or Connect
AppleInsider › Forums › Mac Hardware › Future Apple Hardware › New Powermacs to use Cell Processor?
New Posts  All Forums:Forum Nav:

New Powermacs to use Cell Processor? - Page 5

post #161 of 221
eWeek on Apple & Cell
When they said "Think Different", I ran with it.
Reply
When they said "Think Different", I ran with it.
Reply
post #162 of 221
Quote:
Originally posted by murk
eWeek on Apple & Cell

Great article.
post #163 of 221
Quote:
Originally posted by murk
eWeek on Apple & Cell

Great article indeed. The author (John Rizzo) has outlined the difficulties along the possible Cell implementation by Apple, both from the Apple side ("porting" the software) and IBM side (customizing the chip). The point is:

While the specs of the new chip are impressive, especially with its integrated support for virtualization and speedy video performance, analysts said differences between the Cell and the current PowerPC architectures will make any transition an unlikely prospect for the next few years.

It is hard to disagree with the arguments of the author. However, playing the optimist on this forum I would like to note that there is one week point there .
Let's assume that migration to the Cell requires 4-5 years. The author speculates that IBM may be trying to entice Apple to get involved. The reading of the article leaves the impression that this "convincing game" started around the time of the public introduction of the first Cell implementation this month. It is more logical to assume that the first round of it started 2-3 years ago. We don't know the outcome but it is safe to assume that Apple and IBM do have some definite short and medium term plans regarding Apple processors and some long term projections. We should be aware that the processor performance per clock cycle could be (or, to be more precise, IS) emulated long before the first silicon is ready. There is one BIG unknown at that point - the clock speed which will be achieved (well, and the challenges of the production process which contribute to the amount of the investments needed to get the processor out). So let's analyze the possibilities:

1. Apple was NOT convinced (1.5 - 2 years ago). Steve decided to wait and see and stay with the G5 for the time being. Even if this is the case, Apple would make it's best to play safe (they had enough processor-related problems, didn't they?) and to keep an eye on the development of the project and to move the operating system in a way that will make it easier to adopt the Cell if needed (just in case). Even in this scenario it is not possible to answer the main question "Will Apple use Cell in future PowerMac". However, it puts the timeframe for the possible adoption at least 3-4 years away. Based on the information we have on Cell this is the safest bet. Under this scenario Apple will use the Cell ONLY and IF it gives a definite, measurable advantage over classic design. But Steve knows more about Cell than we do... IF the things we do not know are promising there are two more scenarios.

2. Apple was convinced and is preparing for the Cell. However, it is very unlikely tat Apple will use the version of the Cell which was presented this month. So let's move the 3rd scenario.

3. Apple was interested in the Cell's potential and discussed with IBM (1.5 - 2 years ago) the optimizations they need. Both companies started to work in this direction. The Apple version could include a Power Processor Element which is much more sophisticated than the one shown this month. It could be redesigned G5. The redesign most notably affected the bus and supplemental resources - the cache, registers etc. with the main core (with the instruction scheduling architecture) almost unaffected with multithreading and some power-saving enhancements added. Note that the size and transistor count of the current G5 suggests that to match the demonstrated version of Cell (in terms of size) you can add some 4 SPEs. This is much more effective solution (cost and performance) than use the Cell as a co-processor. The arrival of this new architecture is only 2 years away (tied to Longhorn release?).


So, we don't have internal information and we have to speculate on the publicly available information.
What we know:
1. Steve likes the idea of distributing OS tasks between different subsystems, and it looks like he likes it very much: NEXT boxes used DSPs, and now we have Quartz, Quartz Extreme, Core Image, Core Audio and Core Video - all capable to offload the main processor and to use other subsystem available. One specific aspect of Core Image implementation (may be it is similar in Core Audio and Core Video as well): When editing an image in Photoshop, all filters are applied to the entire image one after another. In Core Image different filters can be "chained" and to be applied to the image virtually "simultaniosly". This could be a good candidate for Cell's streaming capability. More on Core Data later.
2. Apple did NOTHING to move Tiger to 64 bit architecture. Few low level patches, available in Panther as well, for those who need to use 64 bitness - but not available in Cocoa.
3. Apple is aggressively fighting for the living room. It's main consumer applications: iPhoto, iMovie, iTunes, iDVD, iChat AV, GarageBand, Keynote - with increasing efforts on interoperability between all the applications and the OS as well as media streaming capabilities.
4. No need to talk about the Apple pro applications, just to name them: Logic Pro, Motion, Final Cut Pro HD, DVD Studio Pro, Shake - all media related. Plus Xsan.
5. New OS features: OS-level support for Camera RAW format (Finder, Preview.app, we already have the iPhoto; supported by NSImage, CIImage. And who sad the iPhoto/Finder have to make previews of the images one by one?). Spotlight, Core Data - permanently running database engines. Core Data is based on the EOF (Enterprise Objects Framework) found in NEXT/WebObjects with the notable difference that it does not communicate with an external database but uses embedded database engine. A lot of the underlying technics of the Core Data (Faulting, Uniquing, Conflict Detection, Snapshotting) would benefit a lot from a multi-core architecture (dual-core, dual threaded G5/"conventional" G6 as well, for that matter).

May be there is room for the Cell after all!

P.S. Sorry for the long post!
post #164 of 221
Quote:
Originally posted by shadow

2. Apple did NOTHING to move Tiger to 64 bit architecture. Few low level patches, available in Panther as well, for those who need to use 64 bitness - but not available in Cocoa.

Apple has added a significant amount of 64 bit support in Tiger.

There is complete support for 64 bit binaries and 64 bit arithmetic. While it doesn't support Cocoa, so far as we are aware, it amounts to far more than a few low level patches.

The support is there where 64 bit matters: in the Unix layers. I can't think of a reason for the Finder, the system applications (Mail etc) or any user visible application to require 64 bit goodness. Write the 64 bit portion as a daemon or server and call it from a 32 bit client.
post #165 of 221
Quote:
Originally posted by JRG
Apple has added a significant amount of 64 bit support in Tiger.

There is complete support for 64 bit binaries and 64 bit arithmetic. While it doesn't support Cocoa, so far as we are aware, it amounts to far more than a few low level patches.

The support is there where 64 bit matters: in the Unix layers. I can't think of a reason for the Finder, the system applications (Mail etc) or any user visible application to require 64 bit goodness. Write the 64 bit portion as a daemon or server and call it from a 32 bit client.

Ooops!
OK, no 64 bit support in Cocoa. I am not sure that really matters - we all know that 64 bit is not going to provide the leap in performance for all aplications the 16 bit - 32 bit transition did.
post #166 of 221
That article is poor, and just plain inaccurate in some ways. Everybody saying that the Cell's Power core is not adequate for Apple is ignoring the 2:1 clock rate differential, which is just silly. This core will, on average, be about that same speed as current 970 offerings and we don't have much hope for faster 970s. As a result the Cell's bandwidth, 8 vector units, and on-chip memory controller will grossly outmatch the 970-based Macs.
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #167 of 221
Quote:
Originally posted by Programmer
I give up. Lets just wait until they publish more information.

There's no technical reason preventing Apple from using cell. IAW you're right, so take it easy

End of Line
post #168 of 221
Quote:
While the specs of the new chip are impressive, especially with its integrated support for virtualization and speedy video performance, analysts said differences between the Cell and the current PowerPC architectures will make any transition an unlikely prospect for the next few years.

That is my summary on Apple, and Cell also. It's definitely not something we'll be seeing in the next year.
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
post #169 of 221
Quote:
Originally posted by Programmer
That article is poor, and just plain inaccurate in some ways. Everybody saying that the Cell's Power core is not adequate for Apple is ignoring the 2:1 clock rate differential, which is just silly. This core will, on average, be about that same speed as current 970 offerings and we don't have much hope for faster 970s. As a result the Cell's bandwidth, 8 vector units, and on-chip memory controller will grossly outmatch the 970-based Macs.


Thanks. I was hoping you would reply to this. I disagreed with the article when I read it, and thought it was full of logical holes. Yet I didn't know enough about the topic to be confident of my conclusion.

IMO, the dumbest thing the author says is that IBM hopes to interest Apple in the Cell processor. I believe there is a good chance Apple had knowledge of Cell for quite some time, and is preparing Tiger to run on it.
post #170 of 221
Quote:
Originally posted by snoopy
I believe there is a good chance Apple had knowledge of Cell for quite some time, and is preparing Tiger to run on it.

That's hysterical.
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
post #171 of 221
Quote:
Originally posted by snoopy
IMO, the dumbest thing the author says is that IBM hopes to interest Apple in the Cell processor. I believe there is a good chance Apple had knowledge of Cell for quite some time, and is preparing Tiger to run on it.

Two things to be careful of:

1) The Cell is an STI project that Apple has no doubt been aware of, but I doubt they've had terribly accurate knowledge of the details. The AIM partnership always suffered from an inability to share information with other companies because it had to get past 3 sets of management to allow it to go out. I expect STI is much the same, perhaps even worse because they are international and government restrictions on trade & export would come into play at some point. So Apple may just be finding out about this thing now (in detail), but may have had promises from IBM about how applicable it is and the general direction of the architecture.

2) "Preparing Tiger to run on it" can be interpreted in two ways. Option 1, typical of the more literal minded, is that they are writing code that will use the Cell's vector cores and allow Tiger to boot on some mythical Cell-based Mac. I would be extremely surprised if this was the case -- see point 1 for more details. Option 2, for the planners among us, is to build things into Tiger which would allow Apple programmers to quickly take advantage of the vector units in a way that could immediately benefit applications that are taking advantage of current MacOS X technologies, plus those coming in Tiger. OpenGL, Quartz2D, CoreAudio, CoreImage, QuickTime, etc are all design to provide computationally intensive services to an application without that application knowing how the computation is actually being done. These systems can use AltiVec, multiple cores, programmable GPUs (and in the future, Cell vector units) all without explicit knowledge of the application. For all these applications that do this to leverage much of the Cell's power, all that needs to happen is for Apple to write the Cell vector code for each of these modules. This is the direction I would steer Apple if I was at the helm, and this is the direction they appear (to me) to be taking. And if they're not taking it then hopefully SJ is reading this and going "hey, this guy is pretty smart, maybe we'd better do what he says before Sony & IBM steamroller us".
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #172 of 221
Quote:
Originally posted by Programmer

. . . The Cell is an STI project that Apple has no doubt been aware of, but I doubt they've had terribly accurate knowledge of the details. The AIM partnership always suffered from an inability to share information with other companies because it had to get past 3 sets of management to allow it to go out. I expect STI is much the same, perhaps even worse because they are international and government restrictions on trade & export would come into play at some point. So Apple may just be finding out about this thing now (in detail), but may have had promises from IBM about how applicable it is and the general direction of the architecture. . .


This is certainly a logical way to look at the situation, and you may very well be right. On the other hand, STI may have actively sought involvement of Apple and other potential customers for Cell. That's why I give it good chance. Maybe not 50-50 but decent odds.

I considered what STI's motivation might be from what I read. At one extreme, STI could wish to keep it proprietary, for their own advantage. At the other, STI may wish to get Cell into everybody's products and have it become the new world standard. I'm guessing that STI's position is much closer to the second statement. That's why I'm optimistic.

I believe STI has been motivated to get Cell into products quickly. A gradual slow adoption of Cell gives the competition too much time to respond, which will make Cell's adoption even slower. However, If Cell can hit the market and reach "critical mass" quickly, it is in a much better position to duke it out with the king pin chip maker.

So, how would STI achieve such a goal? By talking to key potential customers far enough ahead of time. First, potential customers got STI's sales pitch. Those who wanted to be one of the first to put Cell in a product would have entered into an agreement with STI and have been given early details, which allows product design to proceed.

I think there is a good chance that STI took this aggressive approach.
post #173 of 221
I keep thinking back to the keynote address with the guy from Sony, and Steve saying 'someday maybe music and computers'...

And then the interview with Steve in Fortune where he states that three big name PC manufacturers were courting Apple to license OS X...

And now we have Cell, a 'revolutionary' compute unit from a three-headed consortium of, wait for it, PC manufacturers...

And when they mention DCC workstations based on Cell, who do we know that makes a pretty nice package for DCC work...?!?

Draw your own conclusions, but start out the Mac hardware lineup with a basic Cell-based Mac mini, and scale up from there...
Late 2009 Unibody MacBook (modified)
2.26GHz Core 2 Duo CPU/8GB RAM/60GB SSD/500GB HDD
SuperDrive delete
Reply
Late 2009 Unibody MacBook (modified)
2.26GHz Core 2 Duo CPU/8GB RAM/60GB SSD/500GB HDD
SuperDrive delete
Reply
post #174 of 221
Just a thought, I have been thinking about those posters that say that by the time that Apple would be able to use Cell that the 9xx series could have gotten much faster. They are both moving targets. But if IBM makes a dual core 9xx and it includes an on die memory controller and Apple keeps the line-up with two chip for the pro computers then things would get interesting. We would have 8FPUs, and 8VMX units in that confuguration.

As I understand it there are a few things that hold back the performance of the 970. Compilier, IBMs is apparently is much better, Apple is working on GCC. Auto vectorizing that is now coming to GCC. I think that memory bandwidth could rearing its ugly head when VMX is heavily used. This could get much worse if when we have two chips, four cores, 8 FPUs, and 8 VMX units. I hope tiger is heavily multi threaded. I hope Apple has something up their sleeve to alleviate memory bandwidth issues.
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
post #175 of 221
Quote:
Originally posted by snoopy
I considered what STI's motivation might be from what I read. At one extreme, STI could wish to keep it proprietary, for their own advantage. At the other, STI may wish to get Cell into everybody's products and have it become the new world standard. I'm guessing that STI's position is much closer to the second statement. That's why I'm optimistic.

Keeping it prorpietary and involving other parties in the design process are two very different things. Successful product design projects usually do not directly involve potential customers in the design process. They consider (or ask) what the customer needs, but the project itself is focused and as streamlined as possible. Failure to do that usually results in a project failure.

Quote:
I believe STI has been motivated to get Cell into products quickly. A gradual slow adoption of Cell gives the competition too much time to respond, which will make Cell's adoption even slower. However, If Cell can hit the market and reach "critical mass" quickly, it is in a much better position to duke it out with the king pin chip maker.

Absolutely, but that just means making samples available to customers -- not having them take part in the design. "Too many cooks in the kitchen..."

Besides, between Sony's PS3, Toshiba's TVs, and IBM's workstations and supercomputers I'm sure that their fabs will be busy. That is pretty much a given, otherwise Sony and Toshiba wouldn't be building additional fabs just for making Cell processors.

Quote:
So, how would STI achieve such a goal? By talking to key potential customers far enough ahead of time. First, potential customers got STI's sales pitch. Those who wanted to be one of the first to put Cell in a product would have entered into an agreement with STI and have been given early details, which allows product design to proceed.

Yes, but far enough ahead of time is about now (or maybe a few months ago). Keep in mind that Cell-based products are probably still a year away. ISSCC talks about chips well before consumers can buy products built around them. What did they say? Production starting 2H '05?

Quote:

From Brendon:
Just a thought, I have been thinking about those posters that say that by the time that Apple would be able to use Cell that the 9xx series could have gotten much faster. They are both moving targets. But if IBM makes a dual core 9xx and it includes an on die memory controller and Apple keeps the line-up with two chip for the pro computers then things would get interesting. We would have 8FPUs, and 8VMX units in that confuguration.

Last time I looked the 970 wasn't moving that much. The dual core 970 is probably, but there are zero indications that an on-chip memory controller is coming. And a dual core 970 is only 2 instruction streams, compared to the Cell's 10. Two 970s have 4 FPU execution units, 2 VMX math units and 2 VMX permute units, I don't know where you get 8/8 from. This compares to some unknown (but smaller) number of units in the PPE, and probably 2 per vector core (x8)... for something like 18-20 total. And they operate on different code sequences and are thus more flexible.

Quote:
As I understand it there are a few things that hold back the performance of the 970. Compilier, IBMs is apparently is much better, Apple is working on GCC. Auto vectorizing that is now coming to GCC. I think that memory bandwidth could rearing its ugly head when VMX is heavily used. This could get much worse if when we have two chips, four cores, 8 FPUs, and 8 VMX units. I hope tiger is heavily multi threaded. I hope Apple has something up their sleeve to alleviate memory bandwidth issues.

All the compiler work on auto-vectorization will benefit Cell much more heavily, and I'd guess that the Cell was the real motivation for it. What is really holding back the 970 is its microarchitecture -- it simply isn't scaling up, most likely due to the complexity of the OoOE unit and its automated design. I say that because the Cell's PPE doesn't have one is hand tuned, and runs at >4 GHz.

As for memory bandwidth, the Mac memory subsystem needs faster types of RAM to catch up with the 970's existing FSB. Once that is saturated then it becomes an issue of the 970's design, but that is a ways off yet.

There are more than enough threads in MacOS X (even pre-Tiger) to keep 2 Power threads busy. That's how many both Cell and a 970MP have. A dual 970MP machine might be under-utilized unless you were running multi-threaded apps, but there is no reason to run all the cores at full speed unless you have work for them -- and things like CoreImage are designed to distribute work across many cores (whether they are PPC, GPU, or Cells).
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #176 of 221
Quote:
Originally posted by Programmer
Last time I looked the 970 wasn't moving that much. The dual core 970 is probably, but there are zero indications that an on-chip memory controller is coming. And a dual core 970 is only 2 instruction streams, compared to the Cell's 10. Two 970s have 4 FPU execution units, 2 VMX math units and 2 VMX permute units, I don't know where you get 8/8 from. This compares to some unknown (but smaller) number of units in the PPE, and probably 2 per vector core (x8)... for something like 18-20 total. And they operate on different code sequences and are thus more flexible.

Clarification: If Apple continues to ship the pro computers with two chips and IBM puts two cores on each chip that would be four processors, each with two FPUs and two VMX units for a grand total of 8FPUs and 8VMX units.

Quote:
All the compiler work on auto-vectorization will benefit Cell much more heavily, and I'd guess that the Cell was the real motivation for it. What is really holding back the 970 is its microarchitecture -- it simply isn't scaling up, most likely due to the complexity of the OoOE unit and its automated design. I say that because the Cell's PPE doesn't have one is hand tuned, and runs at >4 GHz.

No argument here.

Quote:
As for memory bandwidth, the Mac memory subsystem needs faster types of RAM to catch up with the 970's existing FSB. Once that is saturated then it becomes an issue of the 970's design, but that is a ways off yet.

Agreed

Quote:
There are more than enough threads in MacOS X (even pre-Tiger) to keep 2 Power threads busy. That's how many both Cell and a 970MP have. A dual 970MP machine might be under-utilized unless you were running multi-threaded apps, but there is no reason to run all the cores at full speed unless you have work for them -- and things like CoreImage are designed to distribute work across many cores (whether they are PPC, GPU, or Cells).

As you point out, the "Cores" (image, audio, video, data) of Tiger are designed for work across many cores, and Apple will have four cores to spread the work across as well as the GPU. Not saying that Cell is not all that but just saying that the 970 ain't dead either. June will be an interesting month and we will know much more then.
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
post #177 of 221
I find it kind of odd that John Rizzo is talking about Cell processors. If this is the John Rizzo of www.macwindows.com fame I think he's a bit out of his element and that would explain some of the innacuracies in the article.

I don't really think that we need to run OSX Tiger for Cell but rather I'd love to see Apple move to Cell for their multimedia. Either as a coprocessor or if Quicktime could be sat on top of Cell for media center like devices.
He's a mod so he has a few extra vBulletin privileges. That doesn't mean he should stop posting or should start acting like Digital Jesus.
- SolipsismX
Reply
He's a mod so he has a few extra vBulletin privileges. That doesn't mean he should stop posting or should start acting like Digital Jesus.
- SolipsismX
Reply
post #178 of 221
Quote:
Originally posted by snoopy
This is certainly a logical way to look at the situation, and you may very well be right. On the other hand, STI may have actively sought involvement of Apple and other potential customers for Cell. That's why I give it good chance. Maybe not 50-50 but decent odds.

There are the obivous public contacts between the companies to support your position. Further I would imagine that there are people getting advance information years before it is made public. Now the question is would the Cellpeople want Apple involved?
Quote:

I considered what STI's motivation might be from what I read. At one extreme, STI could wish to keep it proprietary, for their own advantage. At the other, STI may wish to get Cell into everybody's products and have it become the new world standard. I'm guessing that STI's position is much closer to the second statement. That's why I'm optimistic.

There is little advantage to proprietary parts for high volumn products. The other thing that people seem to be failing to grasp is that STI has very much said publicly that Cell is going to go into alot of products. So like you I'm optimistic. I'm also optimistic that Apple is far along with a Cell based product. Maby it won't be released this half of the year but late in the year is a possiblity.
Quote:

I believe STI has been motivated to get Cell into products quickly. A gradual slow adoption of Cell gives the competition too much time to respond, which will make Cell's adoption even slower. However, If Cell can hit the market and reach "critical mass" quickly, it is in a much better position to duke it out with the king pin chip maker.

Cell doesn't have anyone to duke it out with. I suspect STI's biggest problems will be production ramp and code development.
Quote:

So, how would STI achieve such a goal? By talking to key potential customers far enough ahead of time. First, potential customers got STI's sales pitch. Those who wanted to be one of the first to put Cell in a product would have entered into an agreement with STI and have been given early details, which allows product design to proceed.

Exactly! Everyone involved with STI most likely has product far along in the development cycle. What those products are and how they use Cell is a mystery now, but I could see both Toshiba and Sony applying this technology to a bunch of stuff beyond game boxes. The question becomes is Apple in the loop, I have to say yes at this point.
Quote:

I think there is a good chance that STI took this aggressive approach.

Well I'm going to say right now that I think Apple is in the loop in some fashion. Maybe not the Cell chip we recently seen announced but using the technology certainly. In a nut shell that is what Cell is all about a technology platform more than anything. Apple could very well have this technology ready for the portable platforms or even the Mini, adapted of course for the thermal environments.

Dave
post #179 of 221
Quote:
Originally posted by hmurchison
I find it kind of odd that John Rizzo is talking about Cell processors. If this is the John Rizzo of www.macwindows.com fame I think he's a bit out of his element and that would explain some of the innacuracies in the article.

I don't really think that we need to run OSX Tiger for Cell but rather I'd love to see Apple move to Cell for their multimedia. Either as a coprocessor or if Quicktime could be sat on top of Cell for media center like devices.

I believe that about 5+ years ago there were questions about Quicktime not needing Windows to run on intel computers, there was thought put to including a Unix core to Quicktime and that way bypassing Windows and MS. I don't know what happened to that, but I remember that Apple said that Quicktime was designed for sitting on different hardware and could in time be an operating system as long as it could use Unix for the core functions. Note that this came out about the time that NetScape said that they could make in operating system that would allow the user to never know that windows was running under the browser.

So to start a thread: Is it possible and economically viable for Apple to have a standalone version od quicktime with a unix core so that Quicktime could be included or sold seperately to PS3 owners since Cell should be able to run multiple operating systems.

And is it possible that an economic reason for Apple to be interested in Cell other than speeding up Tigers Cores would be for the ease in porting games from the PS3 and Xbox since they will be written for cell. If this is possible, from what I have heard game porting is not difficult but porting and keeping the speed is very difficult requiring lots of processor specific code to be ported to a different processor, then the Mac paltform would no longer suffer this problem.
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
post #180 of 221
Quote:
Originally posted by hmurchison
I find it kind of odd that John Rizzo is talking about Cell processors. If this is the John Rizzo of www.macwindows.com fame I think he's a bit out of his element and that would explain some of the innacuracies in the article.

I don't really think that we need to run OSX Tiger for Cell but rather I'd love to see Apple move to Cell for their multimedia. Either as a coprocessor or if Quicktime could be sat on top of Cell for media center like devices.

I'm with you. If apple is even thinking of using a cell design yet I think the first product we would see a cell processor in would be a new one.
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
onlooker
Registered User

Join Date: Dec 2001
Location: parts unknown




http://www.apple.com/feedback/macpro.html
Reply
post #181 of 221
Quote:
Originally posted by Programmer
Keeping it prorpietary and involving other parties in the design process are two very different things. Successful product design projects usually do not directly involve potential customers in the design process. They consider (or ask) what the customer needs, but the project itself is focused and as streamlined as possible. Failure to do that usually results in a project failure.

Doesn't this sort of fly in the face of what Cell accomplished or what AIM accomplished with Alt-Vec. Apple was heavily involved in the development of Alt-Vec and could be involved heavily in the development of Cell.

In the general business sense though your statment about customers not being involved in the design process is just wrong. It doesn't matter if it is a ball bearing uint for a car or a wing for an airplane of a PC chip success needs customer involment.
Quote:


Absolutely, but that just means making samples available to customers -- not having them take part in the design. "Too many cooks in the kitchen..."

This simply isn't modern business practice.
Quote:

Besides, between Sony's PS3, Toshiba's TVs, and IBM's workstations and supercomputers I'm sure that their fabs will be busy. That is pretty much a given, otherwise Sony and Toshiba wouldn't be building additional fabs just for making Cell processors.

Yes all these fabs do indicate that a huge number of Cell based porcessors will soon be on the market. This clearly indicates a wide take up of the design.
Quote:


Yes, but far enough ahead of time is about now (or maybe a few months ago). Keep in mind that Cell-based products are probably still a year away. ISSCC talks about chips well before consumers can buy products built around them. What did they say? Production starting 2H '05?

The evidence is clear that Apple knew about the 970 well before it was publicly announced at ISSCC and it was a year after that when we got good hardware. Apple certainly could have been in the loop before the ISSCC debut of Cell.

While it may be a stretch I think the clearest indication yet is the lack of a low power 970 that could be used for Apples other needs. A lower pwoer variant of Cell could very well be intended for Apples low power needs where it seems to be the best offering to date. That is the PPE is apparrently the only low power 64 bit core available at the moment.
Quote:


Last time I looked the 970 wasn't moving that much. The dual core 970 is probably, but there are zero indications that an on-chip memory controller is coming. And a dual core 970 is only 2 instruction streams, compared to the Cell's 10. Two 970s have 4 FPU execution units, 2 VMX math units and 2 VMX permute units, I don't know where you get 8/8 from. This compares to some unknown (but smaller) number of units in the PPE, and probably 2 per vector core (x8)... for something like 18-20 total. And they operate on different code sequences and are thus more flexible.

Actually the fact that they operate on different code sequences makes them less flexible. 970 based multi core products would still have an advantage in some applications. Cell is at an advantage when the SPE's can be used to advantage. It is not clear at all, atleast to me, how effective the SPE's would be with scalar integer ops. That is could one feed a regular PPC thread to them for execution. The indication are that this is impossible, so flexibility is really not there.
Quote:


All the compiler work on auto-vectorization will benefit Cell much more heavily, and I'd guess that the Cell was the real motivation for it. What is really holding back the 970 is its microarchitecture -- it simply isn't scaling up, most likely due to the complexity of the OoOE unit and its automated design. I say that because the Cell's PPE doesn't have one is hand tuned, and runs at >4 GHz.

Well it remains to be seen how well the autovectorizing compiler work in the first place. However I do not see them benefiting Cell if the code has to be dramatically different on Cell.
Quote:

As for memory bandwidth, the Mac memory subsystem needs faster types of RAM to catch up with the 970's existing FSB. Once that is saturated then it becomes an issue of the 970's design, but that is a ways off yet.

An on board memory controller does move that memory closer to the processor even if it is the same old technology. So even today one could see a benefit on the 970 or a follow on. There seems to be little in the rumor mill about a 970 with on board memory controller coming, inidcating to me that 970 technology isn't long for this world. IBM has seen little uptake in the 970 and its I/O bus ouside of Apple.
Quote:

There are more than enough threads in MacOS X (even pre-Tiger) to keep 2 Power threads busy. That's how many both Cell and a 970MP have.

Yes but lets not confuse the two one is a second processor and the other is the result of SMT. The two shouldn't be compared untile we get a handle on how well IBM's SMT works on Cell. Plus we have the possibility that the cores on the MP might be threaded.
Quote:

A dual 970MP machine might be under-utilized unless you were running multi-threaded apps, but there is no reason to run all the cores at full speed unless you have work for them -- and things like CoreImage are designed to distribute work across many cores (whether they are PPC, GPU, or Cells).

Even my single processor Linux machine that I'm writing this on is under-utilized 95% of the time. But and it is a big but I can still load it down from time to time in ways that can be frustrating. Even more frustrating is that there are times when on solution (Cell) would be more efective than the other (an MP) and vs versa. For most of those times though Cell would seem to be the ideal solution for a better computing experience.

Dave
post #182 of 221
Quote:
Originally posted by Programmer


. . . Successful product design projects usually do not directly involve potential customers in the design process. They consider (or ask) what the customer needs, but the project itself is focused and as streamlined as possible. Failure to do that usually results in a project failure. . .


Sorry if I wasn't clear. I have not been advocating customer involvement in the design of Cell, but just getting information to customers, so they can begin work on concepts and product design using Cell. (You are right about "Too many cooks in the kitchen...") Customers will need samples eventually, but since a whole lot of design takes place before samples are needed, STI would want to prevent customers from getting committed to other technology. Receiving information too late could delay Cells introduction in final products by a whole generation. That's product generation, not Biblical generation.

EDIT: Oops, I didn't see the post from wizard69 when I was writing the above. I think there could have been a place for customer involvement in Cell's design, but it would have been in the early stages and no different from how STI might have consulted with many experts before proceeding too far down the design path.
post #183 of 221
Quote:
Originally posted by wizard69

. . . Cell doesn't have anyone to duke it out with. I suspect STI's biggest problems will be production ramp and code development. . .


I'm hoping Cell will eventually let the PPC compete with x86 for the personal computer market. I'm allowed to dream big, right?

Not knowing all that is involved with Cell code development, it seems like Apple's one big software job is to update OS X for Cell's SPE's and thereby accelerate basic OS services. Most applications don't need changing, and those that could benefit form SPEs can be change later, when developers get around to it.
post #184 of 221
Quote:
Originally posted by wizard69
Doesn't this sort of fly in the face of what Cell accomplished or what AIM accomplished with Alt-Vec. Apple was heavily involved in the development of Alt-Vec and could be involved heavily in the development of Cell.

An Apple employee was the project lead. But the nuts and bolts work was done by Motorola and IBM engineers hunkered down in Somerset.

Quote:
In the general business sense though your statment about customers not being involved in the design process is just wrong. It doesn't matter if it is a ball bearing uint for a car or a wing for an airplane of a PC chip success needs customer involment.

At what level? Do you have the customer leaning over an engineer at a CAD station, telling him where to place the traces? Or do you work out the spec with the customer, leave the implementation to the engineers, and then repeat the spec-implement process until the result pleases the customer? Programmer is simply arguing that you leave the customer out of the implementation phase.

Quote:
Yes all these fabs do indicate that a huge number of Cell based porcessors will soon be on the market. This clearly indicates a wide take up of the design.

Or, it clearly indicates a narrow take up into mass-produced products. In terms of manufacturing capacity, 1,000,000 units shipped per month in one product will be indistinguishable from 1,000 units shipped per month in each of 1,000 products.

Quote:
The evidence is clear that Apple knew about the 970 well before it was publicly announced at ISSCC and it was a year after that when we got good hardware.

Apple is on record as asking IBM to build (what became) the 970. Of course they knew about from the inception.

If they weren't one of the customers who approached IBM at about the same time, looking for what would become Cell, then they could have found out at any later date.
"...within intervention's distance of the embassy." - CvB

Original music:
The Mayflies - Black earth Americana. Now on iTMS!
Becca Sutlive - Iowa Fried Rock 'n Roll - now on iTMS!
Reply
"...within intervention's distance of the embassy." - CvB

Original music:
The Mayflies - Black earth Americana. Now on iTMS!
Becca Sutlive - Iowa Fried Rock 'n Roll - now on iTMS!
Reply
post #185 of 221
OK, for the sake of discussion, say Apple is going to use the Cell in their product lines. What would they call the product? Power Mac Cell? PowerBook Cell? Power Mac G6? PowerBook G6?

Lastly, what is the chances Apple will use THIS chip in an Apple product vs a FUTURE Cell? 10% 50% 90%?
You think Im an arrogant [expletive] who thinks hes above the law, and I think youre a slime bucket who gets most of his facts wrong. Steve Jobs
Reply
You think Im an arrogant [expletive] who thinks hes above the law, and I think youre a slime bucket who gets most of his facts wrong. Steve Jobs
Reply
post #186 of 221
Quote:
Originally posted by DHagan4755
OK, for the sake of discussion, say Apple is going to use the Cell in their product lines. What would they call the product? Power Mac Cell? PowerBook Cell? Power Mac G6? PowerBook G6?

Lastly, what is the chances Apple will use THIS chip in an Apple product vs a FUTURE Cell? 10% 50% 90%?

Put me down for the 50% mark because Apple could easily ues this as a coprocessor, but they may ask IBM for a more VMX like SMD since this route would allow for the easiest transition for Apple and the folks that program for the Mac. My guess is more the latter than the former.
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
post #187 of 221
Quote:
Originally posted by Brendon
Clarification: If Apple continues to ship the pro computers with two chips and IBM puts two cores on each chip that would be four processors, each with two FPUs and two VMX units for a grand total of 8FPUs and 8VMX units.

If you get 2 970MP chips, then I get 2 Cell chips... and mine can talk to eachother with something like 60 MB/sec bandwidth, plus an aggregate 50 MB/sec or so of main memory bandwidth. And 20 threads of execution, yadda yadda yadda.

Quote:
As you point out, the "Cores" (image, audio, video, data) of Tiger are designed for work across many cores, and Apple will have four cores to spread the work across as well as the GPU. Not saying that Cell is not all that but just saying that the 970 ain't dead either. June will be an interesting month and we will know much more then.

My guess is that 2005 will see the ultimate development of the 970, and starting in 2006 IBM will be flogging Cell to everyone.



Quote:
From Wizard69:
Doesn't this sort of fly in the face of what Cell accomplished or what AIM accomplished with Alt-Vec. Apple was heavily involved in the development of Alt-Vec and could be involved heavily in the development of Cell.

There is no "A" in STI.

Quote:
This simply isn't modern business practice.

Sure it is, especially with something like Cell where clearly Sony was the major motivator. The members of STI drive a radical new design forward, each with their own requirements and plans. Part of that, for IBM at least, is likely to be how they can sell it to other Power customers as part of the Power strategy/ecosystem/whatever. When it is at a point that samples can be produced and results can be demonstrated then you invite over your other potential customers and try to sell it on its merits. For somebody like Apple they could have been shown a few months before the ISSCC announcements. Before that they probably had bubbles on their internal road maps showing a potential product that would be great if it panned out as expected. Really, however, Apple couldn't do much about it before now anyhow except keep architecting their system services to support "additional computational hardware resources".

Quote:
Yes but lets not confuse the two one is a second processor and the other is the result of SMT. The two shouldn't be compared untile we get a handle on how well IBM's SMT works on Cell. Plus we have the possibility that the cores on the MP might be threaded.

Remember that 90% of compute resources are typically consumed by 10% (or less) of the code. The other 90% of the code is usually scalar crap that has to run reasonably fast... but frankly it does just fine even on a 1 GHz G4, and one thread on a 4+ GHz PPE will be plenty fast. The great (and horrible) thing about this crappy scalar code is that half (or more) of the time the processor is stalled on branches or other things, so your other SMT thread can pick up the slack.

The 10% where performance really really matters is usually on high volume computation that is (on the Mac) most often handled with the VMX vector unit. The Cell's SPEs are vector units and vector algorithms can usually be ported fairly easily between different kinds of vector units -- the hard part is figuring out the vector algorithm in the first place. Given that IBM has designed both the VMX and SPE instruction sets, its likely a good bet that they aren't hugely dissimilar... and if GCC is auto-vectorizing for both then it should cover the differences internally. And apparently the SPEs do double precision as well, so things that have to be in the FPUs on the 970 can be done on the the SPEs in the Cell.

So the result is that the Cell runs the scalar Power code about as well as a current generation PowerPC, plus at the same time it runs the really expensive vector calcs at 8-16 times as fast (8 SPE cores at double the current clock rates). Unless you are bandwidth limited, of course, but then the still Cell has you beat by quite a bit.

No, the Cell won't be faster than the 970 on every single piece of code you throw at it.... but it'll absolutely kill it on most of the code you actually care about.
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #188 of 221
Quote:
Originally posted by Amorph
An Apple employee was the project lead. But the nuts and bolts work was done by Motorola and IBM engineers hunkered down in Somerset.

Yes and I bet htat lead was involved deeply with those engineers at Somerset.

Quote:

At what level? Do you have the customer leaning over an engineer at a CAD station, telling him where to place the traces? Or do you work out the spec with the customer, leave the implementation to the engineers, and then repeat the spec-implement process until the result pleases the customer? Programmer is simply arguing that you leave the customer out of the implementation phase.

Maybe not at that level all the time, but do understand that my experiences are with manufacturing equipment. For core equipment the Engineers would be heavily involved with the vendor. You simply would not give up the heart of your machine and your income stream to any problems.

In the case of the 970 I can not see the project succedding without Apple involved. If we accept that Apple did the bridge ASIC, there would have been a huge amount of interaction right there. Further considering Apples deep interest in Vector support I wouldn't be surprised to find a few Apple engineers involved there.

Quote:


Or, it clearly indicates a narrow take up into mass-produced products. In terms of manufacturing capacity, 1,000,000 units shipped per month in one product will be indistinguishable from 1,000 units shipped per month in each of 1,000 products.

I fully expect strong take up into mass produce products. Especially if it is as easy as STI has indicated to spin variants of the processor. This clearly is one of STI's goals, we will probally be bombarded with CELL inside stickers.

Quote:


Apple is on record as asking IBM to build (what became) the 970. Of course they knew about from the inception.

Thus implying the heavy involvment of Apple.

Quote:

If they weren't one of the customers who approached IBM at about the same time, looking for what would become Cell, then they could have found out at any later date.

I would suspect that Apple has been tuned into Cell for a very long time or at least the PPE technology. One of the reasons here is the very blunt nature with which Apple has squashed rumors about a 970 based laptop. The other is the nature of Apples arraingement with IBM, which is basically one of a single supplier single user, which frankly isn't good for anybody.

Dave
post #189 of 221
Quote:
Originally posted by Programmer
My guess is that 2005 will see the ultimate development of the 970, and starting in 2006 IBM will be flogging Cell to everyone.

There is no "A" in STI.

And just how could Apple publicly get involved in Cell before the issue of compatability could be addressed? Even now Apple announcing the switch to Cell would cause a lot of confusion in the non technical world. Many would see it as similar to the trasnisiton from 68K to PPC, even though it is pretty obvious to those with a technical back ground that it isn't an issue.

Quote:


Sure it is, especially with something like Cell where clearly Sony was the major motivator. The members of STI drive a radical new design forward, each with their own requirements and plans.

Yes and that involves being involved in the design of the processor.

Quote:
Part of that, for IBM at least, is likely to be how they can sell it to other Power customers as part of the Power strategy/ecosystem/whatever. When it is at a point that samples can be produced and results can be demonstrated then you invite over your other potential customers and try to sell it on its merits.

I rather think it is more of a case that IBM had this great little processor in the PPE that drinks few amps that they have attempted to sell. Sony says well that is nice but what we really need is lots of SIMD with this type of functionality - can we work together on this.

IBM gets back with Sony explaining to them that they already have a silent partner that may like to get involved but for undisclosed reasons can't be made public. Sony likewise says that they would like to expand manfacturing capability and involve Japanese producers. So we end up with STI as the public face of Cell.

Meanwhile the PPE is almost ready for the silent partner, but then that partner realizes that some of Cells capability could be leveraged to huge advantage by them. They thus delay the *(&^*Book so that they can play in this new high performance field.

As is common in the industry the designers of Cell want to strut their stuff and do so at ISSCC. The silent partner doesn't want to be exposed so they talk all day and night about the SPE's and ignore the PPE. Meanwhile the silent partner produces a bump of a *(&^*Book that is almsot useless but silences the raving lunatics that they have as customers.

Quote:
For somebody like Apple they could have been shown a few months before the ISSCC announcements. Before that they probably had bubbles on their internal road maps showing a potential product that would be great if it panned out as expected. Really, however, Apple couldn't do much about it before now anyhow except keep architecting their system services to support "additional computational hardware resources".

So you do think Apple is in the loop as it appears that we read the same thing. Maybe we disagree on timing. I do suspect that Apple has known about the PPE for as long as the 970 has been around. Further I suspect htat Apple never had any intentions of producing a 970 based laptop.
Quote:


Remember that 90% of compute resources are typically consumed by 10% (or less) of the code. The other 90% of the code is usually scalar crap that has to run reasonably fast... but frankly it does just fine even on a 1 GHz G4, and one thread on a 4+ GHz PPE will be plenty fast. The great (and horrible) thing about this crappy scalar code is that half (or more) of the time the processor is stalled on branches or other things, so your other SMT thread can pick up the slack.

So you like the thought of Cell and its PPE as much as I do!
Quote:

The 10% where performance really really matters is usually on high volume computation that is (on the Mac) most often handled with the VMX vector unit. The Cell's SPEs are vector units and vector algorithms can usually be ported fairly easily between different kinds of vector units -- the hard part is figuring out the vector algorithm in the first place. Given that IBM has designed both the VMX and SPE instruction sets, its likely a good bet that they aren't hugely dissimilar... and if GCC is auto-vectorizing for both then it should cover the differences internally. And apparently the SPEs do double precision as well, so things that have to be in the FPUs on the 970 can be done on the the SPEs in the Cell.

Well here is what is really interestin to me. Just how similar are the SPE's to a standard PPC. There are the previously stated changes to the register file, but the question in my mind is could the AltVec code be runned unmodified on the SPE's. you certainly won't take advantage of the improvements but some code could end up executing on SPE's very easily.
Quote:

So the result is that the Cell runs the scalar Power code about as well as a current generation PowerPC, plus at the same time it runs the really expensive vector calcs at 8-16 times as fast (8 SPE cores at double the current clock rates). Unless you are bandwidth limited, of course, but then the still Cell has you beat by quite a bit.

No, the Cell won't be faster than the 970 on every single piece of code you throw at it.... but it'll absolutely kill it on most of the code you actually care about.

Well this all depends on what sort of code you care about and just how flexible the SPE's are. This may not be the hardware to run database software on for example. Or it could be very good hardware for database type work if the SPE's have that capability. I could see Apple taking a two prong approach, with processors, until the market is happy with what they see happening with Cell. Yes of some markets Cell will run away at incredible speed.
post #190 of 221
I'm sure I missed some crucial element in this discussion,
but what I've read seems to indicate that the Cell will
be PowerPC controlled, thus giving me the impression
that the best use for Cell would not be as the primary CPU
but as a secondary processor or processing buss dedicated to nearly limitless audio and video processing.

So in my view, it seems more likely that Apple might be looking at various uses for a CELL CARD to handle dedicated audio and video and more, thus replacing the need for PCI, PCI-X AND PCI-Express.

So Apple could safely release a dual core Power PC now
with provisions on the motherboard to accept CELL as a future plug-in.

Does any of this theory hold water?

post #191 of 221
Quote:
Originally posted by FallenFromTheTree
I'm sure I missed some crucial element in this discussion,
but what I've read seems to indicate that the Cell will
be PowerPC controlled, thus giving me the impression
that the best use for Cell would not be as the primary CPU
but as a secondary processor or processing buss dedicated to nearly limitless audio and video processing.

So in my view, it seems more likely that Apple might be looking at various uses for a CELL CARD to handle dedicated audio and video and more, thus replacing the need for PCI, PCI-X AND PCI-Express.

So Apple could safely release a dual core Power PC now
with provisions on the motherboard to accept CELL as a future plug-in.

Does any of this theory hold water?


That has been my thought, but I am just a speculator, how about it Wiz69 and Programmer, what do you think? to me in this scenario Apple gets it all, a great processor for the Cores of Tiger and access to much easier porting and better running games, as well as something for the Science Core.
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
Please consider throwing extra cycles at better understanding Alzheimer's, Mad Cow (CJD), ALS, and Parkinson's disease go here <a href="http://folding.stanford.edu/" target="_blank">http://folding....
Reply
post #192 of 221
Quote:
Originally posted by FallenFromTheTree
I'm sure I missed some crucial element in this discussion,
but what I've read seems to indicate that the Cell will
be PowerPC controlled, thus giving me the impression
that the best use for Cell would not be as the primary CPU
but as a secondary processor or processing buss dedicated to nearly limitless audio and video processing.

So in my view, it seems more likely that Apple might be looking at various uses for a CELL CARD to handle dedicated audio and video and more, thus replacing the need for PCI, PCI-X AND PCI-Express.

So Apple could safely release a dual core Power PC now
with provisions on the motherboard to accept CELL as a future plug-in.

Does any of this theory hold water?


No.

What you are missing is that the Power(PC) is on the Cell chip. What people mean when saying that is that the Power core that is part of the chip is controlling the other cores. It is the ring leader. This has nothing to do with putting the Cell in as a co-processor -- that would actually kill performance because any time you try communicating between chips you have to slow down and take a big latency hit (which really hurts at 4+ GHz).
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #193 of 221
Oh well it sounded cool.

Suddenly a 24 month Apple lease program is sounding very attractive.
post #194 of 221
Quote:
Originally posted by wizard69
I rather think it is more of a case that IBM had this great little processor in the PPE that drinks few amps that they have attempted to sell. Sony says well that is nice but what we really need is lots of SIMD with this type of functionality - can we work together on this.

Yes, except that IBM also had the system-on-chip expertise and ISA design expertise. So hopefully the SPEs aren't the disaster of an ISA that the PS2 vector units are (which is part of why there aren't any C/C++ compilers for the PS2's vector units).

Quote:
IBM gets back with Sony explaining to them that they already have a silent partner that may like to get involved but for undisclosed reasons can't be made public.

This I doubt. They probably didn't even tell Sony they had other customers for this new Power core -- it is an product that IBM created to license to anybody interested. It is very likely they have another big customer for the core... and its not Apple. Apple might be interested, but without the rest of the Cell attached it is much less compelling.

Remember that IBM said explicitly that the 970FX would be a laptop chip. That may have changed since 90nm didn't pan out the way everyone thought, but that doesn't change the fact that they were planning for it to scale down in terms of power.

Quote:
So you do think Apple is in the loop as it appears that we read the same thing. Maybe we disagree on timing. I do suspect that Apple has known about the PPE for as long as the 970 has been around. Further I suspect htat Apple never had any intentions of producing a 970 based laptop.

"In the loop" is very different than being "involved in design". What Apple needed to contribute to the processor could be done between an Apple and an IBM engineer in about 5 minutes of verbal conversation. IBM didn't consult Apple on the design of the 970's core -- they already had it from the POWER4. Apple said "we need VMX", so IBM tacked it on. The bus design was the one point where Apple may have had more influence because they've designed the only 970 northbridge we know of. In the case of Cell, however, that is provided by RamBus.

Quote:
So you like the thought of Cell and its PPE as much as I do!

Probably more. Have I said otherwise? My point is merely that Apple wasn't involved in the chip's development, not that they aren't a likely customer.

Quote:
Well here is what is really interestin to me. Just how similar are the SPE's to a standard PPC. There are the previously stated changes to the register file, but the question in my mind is could the AltVec code be runned unmodified on the SPE's. you certainly won't take advantage of the improvements but some code could end up executing on SPE's very easily.

http://www.realworldtech.com/page.cf...WT021005084318

This article and the one at ArsTechnica should tell you what you need to know. It seems to me that many people that read these articles miss half their content, so go back and read it again... it is packed with useful information.

Quote:
Well this all depends on what sort of code you care about and just how flexible the SPE's are. This may not be the hardware to run database software on for example. Or it could be very good hardware for database type work if the SPE's have that capability. I could see Apple taking a two prong approach, with processors, until the market is happy with what they see happening with Cell. Yes of some markets Cell will run away at incredible speed.

These things are vector processors. It is probably going to be possible to run scalar code on them (IBM said that they were working with "open source compiler writers", i.e. GCC, about providing compilers for SPE), but that doesn't mean it'll run fast.
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #195 of 221
This is a vague memory, but isn't it said somewhere in the ArsTech paper that the SPE draws its origins from a PPC601 core, except with functionality bent toward high performance SIMD operations? I remember reading this somewhere. Like a 601 with all the integer logic torn out and replaced with hardcore floating point plumbing?
Lauren Sanchez? That kinda hotness is just plain unnatural.
Reply
Lauren Sanchez? That kinda hotness is just plain unnatural.
Reply
post #196 of 221
Quote:
Originally posted by Randycat99
This is a vague memory, but isn't it said somewhere in the ArsTech paper that the SPE draws its origins from a PPC601 core, except with functionality bent toward high performance SIMD operations? I remember reading this somewhere. Like a 601 with all the integer logic torn out and replaced with hardcore floating point plumbing?

No, he was just drawing an analogy. It is a completely new piece of hardware.




That's how bad rumors get started.
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #197 of 221
Quote:
Originally posted by Programmer



. . . And apparently the SPEs do double precision as well, so things that have to be in the FPUs on the 970 can be done on the the SPEs in the Cell. . .


This is what makes the most sense to me. With 8 SPEs doing double precision, it seems wasteful to have 2 FPUs. It's my impression that the FPUs in a typical system sit idle much of the time, but when they are needed they work like crazy. If the SPEs do replace the FPUs, does the OS handle the conversion? That is, if an FPU instruction is issued, does the OS convert this to equivalent SPE instructions? My computer science ignorance is showing here.

There seems to be a confusion about this matter. The ars technica article shows two FPUs in the PPE. However, the chip photo in the Real World Technologies article shows a PPE that appears to be only twice the size of the SPE, neglecting caches. Not big enough for 2 FPUs it would seem. Also, I didn't see the FPU mentioned. However, the confusion in this article its mention that the SPE only do single precision, with other limitations.

Maybe we have to wait to get the facts?
post #198 of 221
Quote:
Originally posted by snoopy
This is what makes the most sense to me. With 8 SPEs doing double precision, it seems wasteful to have 2 FPUs. It's my impression that the FPUs in a typical system sit idle much of the time, but when they are needed they work like crazy. If the SPEs do replace the FPUs, does the OS handle the conversion? That is, if an FPU instruction is issued, does the OS convert this to equivalent SPE instructions? My computer science ignorance is showing here.

No, the SPEs are seperate cores. The cost of communicating between cores is usually quite high -- and far more than the cost of sharing data between instructions within a single core. A lot of people don't get this, but it is really quite simple... a core is a set of registers and pipelined execution units that are working on a single stream of instructions. These days the processors take 2-5 instructions per clock cycle and feed them into the pipelines. Those instructions have very high speed access to that core's pool of registers, and the kinds of instructions are mixed together. Imagine a single threaded processor executing a stream of integer (I), floating point (F), vector (V), load/store (L, S), and branch (B) instructions. It might look like this:

... LIIS LFIS LVII IFFS BFLI VSLI LBLS ...

Where each group is sent in a single clock cycle into the core to modify that core's registers (I've envisioned a 4-way dispatch here). The results of one instruction are sent to a register and can be used by a later instruction, and how long that takes is a function of the processor design and the kind of instruction -- primarily the length of the pipeline handling the instruction.

If you have multiple cores they typically communicate by writing values out to the bus, and depending on the system it may have to go all the way to main memory, directly across a shared FSB, or to a shared L2 cache. In the case of the Cell it has to go across the shared on chip bus (EIB, I think they call it?). Access to these things takes anywhere from 10-1000 cycles depending on the hardware, and then the core you are sending the data to has to read it.

You can visualize multiple cores as multiple streams of instructions (here is a 4-way and a couple of 2-way dispatch cores):

... LIIS LFIS LVII IFFS BFLI VSLI LBLS LIBS SIFF VILB SIVL LSIF SSIL ...

... BL IF VL IF BB IS LL LI FV VV SI FI FF II IF FS ...

... LL II SS LL FF IF IF IF IF IB SS SS SS SS II SL IF ...

... IL IL IL VV VV VV VV VV VV VV IS IS IS IS IS IS ...

Sending data to the right takes 1-10 cycles, and sending data up or down takes 10-1000 cycles and requires L and S instructions to do it. It should be clear that trying to redirect one kind of instruction to another core would slow the whole show down enormously. Even worse, if you are trying to do this on existing code then the processor would have to recognize an instruction it couldn't deal with and take special action... and this process is very expensive and usually involves throwing out all the instructions currently in the pipeline and executing another set that knows how to send the information to the other core.

Back in the days of 10-50 MHz processors they used to have some execution units on external chips (e.g. the MMU on the 68020 or the FPU on the 68030. That was possible because the chips were so slow relative to the speeds of the connections between the chips. These days, in the multi-GHz era, even the time to send signals between cores on the same chip is too high compared to how fast the cores are grinding through instructions.

Quote:
There seems to be a confusion about this matter. The ars technica article shows two FPUs in the PPE. However, the chip photo in the Real World Technologies article shows a PPE that appears to be only twice the size of the SPE, neglecting caches. Not big enough for 2 FPUs it would seem. Also, I didn't see the FPU mentioned. However, the confusion in this article its mention that the SPE only do single precision, with other limitations.

Maybe we have to wait to get the facts?

I'd be surprised if the PPE had two FPUs -- it is designed for low power and small size so that they can get as many SPEs on the chip as possible. Also, since it is a dual-issue chip it doesn't make sense for both of the instructions issued to be FPU instructions (that would happen quite rarely). Hopefully they just put one good FPU into the PPE.

As for the SPE confusion -- RWT says that the double precision is much slower than the single precision. This makes me wonder if they are doing some kind of emulation to achieve the higher precision. Apple has some routines for doing quad precision 128-bit floating point operations in the VMX unit... I wonder if this is along the same lines?
Providing grist for the rumour mill since 2001.
Reply
Providing grist for the rumour mill since 2001.
Reply
post #199 of 221
Quote:
Originally posted by Programmer
These days, in the multi-GHz era, even the time to send signals between cores on the same chip is too high compared to how fast the cores are grinding through instructions.

i know you're right on this. but just for the sake of discussion...

i haven't been following the specifics closely, but as i understand it, OS X's rendering technology is evolving to offload vector tasks to the GPU, right? one of the advantages of PCI-express is the bandwidth both to and *from* the GPU, allowing the OS to leverage the GPU's vector logic for non-graphics tasks.

well, the bandwidth from CPU to GPU, even over PCIexpress is going to be significantly slower than what could be arranged on a daughter card, right?

so if apple is already investing in the GPU concept, wouldn't daughter-card coprocessing be worthwhile to explore?

all this said, i completely agree that bundling everything on one die is the way to go. just seems like inefficient coprocessing is already underway.
apple gets "it."
Reply
apple gets "it."
Reply
post #200 of 221
Quote:
Originally posted by snoopy
This is what makes the most sense to me. With 8 SPEs doing double precision, it seems wasteful to have 2 FPUs. It's my impression that the FPUs in a typical system sit idle much of the time, but when they are needed they work like crazy. If the SPEs do replace the FPUs, does the OS handle the conversion? That is, if an FPU instruction is issued, does the OS convert this to equivalent SPE instructions? My computer science ignorance is showing here.

There seems to be a confusion about this matter. The ars technica article shows two FPUs in the PPE. However, the chip photo in the Real World Technologies article shows a PPE that appears to be only twice the size of the SPE, neglecting caches. Not big enough for 2 FPUs it would seem. Also, I didn't see the FPU mentioned. However, the confusion in this article its mention that the SPE only do single precision, with other limitations.

Maybe we have to wait to get the facts?

The SPE's have the capability to do double precision the old fashioned way, multiple single precision ops. DP work on an SPE takes about a 10x speed hit for that conversion. Sure you can do it, but it ain't blindingly fast, just another current generation high end processor. The design was a specific tradeoff by Sony to do the things common in media really fast, while sacrificing the high-end scientific processing niche uber-capability. The right call for their target market and control of design costs.
.
Reply
.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Future Apple Hardware
AppleInsider › Forums › Mac Hardware › Future Apple Hardware › New Powermacs to use Cell Processor?