A thought hit me as I read your post. Sony will use the Cell in the PS3. IBM will produce a workstation based on the Cell. Maybe these are two different chips? IBM and Apple may use a variant of Cell for high end computers that is different from the one Sony will use in the PS3.
two different chips with different SPEs or even different PPE-variants...
IF they will use the Cell processor we won't see it before 2006. but i could imagine that IBM and apple were working on the G5-successor (which - of course - could be based on a implementation-variant of the new Cell-design-ideas). i expect IBM to be very tight-lipped on new CPUs made for apple and bet we will all be surprised when the G5s successor will be presented...
My guess is that the first Cell won't appear until 2006 and Apple may or may not use it right away, but will eventually. I don't really see the need for them to wait for a new version -- they aren't going to get past 3 GHz this year and the Cell will be at 4+ GHz and Tiger should enable it to be well utilized. It is possible Apple will wait for an OoOE PPE version, but that units power consumption will make it not suitable for iMac and portable units.
For this year we will see a successor to the 970FX and it will still be called a G5. The 970GX/MP is still a viable plan.
The next big chip for powermac will be the dual core PPC 970. This chip will give you a big speed bump and will be smaller than the cell designed for the PS3 (100 millions transistors less). Implementing this chip in a powermac, won't recquiere much code tweaking, because mac os X is already good at MP.
Now, the cell is the next chip for the ps3, it's also a brand new architecture, and if it really shine outside game box, there is no reason for not using it in a mac. But this is not a minor transition, but a big one althought smaller than the transition from 68X to the PPC.
We need more informations on this subject to be able to make good guesses. Unfortunately my christal ball is broken
Excuse me for a second.... Did you just say "the cell is not the next chip for the ps3," ?
When did this happen? I thought Cell was to be the processor for the PS3. I'm sure CEll Processors are going into the PS3.. Unless I missed a press release, or something.
Excuse me for a second.... Did you just say "the cell is not the next chip for the ps3," ?
When did this happen? I thought Cell was to be the processor for the PS3. I'm sure CEll Processors are going into the PS3.. Unless I missed a press release, or something.
Who's right? This well written article says Cell's PPE has a very short pipeline. If I have been reading correctly, other's have posted that the pipeline is long and deep, needing optimized code for best performance. I like what this new article says, but getting facts straight is more important than feeling good.
He want's money for that??? It's like a compendium of all the unfiltered Cell fanboi stuff on the net. Nothing really wrong, but lots of unrequieted hyperbole. But hey, anyone ballsy enough to list the Reg and Blanchard as references
short pipeline needing properly scheduled code for best performance. No out-of-order execution logic makes careful coding more important.
Note, for the purposes of considering Cell for Apple, that the G4 is capable of absolutely minimal OOE (roughly 1/10 as many instructions in flight as the 970 can manage) and that all AltiVec implementations to date are strictly in order.
Short pipelines, if true (and I've read that Cell's PPE is short-pipelined from more credible sources than David Every) mean that code optimized for the G4 might run pretty well on the PPE, except for the PPE's horrible penalty for branch misprediction.
However, I have a hard time reconciling "short pipeline" with 4GHz, even if they have hand-tuned the whole thing, and we have another credible source (Programmer) saying they're long...
Why on earth would anyone think that IBM could make a short pipeline design run at 5.6 GHz when their previous short pipeline designs can barely crack 1 GHz and Motorola's only recently reached a paltry 1.67 GHz? This thing is a long pipeline design, end of story. The whole chip is designed for throughput and pipelining while trying to hide long latencies at every turn, what makes you think that the Power core would be any different?
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
Are we talking like "P4-long" kind of long pipelines? What's that like- 20+ stages now?
Is it possible that the stripping away of OoE type of features has alleviated some areas of typical clockrate limits on CPU's of n number of stages? I'm just thinking out loud on this- not really suggesting the idea is based on anything concrete.
Why on earth would anyone think that IBM could make a short pipeline design run at 5.6 GHz when their previous short pipeline designs can barely crack 1 GHz and Motorola's only recently reached a paltry 1.67 GHz? This thing is a long pipeline design, end of story. The whole chip is designed for throughput and pipelining while trying to hide long latencies at every turn, what makes you think that the Power core would be any different?
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
Actually I was starting to wonder about that myself, and started to think of the possibility that if the main PPC core is running at 500MHz, and it has 8 individual cores running off that same frequency it may possibly read, and perform combined as 4GHz.
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
What was the name of the website he used to have? I remember that it was pretty good at one time.
a Playstation is meant for very good grafic (realtime 3D) and audio (Dolby 5.1 etc) output, and we have seen the "proof of concept" of playing with optical input.
Do you expect any "new borders" with cell? (Still waiting for some useful voice-input, as an example)
AAPL stock is up $2.12 in pre open as I write this. Much of that jump is a result of Merrill Lynch target update to $102 (up from $85).
Here is the quote from their upgrade:
Quote:
Merrill Lynch raised its 12-month target price on Apple Computer, Inc (AAPL: news, chart, profile) to $102 from $85 per share on the possible announcement of a partnership with Sony Corp. (SNE: news, chart, profile) . The broker told clients it thinks that Apple could introduce products based on the Cell processor made by IBM, Sony and Toshiba.
So it seems that Apple could benefit from products made by Sony, Toshiba, or even IBM using the Cell chip.
AAPL stock is up $2.12 in pre open as I write this. Much of that jump is a result of Merrill Lynch target update to $102 (up from $85).
Here is the quote from their upgrade:
------------------------------------
quote:
Merrill Lynch raised its 12-month target price on Apple Computer, Inc (AAPL: news, chart, profile) to $102 from $85 per share on the possible announcement of a partnership with Sony Corp. (SNE: news, chart, profile) . The broker told clients it thinks that Apple could introduce products based on the Cell processor made by IBM, Sony and Toshiba.
------------------------------------
So it seems that Apple could benefit from products made by Sony, Toshiba, or even IBM using the Cell chip.
I'd say if the stock jumped on "could" being the focus it was premature, but Sony today did announce a new Cell phone that will also have iTunes on board, that actually plays AAC, and they will be showing it at Cebit in March.
iTunes, and/or iTMS going into a Sony product when sony has their own music store appears to be the true focus of the jump. I think they (marketeers) believe that Sony may be shifting it's position on their music download service, and it's possible they are going to back Apple. If not why wouldn't they use their own instead?
I believe that to be the reason of the stock jump my self. The term "could" doesn't hold that much water in my book.
Comments
Originally posted by snoopy
A thought hit me as I read your post. Sony will use the Cell in the PS3. IBM will produce a workstation based on the Cell. Maybe these are two different chips? IBM and Apple may use a variant of Cell for high end computers that is different from the one Sony will use in the PS3.
two different chips with different SPEs or even different PPE-variants...
Originally posted by Krassy
IF they will use the Cell processor we won't see it before 2006. but i could imagine that IBM and apple were working on the G5-successor (which - of course - could be based on a implementation-variant of the new Cell-design-ideas). i expect IBM to be very tight-lipped on new CPUs made for apple and bet we will all be surprised when the G5s successor will be presented...
My guess is that the first Cell won't appear until 2006 and Apple may or may not use it right away, but will eventually. I don't really see the need for them to wait for a new version -- they aren't going to get past 3 GHz this year and the Cell will be at 4+ GHz and Tiger should enable it to be well utilized. It is possible Apple will wait for an OoOE PPE version, but that units power consumption will make it not suitable for iMac and portable units.
For this year we will see a successor to the 970FX and it will still be called a G5. The 970GX/MP is still a viable plan.
Now, the cell is the next chip for the ps3, it's also a brand new architecture, and if it really shine outside game box, there is no reason for not using it in a mac. But this is not a minor transition, but a big one althought smaller than the transition from 68X to the PPC.
We need more informations on this subject to be able to make good guesses. Unfortunately my christal ball is broken
When did this happen? I thought Cell was to be the processor for the PS3. I'm sure CEll Processors are going into the PS3.. Unless I missed a press release, or something.
Originally posted by onlooker
Excuse me for a second.... Did you just say "the cell is not the next chip for the ps3," ?
When did this happen? I thought Cell was to be the processor for the PS3. I'm sure CEll Processors are going into the PS3.. Unless I missed a press release, or something.
Typo error.
Edited
Sorry for the confusion
Originally posted by Powerdoc
Typo error.
Edited
Sorry for the confusion
Better late than never I always say.
Originally posted by MacRonin
Flames on the fire..
He want's money for that??? It's like a compendium of all the unfiltered Cell fanboi stuff on the net. Nothing really wrong, but lots of unrequieted hyperbole. But hey, anyone ballsy enough to list the Reg and Blanchard as references
Originally posted by Hiro
short pipeline needing properly scheduled code for best performance. No out-of-order execution logic makes careful coding more important.
Note, for the purposes of considering Cell for Apple, that the G4 is capable of absolutely minimal OOE (roughly 1/10 as many instructions in flight as the 970 can manage) and that all AltiVec implementations to date are strictly in order.
Short pipelines, if true (and I've read that Cell's PPE is short-pipelined from more credible sources than David Every) mean that code optimized for the G4 might run pretty well on the PPE, except for the PPE's horrible penalty for branch misprediction.
However, I have a hard time reconciling "short pipeline" with 4GHz, even if they have hand-tuned the whole thing, and we have another credible source (Programmer) saying they're long...
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
Is it possible that the stripping away of OoE type of features has alleviated some areas of typical clockrate limits on CPU's of n number of stages? I'm just thinking out loud on this- not really suggesting the idea is based on anything concrete.
Originally posted by Programmer
Why on earth would anyone think that IBM could make a short pipeline design run at 5.6 GHz when their previous short pipeline designs can barely crack 1 GHz and Motorola's only recently reached a paltry 1.67 GHz? This thing is a long pipeline design, end of story. The whole chip is designed for throughput and pipelining while trying to hide long latencies at every turn, what makes you think that the Power core would be any different?
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
Actually I was starting to wonder about that myself, and started to think of the possibility that if the main PPC core is running at 500MHz, and it has 8 individual cores running off that same frequency it may possibly read, and perform combined as 4GHz.
Just a thought.
Originally posted by Programmer
I used to like David Every's writings, but for the last 6 or so years he really hasn't had much to contribute and tends to have too many errors in his articles (this one in particular has quite a few). Charging for it seems ridiculous.
What was the name of the website he used to have? I remember that it was pretty good at one time.
a Playstation is meant for very good grafic (realtime 3D) and audio (Dolby 5.1 etc) output, and we have seen the "proof of concept" of playing with optical input.
Do you expect any "new borders" with cell? (Still waiting for some useful voice-input, as an example)
Originally posted by Kurt
What was the name of the website he used to have? I remember that it was pretty good at one time.
was it Mackido?
Here is the quote from their upgrade:
Merrill Lynch raised its 12-month target price on Apple Computer, Inc (AAPL: news, chart, profile) to $102 from $85 per share on the possible announcement of a partnership with Sony Corp. (SNE: news, chart, profile) . The broker told clients it thinks that Apple could introduce products based on the Cell processor made by IBM, Sony and Toshiba.
So it seems that Apple could benefit from products made by Sony, Toshiba, or even IBM using the Cell chip.
Originally posted by Aphelion
AAPL stock is up $2.12 in pre open as I write this. Much of that jump is a result of Merrill Lynch target update to $102 (up from $85).
Here is the quote from their upgrade:
------------------------------------
quote:
Merrill Lynch raised its 12-month target price on Apple Computer, Inc (AAPL: news, chart, profile) to $102 from $85 per share on the possible announcement of a partnership with Sony Corp. (SNE: news, chart, profile) . The broker told clients it thinks that Apple could introduce products based on the Cell processor made by IBM, Sony and Toshiba.
------------------------------------
So it seems that Apple could benefit from products made by Sony, Toshiba, or even IBM using the Cell chip.
I'd say if the stock jumped on "could" being the focus it was premature, but Sony today did announce a new Cell phone that will also have iTunes on board, that actually plays AAC, and they will be showing it at Cebit in March.
iTunes, and/or iTMS going into a Sony product when sony has their own music store appears to be the true focus of the jump. I think they (marketeers) believe that Sony may be shifting it's position on their music download service, and it's possible they are going to back Apple. If not why wouldn't they use their own instead?
I believe that to be the reason of the stock jump my self. The term "could" doesn't hold that much water in my book.