Zig&Zag: Now Sony management change, layoffs

Shifty Geezer said:
I haven't a link to hand, and admit maybe I just imagined all those announcements from Sony and Toshiba about wanting to include Cell in CE good like TVs. Perhaps Cell will only ever be used in PS3 then, and Toshiba just wasted loads of resources on it and are never going ot use the technology themselves?

I don't see why anyone would look at a programmable media processor like Cell and want to use it in a fridge though and don't know anyone but yourself who even suggested such a ridiculous notion.

Talk is nice, but lets see how widespread this CELL vision turns out...

Let's see if it's cheaper than processors based on MIPS, ARM, TI, etc.
 
Yes, wait and see is the right approach. Look at possibilities and discuss what may or may not happen, always appreciating that anything may happen. Making ridiculous statements about using Cell in totally inappropriate places doesn't really make for any intelligent discussion though, does it?
 
Shifty Geezer said:
Yes, wait and see is the right approach. Look at possibilities and discuss what may or may not happen, always appreciating that anything may happen. Making ridiculous statements about using Cell in totally inappropriate places doesn't really make for any intelligent discussion though, does it?

I was just talking about the people who've been talking about how CELL could be used in all kinds of device when in reality, it really has limited uses. TVs don't require a CELL. Receivers don't require a CELL. HD DVD players don't require a CELL. Mobile phones don't require a CELL. MP3 players don't require a CELL. Computers don't require a CELL. The list goes on and on. As Randycat99 would say, CELL is a solution looking for a problem. SONY is just blaming their floundering CE business on ICs purchased from others which is a poor excuse. Look at any of the other successful CE companies out there, are their CE business floundering because of outsourced ICs?
 
Why not? They all need a processor of some sort. What's so bad about using Cell? The only obvious concern would be cost, but Sony and Toshiba seem convinced they can save money using their own processors then buying in others. And do you disagree that a uniform processor across different devices woud have benefits in development as has been discussed (mostly by me!) already in this thread?
 
PC-Engine said:
I was just talking about the people who've been talking about how CELL could be used in all kinds of device when in reality, it really has limited uses. TVs don't require a CELL. Receivers don't require a CELL. HD DVD players don't require a CELL. Mobile phones don't require a CELL. MP3 players don't require a CELL. Computers don't require a CELL. The list goes on and on. As Randycat99 would say, CELL is a solution looking for a problem.

They don't require it, fair enough, but maybe, by putting Cell in those kind of devices (not all of those!), those devices will be able to do "more". And maybe provide a better overall experience. Escpecially when the same chip is used for all kinds of devices, i'd think communication would be much easier.

Is it "needed"? No. Will some devices become better devices by including a very powerful processor for all kinds of things that they are not doing now? Most probably.

Is HDTV "needed"? No. But it sure looks pretty.
 
PC-Engine said:
I was just talking about the people who've been talking about how CELL could be used in all kinds of device when in reality, it really has limited uses. TVs don't require a CELL. Receivers don't require a CELL. HD DVD players don't require a CELL. Mobile phones don't require a CELL. MP3 players don't require a CELL. Computers don't require a CELL. The list goes on and on. As Randycat99 would say, CELL is a solution looking for a problem.

lol very true. I'm wondering where exactly CELL will fall into place. It truly seems tailored for the gaming sector (obviously) but its being marketed as more broad then that. I personally find it being used a a computers CPU a bit of a stretch on Sony's and IBM's part (haven't heard to much from Toshiba). Maybe with revisions to CELL (years down the road) it will be optimized to work better with hand made OSs but as far as dramatic changes to CELL..what else can be done.
 
Shifty Geezer said:
Why not? They all need a processor of some sort. What's so bad about using Cell? The only obvious concern would be cost, but Sony and Toshiba seem convinced they can save money using their own processors then buying in others. And do you disagree that a uniform processor across different devices woud have benefits in development as has been discussed (mostly by me!) already in this thread?

I don't think the cost of development is a good excuse at all. I've talked to real people who work for companies who make ICs for different types of workloads and they've told me it's very cheap and the development environment is very mature and very fast.

Is HDTV "needed"? No. But it sure looks pretty.

HDTV is needed because people want it and are asking for it.
 
PC-Engine said:
I don't think the cost of development is a good excuse at all. I've talked to real people who work for companies who make ICs for different types of workloads and they've told me it's very cheap and the development environment is very mature and very fast.



HDTV is needed because people want it and are asking for it.

So we only need what we ask for? We should not get anything that we don't ask for?
 
london-boy said:
So we only need what we ask for? We should not get anything that we don't ask for?

Don't put words into other's mouths. People don't care if their TV has a CELL or a TI DSP which is totally differrent from whether or not they care if it's SDTV or HDTV. You brought up the retarded analogy so I simply replied.
 
PC-Engine said:
I don't think the cost of development is a good excuse at all. I've talked to real people who work for companies who make ICs for different types of workloads and they've told me it's very cheap and the development environment is very mature and very fast.
Sure, but those aren't 11 FO4 pipelined designs. Sticking to a single architecture still saves them some money, and their sophistication means they can get away with being less application specific without necessarily needing more mm2.
 
PC-Engine said:
Don't put words into other's mouths. People don't care if their TV has a CELL or a TI DSP which is totally differrent from whether or not they care if it's SDTV or HDTV. You brought up the retarded analogy so I simply replied.

Welcome back PC-Engine... :rolleyes:

People don't care if TVs have Cell, but that's not the point. They don't need TO KNOW that Cell is there, but the advantages that Cell might bring to the TV (for example) will still be there! That's IF Cell will bring anything new to the table, which we still don't know.

Anyway, me being the HDTV freak i am, i think you guys should see this, it's pretty cool:

http://www.cornbread.org/FOTRCompare/index.html (take the mouse on and off the pictures to see the difference between the HD version and the DVD version)
 
MfA said:
Sure, but those aren't 11 FO4 pipelined designs. Sticking to a single architecture still saves them some money, and their sophistication means they can get away with being less application specific without necessarily needing more mm2.

I don't think the money they will save is as much as they make it out to be, let alone turn their CE business around. It's more of a way to try and market the CELL architecture to sell to other companies like what the EE was supposed to have done.

Anyway, me being the HDTV freak i am, i think you guys should see this, it's pretty cool:

You'd be surprised how even an upscaling DVD player can improve a 480p movie. Upscaling a standard DVD movie from 480p to 1080i gives the picture a noticable improvement. I'd expect true HD movies would look even better than upscaled DVDs.:cool:
 
PC-Engine said:
I don't think the cost of development is a good excuse at all. I've talked to real people who work for companies who make ICs for different types of workloads and they've told me it's very cheap and the development environment is very mature and very fast.
Yes it worlks fine now. The industry has grown up around custom and generic IC's of all sorts of different flavours. However a single unified hardware would mean portability of code. As I said elsewhere, if you develop an audio enhancement filter for a Cell hifi, you can port it directly to PS3 and TVs.

I believe Cell is lookng forward to a different way of doing things, and offering a solution to fulfil problems that aren't very apparent at the moment. It's akin to 3D graphics cards. We got by playing games without 3D graphics cards but once the hardware came along 2D died the death. Right now people can get by without 3D accelerators in their mobile phones and megacomplex DSPs in their home entertainment kit, but in the future we'll have found all sorts of uses for that power. Once upon a time someone rather famous said computers would never need more than 640k of memory. Do you REALLY want to say future CE goods aren't going to ever use more than current levels of processing power? And in that future do you really think developers would prefer to work with 4 different complex chip designs then 1 complex chip design?

Of course Cell might flunk out, and be too costly, and rivals may find another solution. But the concept at least of a scalable unified platform seems valid.
 
Shifty Geezer said:
Yes it worlks fine now. The industry has grown up around custom and generic IC's of all sorts of different flavours. However a single unified hardware would mean portability of code. As I said elsewhere, if you develop an audio enhancement filter for a Cell hifi, you can port it directly to PS3 and TVs.

I believe Cell is lookng forward to a different way of doing things, and offering a solution to fulfil problems that aren't very apparent at the moment. It's akin to 3D graphics cards. We got by playing games without 3D graphics cards but once the hardware came along 2D died the death. Right now people can get by without 3D accelerators in their mobile phones and megacomplex DSPs in their home entertainment kit, but in the future we'll have found all sorts of uses for that power. Once upon a time someone rather famous said computers would never need more than 640k of memory. Do you REALLY want to say future CE goods aren't going to ever use more than current levels of processing power? And in that future do you really think developers would prefer to work with 4 different complex chip designs then 1 complex chip design?

Of course Cell might flunk out, and be too costly, and rivals may find another solution. But the concept at least of a scalable unified platform seems valid.

In theory, Sony could have invented a chip that runs x86 code and still accomplish the same thing. I'm not sure why portability of code is being trumpeted as a revolutionary idea.

My question remains, aside from Sony being able to cut costs, what about the cell is beneficial to the consumer? What can the cell do, that no other chip can do, that me as the consumer will be thinking 'wow this cell processor is for real'?

My conention with sony on the cell is the marketing claims that the cell can deliver something that no other chip can deliver (broadband content, yadda, yadda).

Yes a unified chip with portable code makes sense so they can scale their IP across multiple product lines, that part i get. But why it can only be a "cell" to do these things doesnt make any sense and doesnt have me convinced that in the end it will provide any technological advantage to their products or benefit to the consumer other than price (and Sony competing on price is something i'll believe when i see)

J
 
expletive said:
My question remains, aside from Sony being able to cut costs, what about the cell is beneficial to the consumer? What can the cell do, that no other chip can do, that me as the consumer will be thinking 'wow this cell processor is for real'?
Anything 'media' related. It can do, more effectively than any other processor I know of, image processing, audio processing, audio synthesis, DSP work, and host an OS to provide a GUI. An x86 isn't going to make an effective software synthesizer. A DSP isn't going to powerful a graphical front-end for a media centre. A custom IC for a TV based on a MIPS core isn't going to be any use in a mobile phone with audio, video and 3D gaming features. There's other solutions to do all these things but i doubt any will be as cost effective as an ideally deveoped Cell solution. Let's take 1 million TVs, 1 million handhelds, 1 million consoles and 1 million TiVo like devices. If they all use different chips economies of scale will be on 1 million chips. If they all use Cell economies of scal are on 4 milion chips. In Sony's greatest dream, if everything that could be Cell was Cell, the economies of scale would be enormous (and so would Sony's share price!)

For the consumer it potentially offers lower priced, better performance CE goods.
 
Shifty Geezer said:
Anything 'media' related. It can do, more effectively than any other processor I know of, image processing, audio processing, audio synthesis, DSP work, and host an OS to provide a GUI. An x86 isn't going to make an effective software synthesizer. A DSP isn't going to powerful a graphical front-end for a media centre. A custom IC for a TV based on a MIPS core isn't going to be any use in a mobile phone with audio, video and 3D gaming features. There's other solutions to do all these things but i doubt any will be as cost effective as an ideally deveoped Cell solution. Let's take 1 million TVs, 1 million handhelds, 1 million consoles and 1 million TiVo like devices. If they all use different chips economies of scale will be on 1 million chips. If they all use Cell economies of scal are on 4 milion chips. In Sony's greatest dream, if everything that could be Cell was Cell, the economies of scale would be enormous (and so would Sony's share price!)

For the consumer it potentially offers lower priced, better performance CE goods.

Dont you think its a little premature to think that companies whose core competency is multimedia dedicated chips wont be making products that can do all the things youve already talked about the cell being able to do?

Multiple HD streams, audio processing, etc. There are companies BUILT around these technologies (and big companies too) and products. You think they dont know where the market is going and what the needs are going to be over the next 5 years?

Youre hung up on the cell being able to work in various devices. You know what, as a consumer, I DONT CARE that the same chip is in my DVD player, TV, toaster, blender, i just want the thing to perform well.

Sony isnt marketing the cell as 'a chip that will lower prices' they are marketing it as a chip that will 'change the world of multimedia.'

If they were saying to the public 'this chip will allow us to maintain quality at a lower price' great, no problem with that.

But this whole 'cells working in harmony with broadband content that will make air obsolete' hype is just too much.

I'm happy to be wrong on this i just dont see the technological/performance advantage at all. I'd love for air to be obsolete :)

J
 
Shifty Geezer said:
Anything 'media' related. It can do, more effectively than any other processor I know of, image processing, audio processing, audio synthesis, DSP work, and host an OS to provide a GUI. An x86 isn't going to make an effective software synthesizer. A DSP isn't going to powerful a graphical front-end for a media centre. A custom IC for a TV based on a MIPS core isn't going to be any use in a mobile phone with audio, video and 3D gaming features. There's other solutions to do all these things but i doubt any will be as cost effective as an ideally deveoped Cell solution. Let's take 1 million TVs, 1 million handhelds, 1 million consoles and 1 million TiVo like devices. If they all use different chips economies of scale will be on 1 million chips. If they all use Cell economies of scal are on 4 milion chips. In Sony's greatest dream, if everything that could be Cell was Cell, the economies of scale would be enormous (and so would Sony's share price!)

For the consumer it potentially offers lower priced, better performance CE goods.

MIPS cores can be used with a variety of different devices. A TV using a MIPS core with built-in SGX 3D core for fancy GUI can be used in a cellphone. How is this different from a CELL in a TV vs a CELL in a cellphone?
 
Ok guys, cool party.

Anyway so the battle lines are drawn as far as opinion goes - let's just wait and see how the next couple of years go.

I think Cell has the opportunity to be revolutionary in the CE space; some others don't. But the points we're using to argue with each other are all valid, so I think we're just going to have to wait to see who was 'right.'
 
PC-Engine said:
As Randycat99 would say, CELL is a solution looking for a problem.

Don't put words in my mouth. I never said this- EVER. You really deserve another 1 week ban for this unnecessary remark.
 
randycat99 said:
Don't put words in my mouth. I never said this- EVER. You really deserve another 1 week ban for this unnecessary remark.

I didn't say you said CELL specifically so don't get your panties in a knot.
 
Back
Top