Sony is ailing because its a disfuctional company. The costs of developing and selling the PS3 do not in any way match the revenues it is bringing in. Sony isn't going to share those meagre revenues with anyone. There is no cash benefit to Apple.
If Apple wanted to become a games company, it should do the exact opposite of what Sony has done.
Adding a "proper" OS to the PS3 isn't going to benefit Sony. It plays games, it plays video, it isn't selling. Porting iLife is not going to suddenly make it sell either. There is no cash benefit to Sony.
From October 2007 to November 2007, sales of the PlayStation 3 went up by 285% in the United States.[65] Sony CEO Howard Stringer contributes the growth to a price cut and Nintendo's problems with production of the Wii system.[66]
There are three times as many 360s out there. (now 17million)
A similar amount of WIIs
And remember, every PS3 sold is another $300 Sony dollars down the toilet. This is an unrecoverable position for Sony.
In many ways it is worse than the postion faced with the original Xbox. The small Xbox market share was a problem for games developers, but the architecture of the XBox made it relatively cheap to port PS2 titles onto the XBox.
The bizarre architecture of the PS3 means no one wants to incur the risk of a massive porting exercise for such a small market share.
People might not think Sony and the PS series is a good marriage for Apple but the question remains - what is Apple to do about gaming?
I think getting Autodesk and Maya would be sweet. I could see Apple redoing the interface and dropping the price greatly. It would complement FCP Suite nicely. Also makes sense since Pixar uses Renderman with Maya. They then control the app they use.
The only question is whether Apple is growing Motion into a 3D app that can do most of what Maya does. Maybe that is why Apple didn't touch Alias.
Autodesk and the architect/engineering angle is another area altogether though. Every market Apple can dominate is good for them. This is the same question people ask about Adobe. Apple needs more FCP and Aperture high end apps for the pros. Then they need more iLife programs for the consumers.
But back on topic.
Gaming is huge for teenagers and consoles are cheaper and more stable than PCs. Apple getting onto the PS series would open many youth to the Mac platform. It would also put Apple into the living rooms in a big way.
Look at it this way, how would you like it if the Xbox did all the things Apple should do in the living room - movie and music store, email, video chat, web browser, single remote, wireless 5.1 sound. What if the Zoon did video conferencing from anywhere, and it integrated with PCs and with game consoles? Whould this not be a killer app? I think so.
Some people might say some of this has been tried and it didn't work. WebTV did not take over the world. Yes, for standard def. But HD and progressive displays change all that. Nobody wants to browse the web with a 480i SD screen. But a 720p or even 1080p screen is different. Add resolution independence and one could actually use OS X on a HDTV.
Just because the Newton didn't light the world on fire didn't mean the PDA was a bad idea. Same with media centers and network computers. Apple needs something to dominate the living room and right now games are the biggest obstacle.
I also think Apple needs to see what OS X can do on the PPC and Cell processors. PPC may not be aiming for the desktop market anymore but embedded processors are huge for stuff like handheld devices. Apple uses ARM but who is to say the PPC might not be a better processor for the iPhone now or some day in the future?
Apple also needs to have OS X ready to port to any new processor that comes out. (I think they already are in fact, but they should never lose this capability.) I am not just talking about new processors from other companies like AMD. This also includes any new design from Intel. Apple is not tied down with legacy problems like the Windows/x86 world is. What can Intel do once freed from those bonds? They might not have incentive without a computer to use the chip, but now they have Apple that can use the chip.
The question is what is the future of CPUs? Is it massively parallel processing akin to the Cell processor? If so, Apple needs to get there first. Apple already has dual quad machines - 8 cores. What comes next - 16, 32, 1024 cores? Somehow I doubt x86 is the best implementation for this future.
So is Cell the way of the future? Again, first generation products usually do maximize the potential of a new way of doing things. What if the next Cell processor has 1024 logic cores? Having OS X on PPC/Cell is not a bad hedge.
There are three times as many 360s out there. (now 17million)
A similar amount of WIIs
And remember, every PS3 sold is another $300 Sony dollars down the toilet. This is an unrecoverable position for Sony.
In many ways it is worse than the postion faced with the original Xbox. The small Xbox market share was a problem for games developers, but the architecture of the XBox made it relatively cheap to port PS2 titles onto the XBox.
The bizarre architecture of the PS3 means no one wants to incur the risk of a massive porting exercise for such a small market share.
C.
360 has been aggressively sold globally for 2 - 3 years.
That 17 million isn't squat.
With SONY approaching 6 million mainly 1 year out is more impressive. Why? The total market segments on the PS3 with the PS2 is massive. The PS2 is still a huge sale. The XBox has no XBox180 to compete against itself.
Come 2008 expect it alone to sell 10 - 15 million PS3 units as the price continues to drop.
HALO 3 accounts for over 8 Million units sold of the XBox.
360 has been aggressively sold globally for 2 - 3 years.
That 17 million isn't squat.
With SONY approaching 6 million mainly 1 year out is more impressive. Why? The total market segments on the PS3 with the PS2 is massive. The PS2 is still a huge sale. The XBox has no XBox180 to compete against itself.
Come 2008 expect it alone to sell 10 - 15 million PS3 units as the price continues to drop.
HALO 3 accounts for over 8 Million units sold of the XBox.
SONY will regain it's footing and further drive the demand for BluRay.
What a bizarre post!
Sony started the PS3 project long before Microsoft, but they were unable to finish the hardware. Smart move! Letting Microsoft do all that "aggressive marketing!" Luckily that late launch allowed developers extra time to create killer games.
Oh, that's right, that didn't happen.
Perhaps everything will be alright when Sony slash the price? Yes, I am sure that losing $400 per unit will make Sony much much more profitable.
Walter... Great benchmark man. The real question is...
How exactly does six lighting-fast floating-point units help you to....
d) Run a video game?
The answers are...
d) Help! Anyone know Sony's engineering helpline?
C.
'
I am no game developer, but I call bullshit. Oblivion, GTA 4, RFoM, Assassin's Creed, MGS4, GT 4, Kill Zone 2, SOCOM, Racket and Clank, the list goes on. If Cell sucked so badly these games wouldn't be possible much alone look so good. I am no PS fanboy by any means but if Cell sucks so bad, why don't PS3 games would look like Wii games?
Just saying Cell is a terrible thing doesn't give us a lot to go on. While that might be the correct conclusion, there is very little reason to support that conclusion.
So completing the hardware was long and programming games is tough. I would expect all that with a big switch in paradigms. The question is that all a thing of the past and are future generations of hardware and games going to go smoothly and quickly or not? I don't know, but I wouldn't be too quick to dismiss the Cell until I hear some good data from knowledgeable people in the trenches.
On the other side, if the games look better than everything else, that is an argument in the Cell's favor. Everybody knows it takes time to maximize the potential of a hardware. I think most people think the PS3 is not maximized yet. So maybe the best is still to come.
As for market share, that only counts so much. How much market share did the Wii have when it started out? Just because some console has tons of units doesn't mean it will be around in the future.
There is a point to Apple sitting on the sidelines since the gaming market seems to be losing so much for everybody. But Apple is jumping into the living room and right now, consoles are the elephant in the room. Apple cannot just ignore it.
I'm guessing Apple is going to pull a Wii and fight the battle on different terms. A tv with iSight and a new iPhone with iSight all tied together by iChatAV and some 3G or 4G technology would be an outflanking maneuver around consoles. Movie rentals help too.
I'm guessing AppleTV will be Apple's answer and we will see it soon. Unfortunately it will be just one step at a time. I just wish they would jump tot he endzone rather than go for these first downs.
I am no game developer, but I call bullshit. Oblivion, GTA 4, RFoM, Assassin's Creed, MGS4, GT 4, Kill Zone 2, SOCOM, Racket and Clank, the list goes on. If Cell sucked so badly these games wouldn't be possible much alone look so good. I am no PS fanboy by any means but if Cell sucks so bad, why don't PS3 games would look like Wii games?
Dave
Cell was trumpeted by Sony as a revolutionary technology, superior to everything else. It's true.
Cell is an *incredibly fast* math processor - but as a general-purpose CPU it's about the same speed as a single G5.
Cell is so fast at math, that Sony's engineers thought that they could use it to render the graphics. But late in the development realized that it was not going to work. So added a high-end NVidia GPU. It's a good chip but the late addition meant that there are bandwidth issues.
At the end of the day the PS3 is not seriously underpowered. When you see side-by side comparisons with the same title on the 360, it is hard to tell them apart. But you'll notice that most games in your list were late on the PS3. Kill Zone 2 is still unfinished. GT4 is the same game with new graphics. Oblivion took another 10 months.
The difficulty in writing games undermines profitability for developers. Combine that with the less-than-stellar market share and you can see why the platform is being abandoned.
Is the Cell being abandoned? Is there another generation in the works? Based on what info? Is this fact or speculation?
One could look at the Wright brothers Flyer and say there was not much potential in powered airplanes. But now we have the F-22 Raptor. A first generation anything is usually not the full potential of a concept.
Not saying the Cell itself necessarily is the future, but I think multiple cores are. In some ways hardware is like software. Once you have the design done, stamping out copies is cheap and easy. The big hurdle then becomes bandwidth and intercommunication among the cores.
There is so much parallelism in processes today, especially in audio, video, and graphics. For most other stuff we don't need faster processors. Many hands do make light work. Same goes for computer tasks. Just ask F@H.
Cell was trumpeted by Sony as a revolutionary technology, superior to everything else. It's true.
Cell is an *incredibly fast* math processor - but as a general-purpose CPU it's about the same speed as a single G5.
Cell is so fast at math, that Sony's engineers thought that they could use it to render the graphics. But late in the development realized that it was not going to work. So added a high-end NVidia GPU. It's a good chip but the late addition meant that there are bandwidth issues.
C.
Sony must have some really stupid, high-salaried engineers. How did this happen?
Cell will be around for a while. It's the style of things to come: i.e. application processors. The people I've met that disapprove of it are mainly the software developers who aren't talented enough to rapidly make the mental transition to the Cell's very different architecture. Nonetheless, they are starting to get used to it and better developing tools are starting to become available.
I think it would be a good move although I don't see Apple putting OS X on any hardware they don't have complete control over.
Apple is going to get a lot of competition in the home media delivery market and with stupid products like the Apple TV, they need to start getting something better. The easy option would be to use a product that is already out there that needs a stable, easy-to-use OS that doesn't come from Sony's main rival in the gaming market.
Sony could probably use the interface that OS X offers and it has great PPC support having come from a PPC-only hardware lineup.
So Sony get a commercially supported OS that has been proven to work well on PPC hardware along with the nice interface and developer libraries. Apple get another content delivery platform and possibly some discounts on Blu-Ray drives.
I would like to see Apple and Sony together and it would most certainly help Apple in Japan and China.
Cell will be around for a while. It's the style of things to come: i.e. application processors. The people I've met that disapprove of it are mainly the software developers who aren't talented enough to rapidly make the mental transition to the Cell's very different architecture. Nonetheless, they are starting to get used to it and better developing tools are starting to become available.
Splinemodel... Baby .... You are talking to the wrong guys.
Greybeard software developers *love* these crazy-ass architectures. They get paid $1200 per day to come in to untangle the spaghetti. It's just more grist for the mill. (Whatever the heck "grist" is).
Funny thing is. Despite hiring these guys, the profitability needle is still pointing into the red. Why is that?
The difficulty in writing games undermines profitability for developers. Combine that with the less-than-stellar market share and you can see why the platform is being abandoned.
C.
Riiight...the PS3 is dead and being abandoned.
Sony made money on the PS2 but decided that winning the HD format war was worth taking a loss on the PS3. Sony expects breakeven on the PS3 sometime this year...they claimed by end of this FY but I'm thinking maybe, maybe not given they're selling their fabs to Toshiba (who presumably like getting paid to work).
But hey...the 40GB PS3 has the 65nm Cell so who knows. Maybe they can hit breakeven by March on each unit sold.
Splinemodel... Baby .... You are talking to the wrong guys.
Greybeard software developers *love* these crazy-ass architectures. They get paid $1200 per day to come in to untangle the spaghetti. It's just more grist for the mill. (Whatever the heck "grist" is).
Funny thing is. Despite hiring these guys, the profitability needle is still pointing into the red. Why is that?
C
The trend is that consumers are demanding more and more exotic performance out of our computers and electronics, and that's not going to change. What's also not going to change is that the kind of machines that were optimized to work with the programmatic models of C/C++/Java aren't going to cut it as exotic performers. So you can try to convince consumers that they don't actually want cool stuff or you can try to convince programmers to go back to school. Guess who's going to be conceding this one in the long term?
The gray-beards are the ones who had K&R as TA's, and they realize that stack-centric software paradigm are really, really old and need to go. If you have hardware that's meant to run in parallel, the syntax of your code needs to have intrinsically parallel features. It's very hard to convince people that they need to learn a whole new language: C/C++/Java are all basically the same and unsuited for use with parallel hardware. There's a big learning curve and it's an uphill battle, in terms of quality and economy. Eventually someone will go the distance and champion a programming toolchain and API that is intrinsically parallel, and he'll probably be very profitable in the long run. Without that, you're going to have to deal with ugly C code and expensive consultants. Intrinsically parallel syntax is still pretty new outside of FPGA programming, but it's a growing movement. NesC is a good example.
The game industry is freaking huge now, not to mention that Cell (or similar architectures) have a lot of promise for use in supercomputers and telecom infrastructure. The next generation of XBox and Nintendo are, without question, going to have to go more-parallel or go home. There are compelling business cases for entrepreneurs who think they can find ways to milk Cell, but these things take time to materialize. In the 70's, 80's, and 90's hardware was chasing software. Now, it's the other way around.
The ship's still floating, but the first three compartments are flooded.
Believe it or not games consoles need games to make money. Game developers are not charitable institutions. Can you name a single PS3 game that's made a profit?
Put another way, if you had $10M in lose change would you invest in...
One PS3 game (with 5M units out there)
Two 360 games (with 18M units out there)
Three Wii games (with 19M units out there)
These numbers have been exaggerated to make the point clearer.
The trend is that consumers are demanding more and more exotic performance out of our computers and electronics, and that's not going to change. What's also not going to change is that the kind of machines that were optimized to work with the programmatic models of C/C++/Java aren't going to cut it as exotic performers. So you can try to convince consumers that they don't actually want cool stuff or you can try to convince programmers to go back to school. Guess who's going to be conceding this one in the long term?
You are looking at a technological trend, and mistaking it for a market trend.
Yes, there is a vocal bunch of kids who want better graphics, they want networked, real-time subsurface-scattering, physically-based subdivision boobies bouncing all over their screen. They want their violence in hi-def ultrarealism.
Technologically it *can* be done. We can give them what they ask.
But. That does not constitute a viable business model.
When you say "consumers want something" it's only true if you can succeed in giving them what they want *and* turn a profit doing it. If you lose money, they are not consumers.
And if you look at the numbers there is no evidence to say this is viable. While there is a lot of evidence to support what Nintendo has done.
Sony started the PS3 project long before Microsoft, but they were unable to finish the hardware. Smart move! Letting Microsoft do all that "aggressive marketing!" Luckily that late launch allowed developers extra time to create killer games.
Oh, that's right, that didn't happen.
Perhaps everything will be alright when Sony slash the price? Yes, I am sure that losing $400 per unit will make Sony much much more profitable.
C.
It took from 2001 - 2005 for the Cell Processor to actual be completed. The lead in time for PS 3 vs. XBox is negligible.
Comments
If Apple wanted to become a games company, it should do the exact opposite of what Sony has done.
Adding a "proper" OS to the PS3 isn't going to benefit Sony. It plays games, it plays video, it isn't selling. Porting iLife is not going to suddenly make it sell either. There is no cash benefit to Sony.
Move along, nothing to see here.
From October 2007 to November 2007, sales of the PlayStation 3 went up by 285% in the United States.[65] Sony CEO Howard Stringer contributes the growth to a price cut and Nintendo's problems with production of the Wii system.[66]
http://en.wikipedia.org/wiki/PlaySta...oduction_costs
Lol!
Ten times nothin' is still nothin'.
There are three times as many 360s out there. (now 17million)
A similar amount of WIIs
And remember, every PS3 sold is another $300 Sony dollars down the toilet. This is an unrecoverable position for Sony.
In many ways it is worse than the postion faced with the original Xbox. The small Xbox market share was a problem for games developers, but the architecture of the XBox made it relatively cheap to port PS2 titles onto the XBox.
The bizarre architecture of the PS3 means no one wants to incur the risk of a massive porting exercise for such a small market share.
C.
I think getting Autodesk and Maya would be sweet. I could see Apple redoing the interface and dropping the price greatly. It would complement FCP Suite nicely. Also makes sense since Pixar uses Renderman with Maya. They then control the app they use.
The only question is whether Apple is growing Motion into a 3D app that can do most of what Maya does. Maybe that is why Apple didn't touch Alias.
Autodesk and the architect/engineering angle is another area altogether though. Every market Apple can dominate is good for them. This is the same question people ask about Adobe. Apple needs more FCP and Aperture high end apps for the pros. Then they need more iLife programs for the consumers.
But back on topic.
Gaming is huge for teenagers and consoles are cheaper and more stable than PCs. Apple getting onto the PS series would open many youth to the Mac platform. It would also put Apple into the living rooms in a big way.
Look at it this way, how would you like it if the Xbox did all the things Apple should do in the living room - movie and music store, email, video chat, web browser, single remote, wireless 5.1 sound. What if the Zoon did video conferencing from anywhere, and it integrated with PCs and with game consoles? Whould this not be a killer app? I think so.
Some people might say some of this has been tried and it didn't work. WebTV did not take over the world. Yes, for standard def. But HD and progressive displays change all that. Nobody wants to browse the web with a 480i SD screen. But a 720p or even 1080p screen is different. Add resolution independence and one could actually use OS X on a HDTV.
Just because the Newton didn't light the world on fire didn't mean the PDA was a bad idea. Same with media centers and network computers. Apple needs something to dominate the living room and right now games are the biggest obstacle.
I also think Apple needs to see what OS X can do on the PPC and Cell processors. PPC may not be aiming for the desktop market anymore but embedded processors are huge for stuff like handheld devices. Apple uses ARM but who is to say the PPC might not be a better processor for the iPhone now or some day in the future?
Apple also needs to have OS X ready to port to any new processor that comes out. (I think they already are in fact, but they should never lose this capability.) I am not just talking about new processors from other companies like AMD. This also includes any new design from Intel. Apple is not tied down with legacy problems like the Windows/x86 world is. What can Intel do once freed from those bonds? They might not have incentive without a computer to use the chip, but now they have Apple that can use the chip.
The question is what is the future of CPUs? Is it massively parallel processing akin to the Cell processor? If so, Apple needs to get there first. Apple already has dual quad machines - 8 cores. What comes next - 16, 32, 1024 cores? Somehow I doubt x86 is the best implementation for this future.
So is Cell the way of the future? Again, first generation products usually do maximize the potential of a new way of doing things. What if the next Cell processor has 1024 logic cores? Having OS X on PPC/Cell is not a bad hedge.
Lol!
Ten times nothin' is still nothin'.
There are three times as many 360s out there. (now 17million)
A similar amount of WIIs
And remember, every PS3 sold is another $300 Sony dollars down the toilet. This is an unrecoverable position for Sony.
In many ways it is worse than the postion faced with the original Xbox. The small Xbox market share was a problem for games developers, but the architecture of the XBox made it relatively cheap to port PS2 titles onto the XBox.
The bizarre architecture of the PS3 means no one wants to incur the risk of a massive porting exercise for such a small market share.
C.
360 has been aggressively sold globally for 2 - 3 years.
That 17 million isn't squat.
With SONY approaching 6 million mainly 1 year out is more impressive. Why? The total market segments on the PS3 with the PS2 is massive. The PS2 is still a huge sale. The XBox has no XBox180 to compete against itself.
Come 2008 expect it alone to sell 10 - 15 million PS3 units as the price continues to drop.
HALO 3 accounts for over 8 Million units sold of the XBox.
http://en.wikipedia.org/wiki/Xbox_360#Sales
Both haven't made a bid for domination.
SONY will regain it's footing and further drive the demand for BluRay.
- what is Apple to do about gaming?
Nothing. As long as Microsoft and Sony are losing billions. The current strategy isn't working. Only Nintendo are making money.
So is Cell the way of the future?
No. It isn't. It's a terrible terrible technology. It's like the G5 but worse.
It sucks down a billion watts. The failure rate is huge, and no one can afford to write programs which exploit its brain-damaged architecture.
C.
360 has been aggressively sold globally for 2 - 3 years.
That 17 million isn't squat.
With SONY approaching 6 million mainly 1 year out is more impressive. Why? The total market segments on the PS3 with the PS2 is massive. The PS2 is still a huge sale. The XBox has no XBox180 to compete against itself.
Come 2008 expect it alone to sell 10 - 15 million PS3 units as the price continues to drop.
HALO 3 accounts for over 8 Million units sold of the XBox.
http://en.wikipedia.org/wiki/Xbox_360#Sales
Both haven't made a bid for domination.
SONY will regain it's footing and further drive the demand for BluRay.
What a bizarre post!
Sony started the PS3 project long before Microsoft, but they were unable to finish the hardware. Smart move! Letting Microsoft do all that "aggressive marketing!" Luckily that late launch allowed developers extra time to create killer games.
Oh, that's right, that didn't happen.
Perhaps everything will be alright when Sony slash the price? Yes, I am sure that losing $400 per unit will make Sony much much more profitable.
C.
Walter... Great benchmark man. The real question is...
How exactly does six lighting-fast floating-point units help you to....
d) Run a video game?
The answers are...
d) Help! Anyone know Sony's engineering helpline?
C.
'
I am no game developer, but I call bullshit. Oblivion, GTA 4, RFoM, Assassin's Creed, MGS4, GT 4, Kill Zone 2, SOCOM, Racket and Clank, the list goes on. If Cell sucked so badly these games wouldn't be possible much alone look so good. I am no PS fanboy by any means but if Cell sucks so bad, why don't PS3 games would look like Wii games?
Dave
So completing the hardware was long and programming games is tough. I would expect all that with a big switch in paradigms. The question is that all a thing of the past and are future generations of hardware and games going to go smoothly and quickly or not? I don't know, but I wouldn't be too quick to dismiss the Cell until I hear some good data from knowledgeable people in the trenches.
On the other side, if the games look better than everything else, that is an argument in the Cell's favor. Everybody knows it takes time to maximize the potential of a hardware. I think most people think the PS3 is not maximized yet. So maybe the best is still to come.
As for market share, that only counts so much. How much market share did the Wii have when it started out? Just because some console has tons of units doesn't mean it will be around in the future.
There is a point to Apple sitting on the sidelines since the gaming market seems to be losing so much for everybody. But Apple is jumping into the living room and right now, consoles are the elephant in the room. Apple cannot just ignore it.
I'm guessing Apple is going to pull a Wii and fight the battle on different terms. A tv with iSight and a new iPhone with iSight all tied together by iChatAV and some 3G or 4G technology would be an outflanking maneuver around consoles. Movie rentals help too.
I'm guessing AppleTV will be Apple's answer and we will see it soon. Unfortunately it will be just one step at a time. I just wish they would jump tot he endzone rather than go for these first downs.
'
I am no game developer, but I call bullshit. Oblivion, GTA 4, RFoM, Assassin's Creed, MGS4, GT 4, Kill Zone 2, SOCOM, Racket and Clank, the list goes on. If Cell sucked so badly these games wouldn't be possible much alone look so good. I am no PS fanboy by any means but if Cell sucks so bad, why don't PS3 games would look like Wii games?
Dave
Cell was trumpeted by Sony as a revolutionary technology, superior to everything else. It's true.
Cell is an *incredibly fast* math processor - but as a general-purpose CPU it's about the same speed as a single G5.
Cell is so fast at math, that Sony's engineers thought that they could use it to render the graphics. But late in the development realized that it was not going to work. So added a high-end NVidia GPU. It's a good chip but the late addition meant that there are bandwidth issues.
At the end of the day the PS3 is not seriously underpowered. When you see side-by side comparisons with the same title on the 360, it is hard to tell them apart. But you'll notice that most games in your list were late on the PS3. Kill Zone 2 is still unfinished. GT4 is the same game with new graphics. Oblivion took another 10 months.
The difficulty in writing games undermines profitability for developers. Combine that with the less-than-stellar market share and you can see why the platform is being abandoned.
C.
One could look at the Wright brothers Flyer and say there was not much potential in powered airplanes. But now we have the F-22 Raptor. A first generation anything is usually not the full potential of a concept.
Not saying the Cell itself necessarily is the future, but I think multiple cores are. In some ways hardware is like software. Once you have the design done, stamping out copies is cheap and easy. The big hurdle then becomes bandwidth and intercommunication among the cores.
There is so much parallelism in processes today, especially in audio, video, and graphics. For most other stuff we don't need faster processors. Many hands do make light work. Same goes for computer tasks. Just ask F@H.
Cell was trumpeted by Sony as a revolutionary technology, superior to everything else. It's true.
Cell is an *incredibly fast* math processor - but as a general-purpose CPU it's about the same speed as a single G5.
Cell is so fast at math, that Sony's engineers thought that they could use it to render the graphics. But late in the development realized that it was not going to work. So added a high-end NVidia GPU. It's a good chip but the late addition meant that there are bandwidth issues.
C.
Sony must have some really stupid, high-salaried engineers. How did this happen?
Apple is going to get a lot of competition in the home media delivery market and with stupid products like the Apple TV, they need to start getting something better. The easy option would be to use a product that is already out there that needs a stable, easy-to-use OS that doesn't come from Sony's main rival in the gaming market.
Sony could probably use the interface that OS X offers and it has great PPC support having come from a PPC-only hardware lineup.
So Sony get a commercially supported OS that has been proven to work well on PPC hardware along with the nice interface and developer libraries. Apple get another content delivery platform and possibly some discounts on Blu-Ray drives.
I would like to see Apple and Sony together and it would most certainly help Apple in Japan and China.
Cell will be around for a while. It's the style of things to come: i.e. application processors. The people I've met that disapprove of it are mainly the software developers who aren't talented enough to rapidly make the mental transition to the Cell's very different architecture. Nonetheless, they are starting to get used to it and better developing tools are starting to become available.
Splinemodel... Baby .... You are talking to the wrong guys.
Greybeard software developers *love* these crazy-ass architectures. They get paid $1200 per day to come in to untangle the spaghetti. It's just more grist for the mill. (Whatever the heck "grist" is).
Funny thing is. Despite hiring these guys, the profitability needle is still pointing into the red. Why is that?
C
The difficulty in writing games undermines profitability for developers. Combine that with the less-than-stellar market share and you can see why the platform is being abandoned.
C.
Riiight...the PS3 is dead and being abandoned.
Sony made money on the PS2 but decided that winning the HD format war was worth taking a loss on the PS3. Sony expects breakeven on the PS3 sometime this year...they claimed by end of this FY but I'm thinking maybe, maybe not given they're selling their fabs to Toshiba (who presumably like getting paid to work).
But hey...the 40GB PS3 has the 65nm Cell so who knows. Maybe they can hit breakeven by March on each unit sold.
Splinemodel... Baby .... You are talking to the wrong guys.
Greybeard software developers *love* these crazy-ass architectures. They get paid $1200 per day to come in to untangle the spaghetti. It's just more grist for the mill. (Whatever the heck "grist" is).
Funny thing is. Despite hiring these guys, the profitability needle is still pointing into the red. Why is that?
C
The trend is that consumers are demanding more and more exotic performance out of our computers and electronics, and that's not going to change. What's also not going to change is that the kind of machines that were optimized to work with the programmatic models of C/C++/Java aren't going to cut it as exotic performers. So you can try to convince consumers that they don't actually want cool stuff or you can try to convince programmers to go back to school. Guess who's going to be conceding this one in the long term?
The gray-beards are the ones who had K&R as TA's, and they realize that stack-centric software paradigm are really, really old and need to go. If you have hardware that's meant to run in parallel, the syntax of your code needs to have intrinsically parallel features. It's very hard to convince people that they need to learn a whole new language: C/C++/Java are all basically the same and unsuited for use with parallel hardware. There's a big learning curve and it's an uphill battle, in terms of quality and economy. Eventually someone will go the distance and champion a programming toolchain and API that is intrinsically parallel, and he'll probably be very profitable in the long run. Without that, you're going to have to deal with ugly C code and expensive consultants. Intrinsically parallel syntax is still pretty new outside of FPGA programming, but it's a growing movement. NesC is a good example.
The game industry is freaking huge now, not to mention that Cell (or similar architectures) have a lot of promise for use in supercomputers and telecom infrastructure. The next generation of XBox and Nintendo are, without question, going to have to go more-parallel or go home. There are compelling business cases for entrepreneurs who think they can find ways to milk Cell, but these things take time to materialize. In the 70's, 80's, and 90's hardware was chasing software. Now, it's the other way around.
Riiight...the PS3 is dead and being abandoned.
The ship's still floating, but the first three compartments are flooded.
Believe it or not games consoles need games to make money. Game developers are not charitable institutions. Can you name a single PS3 game that's made a profit?
Put another way, if you had $10M in lose change would you invest in...
One PS3 game (with 5M units out there)
Two 360 games (with 18M units out there)
Three Wii games (with 19M units out there)
These numbers have been exaggerated to make the point clearer.
C.
The trend is that consumers are demanding more and more exotic performance out of our computers and electronics, and that's not going to change. What's also not going to change is that the kind of machines that were optimized to work with the programmatic models of C/C++/Java aren't going to cut it as exotic performers. So you can try to convince consumers that they don't actually want cool stuff or you can try to convince programmers to go back to school. Guess who's going to be conceding this one in the long term?
You are looking at a technological trend, and mistaking it for a market trend.
Yes, there is a vocal bunch of kids who want better graphics, they want networked, real-time subsurface-scattering, physically-based subdivision boobies bouncing all over their screen. They want their violence in hi-def ultrarealism.
Technologically it *can* be done. We can give them what they ask.
But. That does not constitute a viable business model.
When you say "consumers want something" it's only true if you can succeed in giving them what they want *and* turn a profit doing it. If you lose money, they are not consumers.
And if you look at the numbers there is no evidence to say this is viable. While there is a lot of evidence to support what Nintendo has done.
C.
What a bizarre post!
Sony started the PS3 project long before Microsoft, but they were unable to finish the hardware. Smart move! Letting Microsoft do all that "aggressive marketing!" Luckily that late launch allowed developers extra time to create killer games.
Oh, that's right, that didn't happen.
Perhaps everything will be alright when Sony slash the price? Yes, I am sure that losing $400 per unit will make Sony much much more profitable.
C.
It took from 2001 - 2005 for the Cell Processor to actual be completed. The lead in time for PS 3 vs. XBox is negligible.