (anywhere from 1.4-2.3GHz), the new 970s will be a lot better than the G4.
I wish I had those spec scores....I think the 970 was on par with the AMD Opteron (sp?), had faster fp than a 3GHz P4, but slower integer than the P4. The G4 was around a magnitude of 3-4x slower than any of these.
If you focus on the chip only..................
Originally posted by Anders the White
While I agree on the principle I would say that Osborn Computer was in a diffferent situation than Apple is now. They practically made the next 1000 computers from the money they earned from the previous 1000. Had no reserve at all. Different with Apple (unless of course they also bought vivendi at the time they preannounced the 970)
You will be amazed at how fast you can burn thru 4.5 billion - without buying any companies.
IBM PPC 970:
SPECint2000 937 @ 1.8 GHz
SPECfp2000 1051 @ 1.8 GHz
Pentium IV @ 3.06 GHz:
SPECint2000 1032 @ 3.06 GHz
SPECfp2000 1092 @ 3.06 GHz
I do not have the Opteron scores but as I recall they were in the 1000 -1100 range @1.8 GHz
Motorola G4 (for the fun of it I calculated the scores a 3 Ghz G4 would have. It would still trail all of them... )
SPECint2000 418 @ 1.4 GHz 913@3.06
SPECfp2000 248 @ 1.4 GHz 542@3.06
Originally posted by DrBoar
for the fun of it...
Thats not funny
Regarding the 970 and pentium comparisment: That benchmark doesn´t use altivec does it? What abou the pentium? Does that have anything that isn´t mesaured in the test?
Even at 3 gig, it aint competing on Spec. However, I'm sure in real world performance it would hold its own and outside of FP, the performance difference would be negligable.
However, that's 'maybe'.
What is clear is that the 970 will hammer the G4 in FP performance. Four times the FP at a given clock? So at 1.8 gig? Ouch time. Dual that and you have a multimedia powerhouse!
I'm not sure that the Spec scores give the real picture for the 970's real world performance. Cue Altivec to cut loose and I'm sure we're looking at a remarkable next gen' machine that will make make 'snap' yesterday's topic in both Web page rendering and 'X' finder performance outside of any code improvements.
Now, Lightwave is heavily optimised by Newtek for the Pentium 4. (Athlon's best doesn't come close...) So, given the raw FP performance of the 970, altivec and bandwidth enhancements and ANYTHING Newtek do...plus multiprocessing...we're probably looking at least a four fold improvement in rendering times. Shaving hours and hours off LW rendering times. I can't wait.
The implications for Photoshop performance? The 970 will be THE PS machine.
Given a 10% margin of error on IBM's conservative Spec estimate...you'd hardly notice the difference in Hammer/970 scores. But I sense a 'sucker punch' from IBM here. Factor in Altivec on THAT bus and whew, we could have a big can of whoopass on that Prescott chip. We'll have to see.
As for the thread title. Jobs hint at Motorola relationship. Sounds ominous. This has been a long, and of late, arduous, relationship. I'm note sure this relationship will recover from the G4 debacle. Cancelling the G5 at the last minute isn't the kind of thing I'd imagine Steve Jobs would forgive you for and sought to serve Apple with a warning. You can't just depend on one supplier (especially if they're incompetant...) If Apple wan't market share, I think they've got no choice but to go multi-platform. Maybe not this year or next but relatively soon. They're still losing market share. A sever or two on Intel hardware/Hammer line of PCs etc. As long as the ever ingenious Apple can maintain control of motherboard and find a way to make software performance acceptable via emulation (and the whispers say they can!) then it doesn't matter. A sale is a sale is a sale.
Motorola, an embedded cpu company will find themselves run down by the IBM juggernaut. Moto may become a 3rd tier phone cpu supplier with some games services. They may do well in that space. Either way, I think they're increasingly irrelevant to Apple.
Apple updated almost all their line earlier this year.
That sent a clear signal, re-hashing the same G4s...that Apple will do the same later this year or early next...and Apple will heave-ho Moto's ass out of their computers. The way they've drilled Apple's ass on this 'exclusive' contract...the way they've hurt Apple's 'power'Mac sales and marketshare...I don't see anyway back. A lawsuit? Most probable.
I see a significant transition within a year. Consumer line G3++. 2 gig. Rio. 200 bus minimum. From ibooks, to emacs, to low end iMacs.
I see 970s: Towers, Powerbooks and highend iMacs.
It's plain to me that Apple have took it up the ass and gathered evidence on Moto's behaviour. If I was Steve, I'd be after at least half a billion of cash to add to the Apple stock pile. That's alot of 'X' ad' revenue...
One years R&D...
I could be wrong. But this is the way I'm thinking at the mo'...
Lemon Bon Bon