Realistic Expectations

[gosh]

Apologies if all of the following has been posted/discussed before.

Regarding all of the PS3 and Xbox2 hubub I have observed a number of people run around with a number of predictions/assumptions, without ever stating how they came to these conclusions.

1) Xbox2s tri-core CPU will run at 3.5Ghz+ :
Currently dual core Power5s operate at about 2Ghz on a 130nm process, I believe it is a given the first Xbox2s will ship on the 90nm process. Is it realistic to add a third core and shrink to 90nm and it will still be able to run at 3.5Ghz ? Admittedly the Power5 has 1.9MB of L2 cache but if even AMD which has a single core Athlon64 which is on 90nm is yet to release chips past 2.6Ghz without a new core stepping and most likely will not release 3Ghz+ part till the end of the year/ next year can we really expect a tri-core Power5 derivative to run at 3.5Ghz+ ?

2)Cell architecture chips may/will eventually end up in personal computers:
Why on earth would Sony even bother to undertake the aforementioned project ? I really do not see PC users jumping all over such a product. I mean zero compatibility with market dominating Microsoft products and a completely new OS, not to mention very little in the way of speed gains for web browsing or document editing anyway. The only way to push it on the desktop would be to use Linux and even then there is no incentive to stop buying a conventional PC.

3) PS3s final implementation details:
Now I am no expert on chips but a cursory glance at www.sandpile.org reveals that Athlon64s and P4s are just less than a third the size of 1 PE (?) so I expect the cell to be a big chip is it still within the realms of reality that Sony will launch such a chip with a fully fledged BR drive and NV50 ? People are also bandying about 24 pipe configurations for GPUs in both the Microsoft and Sony consoles, is a 24 pipe card really necessary at 720p. I did a few calculations and peak pixel out put is quite high for a 24 pipe card at 500 - 600 MHz.

Is it possible that Sony will simply remove vertex shaders from Nvidias "next generation GPU" and use the PE to do the vertex shading or remove some of the vertex shaders. Is it worthwhile, how much will be saved or is it just plain stupid from a developer point of view. Could any of the resident developers give a ballpark figure on the percentage performance drop from the lack of out of order execution on the PU.

Thanks.

 

[psurge]

1) Xbox2s tri-core CPU will run at 3.5Ghz+ :
can we really expect a tri-core Power5 derivative to run at 3.5Ghz+ ?

Whether or not the CPU is a Power5 derivative is unknown. If, for example, the Xbox2 CPU consists of multiple copies of the PPE found in the disclosed Cell chip (Power ISA, 2 way SMT, in-order execution, support for VMX, 512K L2), then (eye-balling it ) you could fit 3 or 4 of them on a die about the size of the Cell chip. These PPEs are running at 4+ GHz already, so the rumored number of cores and clock-speed looks to be feasible (power consumption might limit the number/frequency of the on-die cores).

3) PS3s final implementation details:
Now I am no expert on chips but a cursory glance at www.sandpile.org reveals that Athlon64s and P4s are just less than a third the size of 1 PE (?) so I expect the cell to be a big chip is it still within the realms of reality that Sony will launch such a chip with a fully fledged BR drive and NV50 ? People are also bandying about 24 pipe configurations for GPUs in both the Microsoft and Sony consoles, is a 24 pipe card really necessary at 720p. I did a few calculations and peak pixel out put is quite high for a 24 pipe card at 500 - 600 MHz.

Well, the cell is > 200mm2 at 90nm. Considering that the EE was larger (at PS2 launch time), it seems possible. Also the concept of "pipe" in a GPU is kind of vague. For example, you might have an architecture capable of outputting 8, 12, or 16 pixels per clock cycle, but which can work on 24 or more pixels at a time.

 

[a688]

Apologies if all of the following has been posted/discussed before.

Regarding all of the PS3 and Xbox2 hubub I have observed a number of people run around with a number of predictions/assumptions, without ever stating how they came to these conclusions.

1) Xbox2s tri-core CPU will run at 3.5Ghz+ :
Currently dual core Power5s operate at about 2Ghz on a 130nm process, I believe it is a given the first Xbox2s will ship on the 90nm process. Is it realistic to add a third core and shrink to 90nm and it will still be able to run at 3.5Ghz ? Admittedly the Power5 has 1.9MB of L2 cache but if even AMD which has a single core Athlon64 which is on 90nm is yet to release chips past 2.6Ghz without a new core stepping and most likely will not release 3Ghz+ part till the end of the year/ next year can we really expect a tri-core Power5 derivative to run at 3.5Ghz+ ?

2)Cell architecture chips may/will eventually end up in personal computers:
Why on earth would Sony even bother to undertake the aforementioned project ? I really do not see PC users jumping all over such a product. I mean zero compatibility with market dominating Microsoft products and a completely new OS, not to mention very little in the way of speed gains for web browsing or document editing anyway. The only way to push it on the desktop would be to use Linux and even then there is no incentive to stop buying a conventional PC.

3) PS3s final implementation details:
Now I am no expert on chips but a cursory glance at www.sandpile.org reveals that Athlon64s and P4s are just less than a third the size of 1 PE (?) so I expect the cell to be a big chip is it still within the realms of reality that Sony will launch such a chip with a fully fledged BR drive and NV50 ? People are also bandying about 24 pipe configurations for GPUs in both the Microsoft and Sony consoles, is a 24 pipe card really necessary at 720p. I did a few calculations and peak pixel out put is quite high for a 24 pipe card at 500 - 600 MHz.

Is it possible that Sony will simply remove vertex shaders from Nvidias "next generation GPU" and use the PE to do the vertex shading or remove some of the vertex shaders. Is it worthwhile, how much will be saved or is it just plain stupid from a developer point of view. Could any of the resident developers give a ballpark figure on the percentage performance drop from the lack of out of order execution on the PU.

Thanks.

Thanks for being down to earth with your ideas.

1) Personally, I don't think it'll be a tri core CPU. Also the processor is a "clean slate" according to DeanoC so allows there to be a sizable difference from the current Power5 in speed in "power'" either way.

2) I don't believe that cell will really go the way of becoming a standard pc. If IBM wants to sell this chip in a server, then creating relatively powerful PCs for cheap would only cut into it's profits for the line since computers can put in a grid relatively "easy". Maybe as an extremely expensive workstation but not a standard PC.

3) The cell shown probably won't be the same as the one in the PS3. What they described is the top of the line cell they had. It isn't necessarily what will go in the PS3 even with a shrink. Because of heat output I believe that the clock will be lower or there will be less spus (whatever they're called), expecially if they launch with 90nm chips and only shrink later.

 

[Alejux]

2)Cell architecture chips may/will eventually end up in personal computers:
Why on earth would Sony even bother to undertake the aforementioned project ? I really do not see PC users jumping all over such a product. I mean zero compatibility with market dominating Microsoft products and a completely new OS, not to mention very little in the way of speed gains for web browsing or document editing anyway. The only way to push it on the desktop would be to use Linux and even then there is no incentive to stop buying a conventional PC.

Imagine a linux desktop, that runs all the current linux applications at a decent speed, and still has the hability to play all the game titles of the PS3!
The main problem with linux at the moment, IMO, is the lack of games. Otherwise, you already have a wide range of software to use. At least the the majority of users.

I'm not saying that the PS3 will kill the PC, but I think it could take a decent chunk of the PC market.

 

[London Geezer]

Imagine a linux desktop, that runs all the current linux applications at a decent speed, and still has the hability to play all the game titles of the PS3!
The main problem with linux at the moment, IMO, is the lack of games. Otherwise, you already have a wide range of software to use. At least the the majority of users.

I'm not saying that the PS3 will kill the PC, but I think it could take a decent chunk of the PC market.

How?
The main problem with Linux is NOT the lack of games. The problem with Linux is that most other programs run exclusively on Windows.
There are 1Billion PCs in the world, only a ridiculously tiny percentage of those are used to play games like we do here. Just think, a game like HL sells less than 5 million copies. Nothing compared to how many PCs there are in the world.

 

[PC-Engine]

Imagine a linux desktop, that runs all the current linux applications at a decent speed, and still has the hability to play all the game titles of the PS3!
The main problem with linux at the moment, IMO, is the lack of games. Otherwise, you already have a wide range of software to use. At least the the majority of users.

I'm not saying that the PS3 will kill the PC, but I think it could take a decent chunk of the PC market.

How?
The main problem with Linux is NOT the lack of games. The problem with Linux is that most other programs run exclusively on Windows.
There are 1Billion PCs in the world, only a ridiculously tiny percentage of those are used to play games like we do here. Just think, a game like HL sells less than 5 million copies. Nothing compared to how many PCs there are in the world.

I must admit that lb has a very valid point.

 

[London Geezer]

I must admit that lb has a very valid point.

R u feeling ok?

 

[rabidrabbit]

If PS3 comes with internet browsing, emailing, photo/video editing, printing as standard the day you buy it, then it has a chance to become a personal computer of choice at homes that don't yet have a PC.

But realistically, PS3 only has a chance to be a "second pc" at homes, that is used to play PS3 games and browse the internet at living room. All other computing tasks would be done on the "old" PC.

If Cell processor based PC comes to market, it has a chance to replace PC's of today. Cell should be able to run various OS's, who knows if Windows will be converted to run on it. It would not run the same apps, but as a family PC it really doesn't need but some word processing, internet+email, photo editing, maybe burning and media softwares.
Those could be bundled under one, like Apple's iLife.
If that software was bundled in a sub $1000 Cell PC, I can see it has a chance to have an impact.
Yes, I know most of us like to have much more sw in our PC's, like various ripping sw, media players that play even the most exotic formats, tweaking sw etc.. but those are mostly specialized hobbyists.

It would be nigh on impossible to replace PC's on larger companies, but smaller firms and firms that would need the graphics processing power (if there'll be such power that is significantly better than PC's then) could be interested in a Cell based PC.

Or maybe the Cell PC has brighter future as a part of some server farm, or render farm or such.

 

[London Geezer]

If a Cell computer comes out with Sony's bundled software they make now for PC, then we all know what success it will have. None. Sorry to say, Sony, but your programs (imaging and other ones) are just C R A P.

Taking over MS, Intel and AMD in the PC market would take the resources only a big country or a large group of the most powerful companies in the world would have, certainly not a company like Sony. And even then, it would take years.

 

[rabidrabbit]

What about IBM?
The little time of what Ihad tried OS/2, I remember it was quite nice.

Haven't followed their software much since then, do they still have an OS? What (if any) (consumer) software does IBM have?

 

[London Geezer]

What about IBM?
The little time of what Ihad tried OS/2, I remember it was quite nice.

Haven't followed their software much since then, do they still have an OS? What (if any) (consumer) software does IBM have?

OS2 had a HUGE advertising campaign, and still it went nowhere.
IBM is co-developing the software for Cell, but it's still early, nt much is known at the moment.

 

[Guden Oden]

OS2 had a HUGE advertising campaign, and still it went nowhere.

Mainly because MS worked actively to kill it, I might add. Technically, OS/2 was far, far superior to anything they had at the time.

 

[London Geezer]

OS2 had a HUGE advertising campaign, and still it went nowhere.

Mainly because MS worked actively to kill it, I might add. Technically, OS/2 was far, far superior to anything they had at the time.

Not sure about that, but i guess if that's true, it only shows that technological superiority will not be enough to change the current situation in the PC market.

 

[Kalin]

Imagine a linux desktop, that runs all the current linux applications at a decent speed, and still has the hability to play all the game titles of the PS3!
The main problem with linux at the moment, IMO, is the lack of games. Otherwise, you already have a wide range of software to use. At least the the majority of users.

I'm not saying that the PS3 will kill the PC, but I think it could take a decent chunk of the PC market.

How?
The main problem with Linux is NOT the lack of games. The problem with Linux is that most other programs run exclusively on Windows.
There are 1Billion PCs in the world, only a ridiculously tiny percentage of those are used to play games like we do here. Just think, a game like HL sells less than 5 million copies. Nothing compared to how many PCs there are in the world.

Wrong.

The main problem with Linux is FUD, as it always was with MS competitors.

There are distros out there that run on both x86 and PPC, perform substantially better than XP in the usability, stability (and don't start me on security) department, and offer everything a home PC user might want out of the box, charging nothing.

 

[London Geezer]

Wrong.

The main problem with Linux is FUD, as it always was with MS competitors.

There are distros out there that run on both x86 and PPC, perform substantially better than XP in the usability, stability (and don't start me on security) department, and offer everything a home PC user might want out of the box, charging nothing.

I never said Windows performs better. I never commented on how Linux and Windows work. I only commented on the commrecial situation of the PC market.
I said it will take a lot to get all the PCs in the world to switch over, and the lack of "popular" software isn't helping.

 

[Alejux]

How?
The main problem with Linux is NOT the lack of games. The problem with Linux is that most other programs run exclusively on Windows.
There are 1Billion PCs in the world, only a ridiculously tiny percentage of those are used to play games like we do here. Just think, a game like HL sells less than 5 million copies. Nothing compared to how many PCs there are in the world.

Linux today, does not particularly attract the common user. Most people I've seen changing to Linux, are computer geeks, who still keep a windows partition for games and a few apps, and some companies, because of it's cost reduction.

The disadvantages of a linux system today, are still too big for the system to become attractive to the common user. Some of them are damn hard to install and configure; There are practically no games; and lacks some attractive software , like photoshop and a few others.

There have been some nobel efforts in the past to create a cheap and popular system linux. But these systems were both cheap in price as well as in performance.

I believe, that if SONY, IBM , APPLE or whoever, decided to create a cell-based desktop, with practically super-computing power, easy to use as a Macintosh and with the whole gaming force of the PS3 backing it up...I think we would have a killer-product, that would attract millions of PC users worldwide.

 

[MfA]

I dont think a Cell based desktop computer will penetrate the market very quickly. If they just put a DVI connector on the PS3 and allow you to use it as a homecomputer with an external HD that could be interesting though.

The problem with running linux on the PS3 is that they cannot allow free distribution of binaries which can just run on PS3s ... because it has the potential to undermine their business model (not because of piracy, which could be protected against, but because software distributors could just go around them). I wonder if they could somehow arrange it such that all binaries distributed to run in an "open" PS3 mode would have to be freely copyable and distributeable.

Still, they could only do that once the production costs fell enough to give them some margin on the PS3.

IMO this is far too revolutionary to expect from Sony though, even if they do end up using Linux.

 

[rabidrabbit]

Many people just think they need all the huge software base a PC offers, when in reality they really use only some 5 different software (which might even be bundled under one "package").

Why we have so much software in our PC's? Because we either need compatibility that some other sw doesn't have, or software doesn't crash the PC like the one I used earlier did

In reality, what prevents a cell based PC, or PS3 from taking over the PC is DRM, that a "Cell" PC would probably not run pirated software and eMule (at least not for a long time after launch).

 

[London Geezer]

How?
The main problem with Linux is NOT the lack of games. The problem with Linux is that most other programs run exclusively on Windows.
There are 1Billion PCs in the world, only a ridiculously tiny percentage of those are used to play games like we do here. Just think, a game like HL sells less than 5 million copies. Nothing compared to how many PCs there are in the world.

Linux today, does not particularly attract the common user. Most people I've seen changing to Linux, are computer geeks, who still keep a windows partition for games and a few apps, and some companies, because of it's cost reduction.

The disadvantages of a linux system today, are still too big for the system to become attractive to the common user. Some of them are damn hard to install and configure; There are practically no games; and lacks some attractive software , like photoshop and a few others.

There have been some nobel efforts in the past to create a cheap and popular system linux. But these systems were both cheap in price as well as in performance.

I believe, that if SONY, IBM , APPLE or whoever, decided to create a cell-based desktop, with practically super-computing power, easy to use as a Macintosh and with the whole gaming force of the PS3 backing it up...I think we would have a killer-product, that would attract millions of PC users worldwide.

And i'd be very happy if that happens and it doesn't cost and arm and a leg and i can use ALL of the programs i use today, or programs that let me do the same thing without having to learn new methods.

Price is a huge factor as well. Apple hasn't taken over the PC market becasue of all these factors, price, software availability, public acceptance (expecting a window user to switch over is a huge task, let me tell u).

Personally i think it's more likely that MS will notice how things are turning out and try to butt in. I can see a huge MS/IBM/Sony group one day in the future. That's the easiest way to progress. But it won't happen, anytime soon at least.

 

[Kalin]

Wrong.

The main problem with Linux is FUD, as it always was with MS competitors.

There are distros out there that run on both x86 and PPC, perform substantially better than XP in the usability, stability (and don't start me on security) department, and offer everything a home PC user might want out of the box, charging nothing.

I never said Windows performs better. I never commented on how Linux and Windows work. I only commented on the commrecial situation of the PC market.
I said it will take a lot to get all the PCs in the world to switch over, and the lack of "popular" software isn't helping.

What I meant is that the software would not be a problem. The thing lacking is the proper PR, to steal some mind share, and in this department I'd trust Sony and IBM to do well.