Zig&Zag: Now Sony management change, layoffs

[Shifty Geezer]

You're forgetting Cell's scalability. Sony would be daft to put a 1:8 BE into a TV, but they can use a 1:1 Cell chip, and print them off at 3x the density of the BE at 1/3rd the price. They can use the same software across all Cell platforms to perform the same task. eg. A scaling routine running on a TV can be used in a handheld with Cell. An image filtering process can be used on a TV, Cell editing suite (or video camera), and PS3's EyeToy. Cell based drawing functions (OpenGL presumably) can be used on Cell TVs, media players, and mobile phones providing a uniform interface, regardless of whether the hardware has 1, 2, 4 or 8 SPE's. In future Sony can improve manufacturing of Cell, lowering the price and providing much greater profitability than buying in components, and use the same software development strategies. In 10 or 20 years the code and systems used now on Cell will still be in operation, only the processors will just have more SPE's and run faster. Without needing to worry about developing new chips (just add some more SPE's) Sony can focus on the software that's driving these TV sets, mobile devices, media devices, holographics projectors and so forth.

The key point is scalability. Whereas in the past new tech has required new processors, Sony are betting on Cell being a universal solution, and in the long term they'll be able to make these cheaper then any other new scalable solution and have the headway in the software department too. That's where they hope to make back on these huge investments of their's.

 

[expletive]

So the main benefit of the cell in consumer electronics is a long-term cost savings one?

J

 

[Lazy8s]

The PlayStation division had been losing the company money in recent quarters.

 

[scooby_dooby]

If not, i'm sure Sony's philosophy (cram everything in one piece of hardware) will be helped. You might see Cell powered TVs with online components, integrated BDROM, MS/SD/CF slots, 29 USB ports, 3 HDDs, 5" 1/4 Floppy disk drive (who remembers them?), and ice maker.

lol
I was just thinking it seems like a mighty expensive chip for a TV, Even scaled down, you're still talking about a PowerPC with multiple slave processors...couldn't a simple specialized processor be used for much cheaper?

For example, does anyone know the size and cost of current proccessors in TV's?

 

[expletive]

Some of the best video processing available today is done on commoditzed FPGAs and the real competitive advantage is all in the proproetray software algortihms.

J

 

[Josh378]

Who do I have a feeling that when I watch televison with a Cell in it, I will have a sudden urge to buy everything "Sony" or "toshiba" products with a CELL in it.

Subliminal Messages...Rocks!!!!

( I think it is illegal to do that, but if not...it is the beginning of the end of my bank account)

-Jsoh378

 

[Titanio]

It'll be interesting to see how broad and deep co-operation between SCE and the movies and music divisions will be. I'd heard in the past that such co-operation was virtually non-existant (apparently many within other divisions resented how quickly SCE rose and gained influence in the company).

I mean, will this co-operation "simply" be about a few more movie conversions, and better game soundtracks and selling sony movie/music content through PS3/PSP etc. or will SCE start getting first refusal on all Sony movie-to-game rights, for example?

 

[dukmahsik]

sony knows they're too bloated as a company, they are going back to more of the roots which made them so successful, this is good news for the company but bad news for the employees

 

[Acert93]

You're forgetting Cell's scalability. Sony would be daft to put a 1:8 BE into a TV, but they can use a 1:1 Cell chip, and print them off at 3x the density of the BE at 1/3rd the price.

If what Intel has recently said is true, it would be even more cost effective. Intel noted that there is not a linear relation to die size reduction and yields. If you reduce the size of a die 10% the yields go up MORE than 10%.

So you may cut the die size down to 1/3, but get 4x as many chips. But yeah, for HD decoding another beyong a 1:2 CELL would seem overkill. Also, a large Cell would be expensive, draw a bit of power, produce heat, and possibly require active cooling. (On a side note, this is why the 2 die Xenos is a good idea. 332M transistors is a large chip. Breaking it up into to 2 smaller chips, while having some additional packaging costs, will be a perk in yields).

 

[expletive]

From jsut what ive read here, it seems Sony is confident of 2 things:

1. They will be able to produce cell chips cheaper they could buy specialized chips for their consumer electronics devices.

2. Their own video (and maybe audio) processing software/algorithms will be superior than what is available to license from another company. If its not superior it will be cheaper.

These seem like very bold predictions from my standpoint becuase i think its very risky to assume that:

1. A cell will be cheaper to produce than buying an application specific commoditized chip.

2. They will be able to keep up with advances in video processing technology internally. Companies that have IP in this space are profitiable because they sell to the entire industry, i dont see how Sony can be cutting edge when their only consumer is themselves. They can certainly create software that will perform the function but they count themselves out of the 'best of breed' solutions by doing this, negating other competitive advantages.

There's got to be more to the cell than this?

J

 

[Titanio]

There's got to be more to the cell than this?

Yes, it's called Playstation

It's not unlike Sony to bet the farm JUST on Playstations. Getting more out of their investments is a bonus, relatively But yeah, it's the smarter way to do it, and the way they should do it.

 

[Carl B]

Expletive, I think you're underestimating Cell's abilities as a media chip. The fact that it's fully programable just takes it one step further as being able to fill the roles of several chips that before may have taken several different specialized processors.

Like Shifty said, since Cell is - supposedly - here for the long haul, what you create is a base architecture with an ever expanding knowledge base familiar with it's utilization, such that many years from now Cell may have become the gold standard via which all other media chips are judged.

Consider also that Cell had the R&D opportunity that few other chips enjoy. It was created jointly by three companies that each have something to gain from it, there is a minimum volume that can be expected due to it's placement in PS3 (which will improve familiarity with the manufacture of the chips and grant economies of scale right off the bat), and the size of the companies allowed for a large expenditure to reach their objective.

For one of the smaller firms that specialize in DSP's to undertake that sort of effort would likely burden them beyond what they could endure, and the fact that any sort of massive effort of the sort would have no guaranteed ROI would likely leave them reticent to attempt it.

I think Cell will fill it's role well - the only question is whether it's cost per chip will be worth it compared to cheaper alternatives. I imagine though that it will, and of course Sony and Toshiba hope the costs per chip will fall significantly as time goes on, to say nothing of the comments in above posts on yields and reduced SPE-counts.

 

[MfA]

They can certainly create software that will perform the function but they count themselves out of the 'best of breed' solutions by doing this, negating other competitive advantages.

You might have a point here if the big players weren't already using their own motion compensated deinterlacing, which at the moment represents the most complex algorithm in TVs (the next big thing is multi-frame super-resolution).

 

[expletive]

You might have a point here if the big players weren't already using their own motion compensated deinterlacing, which at the moment represents the most complex algorithm in TVs (the next big thing is multi-frame super-resolution).

So youre saying that companies like Samsung, Panasonic, Mitsubish are all using proprietary scaling/deinterlacing solutions on the hardware and software side?

J

 

[Phil]

I'd like to add to xbdestroya's post that CELL will enable Sony as a company to so much more than what those cheaper "dedicated" chips are doing now. There's a shift occuring, the shift that information is going to be sold via broadband networks - and with that, you have much more possibilities to market your product - AND - offer more.

Sony believes there will be a shift in content distribution - CELL will play a key point in tihs.

 

[expletive]

I'd like to add to xbdestroya's post that CELL will enable Sony as a company to so much more than what those cheaper "dedicated" chips are doing now. There's a shift occuring, the shift that information is going to be sold via broadband networks - and with that, you have much more possibilities to market your product - AND - offer more.

Sony believes there will be a shift in content distribution - CELL will play a key point in tihs.

Ive heard this before Phil but have yet to hear a real example of what in the world theyre talking about. I'm not saying there isnt one, just that no one has been able to produce one for me yet.

Shift in content distribution? Meaning what?

My feeling is that if there is a consumer need, or desire, there will be a cheaper 'dedicated' chip to meet it.

J

 

[Shifty Geezer]

Here's some examples of where I think Cell could/might end up.

1) PS3. A lot of media work will happen here. A lot of algorithms will be broken in with this flagship developemnt environment

2) Professional Media equipment. Sony are big suppliers of professional AV equipment. Cameras, TVs, and all sorts of gear. Cell is ideally suited to realtime special effects which'll speed up development and cost a lot less then alternative multiprocessor solutions.

3) TVs. TV's no longer just take an image and show it. They can do all sorts of tidy-up, especially important working on compressed signals. There'll be different source resolutions that'll need scaling to the screen's native resolution, and effective scaling algorithms can be very demanding yet ideally suited to Cell it would seem. There's also audio decoding to be done, and a general purpose solution like Cell could be software upgraded to new standards where a hardware solution cannot.

4) Audio work. Software synthesizers are another area Cell would be fantastic at, and Cell would offer a cheap and very powerful synthesizer that'll perform at least as well as any custom hardware solution. A to that digital sound processing and you'll have a one chip solution that can handle any effects you throw at it. Rather then needing several boxes each with custom hardware a one box solution would be cheaper and probably better too.

A key point of the above is that developments in one area can benefit another. Audio effects running on a Cell Audio Mixer can be incorporated into your TV's audio decoder and used in PS3 games. Image processing like background removal will be developed and used in EyeToy and professional video equipment where blue-screening can be replaced with easier-to-work-with props and markers. The idea is that processes can be copied and dropped into other appliances. At the moment if someone develops an audio filtering algorithm, it needs to be rewritten for a dozen different processors to use. A single uniform hardware platform elliminates that, focusses on maximum production enhancement as you are only producing one line of chips, not dozens, and provides developers with a predictable future hardware base. If you can write for Cell in PS3, you can write for Cell in a mobile phone, Cell in a TV, Cell in an audio mixer. Instead of umpteen segregated hardware industries all developing individually, all those advances could be concentrated onto one platform.

Certainly I rate it a good idea with some very long term vision behind it.

 

[Phil]

Expletive,

Ive heard this before Phil but have yet to hear a real example of what in the world theyre talking about. I'm not saying there isnt one, just that no one has been able to produce one for me yet.

Shift in content distribution? Meaning what?

This was actually a very heated topics back around 2, 3 years ago and I remember quite well how a member here described the fundemental shift in consumer demands the best. Since it's so much better worded, let me allow to quote some of the relevant points our member vince pointed out (I only found these but I might find some more at a later time, but this should definetly get you started):


--------------------------

http://www.beyond3d.com/forum/showpost.php?p=32735&postcount=47

Sony's already been rearanging itself since their Sony Meeting in Summer 2001 where they disguessed the new corperate strategy under Ando:

"Using graphs and diagrams to drive home his point, Ando insisted that Sony, the world's premier gizmo maker, was better positioned to triumph in this broadband world than any of its rivals. He went on to argue that nobody--not Samsung (SSNLF ), not Microsoft (MSFT )--had a sharper vision of how consumers would navigate superfast networks in which a single fat wire, or a sliver of radio frequency, would handle multiple layers of voice, data, and video. Already, he noted, Sony's violet-gray Vaio laptops were hits with the digerati, who liked to edit their own photos and music files and exchange them over the Web. Once broadband networks were ubiquitous, all of Sony's cameras and audio devices would meld into a seamless distribution network for Sony's movies, music, and games, supported by the company's own online shopping and financial services. "Ando's message was clear and aggressive," says Hiro Uchida, Sony's general manager for strategic ventures. "Sony faces big challenges, but Ando showed us that we're getting activated."

You know, if you guys actually did some research and used some intelligence, they're would be no fighting here....

I won't even talk of Conrgressional oversight when talking about an alliance of rivals of Microsoft who are supporting Open Source projects like Linux and OGSA.

http://www.beyond3d.com/forum/showthread.php?p=58515&highlight=digital#post58515

Thats definatly a cool aspect. The whole idea behind using a cellular architecture was an excellent choice IMHO. I've talked about it before and people just shruged me off, but starting in 2004/2005 Sony and perhaps even Toshiba and Matsushita, will begin making products based around Cell and the era of truely pervasive computing will begin.

One of my favorite books is Visions by the string physicist Michio Kaku, and was taken back at his predictions of the future and just how prevelent computing will be. Fast forward 6 years and I'm sitting here, seeing the beginning of this forming - and at a pace thats far outstripped what he thought was even possible.

Thats what I see as the true revolution, that a single ISA/Core, ect will flawlessly be the backbone of a networking faric that connects our PDAs, PCs, TVs, Cell Phones, and can distribute digital content flawlessly. The limit for this type of thing is only the human imagination... and Cell is barely a first step, but it's a start.

IMHO, thats whats impressive. I'd much rather buy a Sony PS3 thats less powerful than a Xbox Next or Cube2, but allows me to seemlessly interact with all my other little gadgets and get digital content (movies, music, ect).

This is important because I beleive, as Kutaragi has stated and Kaku has written about, that the future will be one of pervasive computing. Where almost every item we use will in someway be networked together and contain low-cost processing elements. It seems alien today, but all truely forward looking staements do. The problem, as Kutaragi talked about in that one interview, is that the internet of today is more like a bunch of stand alone islands with their own ISA, OS, procesing elements, ect. Cell, atleast for Sony and anyone who adopts this, will eliminate those burdens - without software thats costly in it's incompatability problems and/or preformance sacrifices.

So, what’s going to happen is - Sony will start releasing products with Cell and a way to communicate; Wireless, Broadband, et al. People will buy them. People will want other electronic products. People and Salesman says, "Wow, If I buy the Sony product, it'll work with my other one and allow me to do this.. Or that... Easily." People buy more Sony products. Sony makes more money. Other companies want to make more money. Other companies see Sony Advantage (eg. Trinitron). Other companies license Cell from Sony. Sony makes more money.

http://www.beyond3d.com/forum/showpost.php?p=92296&postcount=334

As many here have been saying for damn-near 2 years now - Cell is an architecture designed under the aegis of Kutaragi that will allow for a synergy of Sony's future products in a broadband world where pervasive computing is commonplace. Go attand any lecture at any top-10 university in the United States on computing and your bound to hear of a form of pervasive computing and the necessary requirments.

Cell, in one incarnation or another, will be the base of allmost all future Sony products and will allow threw wireless, power and/or broadband for the Sony Group to controll the entire Digitial Media and Content pipeline. By seemlessly allowing all their devices to interact threw one architecture, one ISA - they simplify grately the interconnectivity of they're devices. They day is approaching where you'll buy a PS3 powered by the Cell architecture, and download Digital Content (eg. Spiderman2) created in a Sony Cinema Production House or Music (eg. Pete Yorn's latest album) and using some sort of DRM software like Sony's OpenMG save it to a Blue-Ray disc and/or send it to your Cell powered Clie, or your Cell powered 50" WEGA Plasma, or your Cell powered Sony Shelf-Audio system to play... all with one or two clicks on a remote.

Or take what you can already do using the Infared Port/MagicGate to the next level and take your Digital Pictures and/or Video on a Sony device of your ass pounding down the Tequilla and licking your friend's "chest" clean of salt and lime juice and send it anywhere... either saved on a R-Media or send it threw the 'net - just without the typical PC problems inherient to the very nature of a PC, OS and HAL.

To say that PS3 doesn't fit into this scheme - when the Cell project was initiated under the guise of SCE to power PS3 and only in 2001 when the project was sufficiently proven did Sony Group sign on - is insane. This man's an idiot, he deserves to be chained to a chair and put in a room with Chappers.

I think it was Ken Kutaragi himself who said that Sony to survive in the 21st century needs to differentiate itself from its competition - otherwise it will become another Toshiba or Mashushita that's profits are influenced by the endless productivity wars because they use 3rd party chips. The man's not as stupid as many here make him out to be... I'd say he borders on the brilliant, not "Witten" brilliant, but above most.

Personally, I think this last quote answers your question best - the key words being digital content distribution and pervasive computing.

CELL is a key point in this strategy - a strategy that can combine all CELL powered products and allow for immense content distribution. Sony owns the ISA (CELL) while they also have a large stake in the backend (Sony Pictures, Sony Music, PlayStation). This could mean Sony could control the flow of information in one go -> from the studios right directly into the livingroom through their products - all by the touch of a button for the customer.

 

[dskneo]

The PlayStation division had been losing the company money in recent quarters.

actualy, at the end of the 4 quarters SOny is winning money, not losing.

they are losing money in the first quarters because of the next gen system stuff. (it happen the same in 1999/2000), by christimas time they win all the money back.

i'm not gonna even bother to search again and post here the figures. go google.

sony its not losing money. Quarters are just that, quarters.

 

[Shifty Geezer]

Tye've been taking record profits from hardware and software sales last I read, but not making as much net operational profit because they keep spending all of it!