Toshiba, Sony close to 65nm sample production

MrSingh said:
2006 earliest. Place your bets everyone!


My bet.

Sony delays the PSP until .65 nm is ready. Additional edram will be the only major upgrade. They'll go with an early to middle of the year launch in 2005.

Then in 2006 Sony prepares to release the PS3 with a retail price tag of $399.
 
Brimstone said:
MrSingh said:
2006 earliest. Place your bets everyone!


My bet.

Sony delays the PSP until .65 nm is ready. Additional edram will be the only major upgrade. They'll go with an early to middle of the year launch in 2005.

Then in 2006 Sony prepares to release the PS3 with a retail price tag of $399.


Interesting! PSP at .65...mmmm...
Gosh... 2006 seems so far away thought.. It's not even 2004 now.. That's 2 years!! That's a tenth of my life so far! A 50th of my expected life duration!!
 
Fafalada said:
It's not even 2004 now..
My bet, you're going to wait 2 years(or more) for Any of the new consoles (Not counting anything bizarre like a surprise Nuon2 release :p).

I can see nintendo launching in 2005 for x mass with ms launching around the ps3 .
 
london-boy said:
Brimstone said:
MrSingh said:
2006 earliest. Place your bets everyone!


My bet.

Sony delays the PSP until .65 nm is ready. Additional edram will be the only major upgrade. They'll go with an early to middle of the year launch in 2005.

Then in 2006 Sony prepares to release the PS3 with a retail price tag of $399.


Interesting! PSP at .65...mmmm...
Gosh... 2006 seems so far away thought.. It's not even 2004 now.. That's 2 years!! That's a tenth of my life so far! A 50th of my expected life duration!!


If Sony decided to wait for .65 nm, how much more edram could they include before the die would become too big? Also wouldn't it sort of make sense for them to do something like this? The Sony fans here have brought up many times how Sony/Toshiba has the smallest edram cell size at .65nm.

This current console generation isn't even close in terms of installed base. Sony is so far ahead, they have no pressure to release another console. They could concentrate on the PSP and wait for .45nm to arrive around 2007. A PS3 at that time could take advantage of the latest edram technologies from Toshiba. SOI edram without a capacitor. Panajev posted about this a while back.



http://www.beyond3d.com/forum/viewtopic.php?t=7773&postdays=0&postorder=asc&start=0


I doubt the PS3 will come out anytime soon.
 
Brimstone said:
If Sony decided to wait for .65 nm, how much more edram could they include before the die would become too big? Also wouldn't it sort of make sense for them to do something like this? The Sony fans here have brought up many times how Sony/Toshiba has the smallest edram cell size at .65nm.

This current console generation isn't even close in terms of installed base. Sony is so far ahead, they have no pressure to release another console. They could concentrate on the PSP and wait for .45nm to arrive around 2007. A PS3 at that time could take advantage of the latest edram technologies from Toshiba. SOI edram without a capacitor. Panajev posted about this a while back.



http://www.beyond3d.com/forum/viewtopic.php?t=7773&postdays=0&postorder=asc&start=0


I doubt the PS3 will come out anytime soon.


Anything is possible, and i agree that Sony are in the position to wait and wait, but only until MS and N start doing something right.... PS2 will outsell next gen consoles in the beginning just like PS1 was outselling all new consoles put together in the first year. they can't keep selling ps2's while MS and N have both next gen consoles out, and hoping to remain at the top... 2007 is pushing it IMO
 
london-boy said:
Brimstone said:
If Sony decided to wait for .65 nm, how much more edram could they include before the die would become too big? Also wouldn't it sort of make sense for them to do something like this? The Sony fans here have brought up many times how Sony/Toshiba has the smallest edram cell size at .65nm.

This current console generation isn't even close in terms of installed base. Sony is so far ahead, they have no pressure to release another console. They could concentrate on the PSP and wait for .45nm to arrive around 2007. A PS3 at that time could take advantage of the latest edram technologies from Toshiba. SOI edram without a capacitor. Panajev posted about this a while back.



http://www.beyond3d.com/forum/viewtopic.php?t=7773&postdays=0&postorder=asc&start=0


I doubt the PS3 will come out anytime soon.


Anything is possible, and i agree that Sony are in the position to wait and wait, but only until MS and N start doing something right.... PS2 will outsell next gen consoles in the beginning just like PS1 was outselling all new consoles put together in the first year. they can't keep selling ps2's while MS and N have both next gen consoles out, and hoping to remain at the top... 2007 is pushing it IMO

It depends on what Microsoft does. If the Xbox 2 comes out in late 2005 that might be pushing it. If the Xbox 2 came out in 2006 it would be a scenario like the Dreamcast. Sony follows up a year later with an avalanche of hype in 2007. If Sony is really serious about Blu-ray, then 2007 dovetails price wise with that launch window. The costs of Blu-ray should be more feasible by 2007.
 
Brimstone said:
It depends on what Microsoft does. If the Xbox 2 comes out in late 2005 that might be pushing it. If the Xbox 2 came out in 2006 it would be a scenario like the Dreamcast. Sony follows up a year later with an avalanche of hype in 2007. If Sony is really serious about Blu-ray, then 2007 dovetails price wise with that launch window. The costs of Blu-ray should be more feasible by 2007.


Oh Lord... Please, Sony don't let me wait another 3 flippin years!!! Give me a starter, a PS2.75, i'll buy it! I'll buy 2 in fact... Just gimme something!! :D

Really, i agree a 2007 PS3 would be much better than a 2005 PS3, of course it would be, but come one that would give PS2 a 7-8 years life.... PS2 is already old-ish (safe from some neat tricks and WOWs here and there), and although it apparently still has some juice left, there's only so much they can do with the hardware, i can only imagine what it will be like to play ps2 games in 2006...
 
They could make PS2 as cheap as hell (first to $99, then to $50 some), and with additional support of developers + PSP link-up options they *could* be able to hang in there for 3 more years and release PS3 with .45nm

But, would Sony be able to catch MS & Nintendo saleswise?
 
I'd still like a 2004-2005Big N launch, followed by a 2005-2006 sony launch, followed by a 2006-2007 MS launch... that would be ideal...

I know my wallet would agree.


It would be pretty interesting to see what would happen to the market if every 2 years one of the 3 console manufactureres released another console (vs all trying to launch at the same time.. or a year apart.. like if Nintendo had launched now with something lots neater than PS2 or Xbox for example). It would add another dynamic to the market.. could perhaps be a good strategy for the underdog.


But, would Sony be able to catch MS & Nintendo saleswise?

I think they'd catch up in a matter of months, given no hardware shortages.
 
I don't think any system will launch after 2006.

It is generally agreed that Sony can play the waiting game longer, and can reap as much profits as possible this gen to cover CELL investments. However, there is definitely a deadline as to when the x billion poured into CELL must start seeing returns. The new fabs that are being contructed should not have idle capacity upon completion.

It's not just a matter of being able to wait longer or not. It's about $$$(isn't it always?). The ghosts of the x billion invested are looming. The new fab will eventually be completed. Red numbers on financial reports will get bigger. Returns to the mammoth sums invested must start sometime, and I don't think Sony can hold out till 2007.

This places launch in 2006. Possibly late 2005 for Japan.

I don't think MS or N will want to launch too late after Sony, after the lessons of this generation. This places them in the 2005/2006 frame as well.

An exception to this opinion is of course, Europe. Which will probably launch later, as usual. The idea of a worldwide launch is unlikely. Just pause and think about the number of units required, money and time required to produce those numbers and NOT sell any of it in the meantime to cover costs, plus growing inventory costs, etc.
 
Oh... Missed that!

Well, that's not the only thing you missed. :LOL:

Really that's not the only bit that suggested it. Even the original patent suggested it.

Also, a number of PEs, such as PE 201, may be joined or packaged together to provide enhanced processing power.

For example, as shown in FIG. 3, four PEs may be packaged or joined together, e.g., within one or more chip packages, to form a single processor for a member of network 104. This configuration is designated a broadband engine (BE).

BTW I just look up the patent for this quote, instead of using the published patent application, I note this in the reference section.

Mamoru Maekawa, Bunsan Operating System (Distributed Operating System), Tokyo: Kyoritsu Shuppan Co., Ltd., Dec. 25, 1991, pp. 175-182, ISBN 4-320-02570-9.
IBM System/370 Kakucho Architecture (SA22-7085-0 IBM System/370 Extended Architecture Principles of Operation, IBM Japan, Apr. 1984, pp. 3-8 to 3-10.
"IBM Wins Playstation 3 Contract," BBC News, Mar. 12, 2001.

Edit: For compleness, here is the link.

http://news.bbc.co.uk/1/hi/business/1216551.stm

Of course, none that we don't know already.
 
nondescript said:
I think that risk production starting next year (for system integration and SDK) and full production in Q3 2005 means that Sony has the capability to go for a late 2005 launch.

Good call. :)

As for my bet:
  • 2H2004 PSP launch.
  • 2H2005 PS3 launch.
To anyone who says anything else (2007, Whoa..), I think you need to clear your mind and look at this again. That would mean they are going to produce PS3's at two fab's for a year (2H2005 to 2H2006) to launch in the US. Yet, mysteriously, Microsoft ramped the XBox in 3 months, heh. In 12 months, you basically lose 100% of your process advantage (IIRC - maybe 18months) for a set-piece IC.
 
Not to mention PS3's memory(Rambus XDR) is going to start production late 2004 and ramp up to full speed in early 2005. Please noone tell me they are going to sit on millions of memory chips for 2 years ;)
 
Here is my bet: Xbox2 and PS3 will launch and be available is some quantities in time for holiday season 2005.
 
passerby said:
I don't think any system will launch after 2006.

It is generally agreed that Sony can play the waiting game longer, and can reap as much profits as possible this gen to cover CELL investments. However, there is definitely a deadline as to when the x billion poured into CELL must start seeing returns. The new fabs that are being contructed should not have idle capacity upon completion.

It's not just a matter of being able to wait longer or not. It's about $$$(isn't it always?). The ghosts of the x billion invested are looming. The new fab will eventually be completed. Red numbers on financial reports will get bigger. Returns to the mammoth sums invested must start sometime, and I don't think Sony can hold out till 2007.

This places launch in 2006. Possibly late 2005 for Japan.

I don't think MS or N will want to launch too late after Sony, after the lessons of this generation. This places them in the 2005/2006 frame as well.

An exception to this opinion is of course, Europe. Which will probably launch later, as usual. The idea of a worldwide launch is unlikely. Just pause and think about the number of units required, money and time required to produce those numbers and NOT sell any of it in the meantime to cover costs, plus growing inventory costs, etc.

CELL will see its market introduction before PlayStation 3, be it in the next Wega Engine or what.

The market is presssuring Sony to show CELL as investors want to see where all the money is going into and they will get a taste next year.

Right now, all the Semiconductor R&D and the fabs that are in construction have one goal, now that Kutaragi's SSNC has the complete control over the company's Semiconductor R&D, and this is to increase the current profit margins of Sony as much as they can.

PSP will see a die shrink to 65 nm ( not too far from its Q4 2004 launch IMHO ) and it is very likely that we will see an EE+GS@65 nm as I think this is what they are sampling now for their 65 nm manufacturing process.

With the sales of PlayStation 2 growing for the Christmas season and the big AAA software releases in 2004 that should assure good sales of PlayStation 2 Hardware, the EE+GS@90 nm should be a life-saver as it should allow them to leave the tag price at $199 and increase a lot their profit margins, at least till next E3 if not later and anyways they would still make profits ith a moderate price cut as the savings given by the EE+GS@90 nm + the new Dragon I/O chip should be quite nice.

Sony corp. needs its cash cow to rank in as much of a profit as they can and will fight to make sure this happens and they are doing good IMHO.
 
The Playstation 2 is out pacing both the Gamecube and Xbox in terms of sales by a wide margin. The PS2 hasn't even dropped down to $99 yet and still strong consumer demand exists. Plenty of profit is to be had from the PS2 platform, why launch a PS3 or even talk about it?

Having the PSP become a big hit has to be the primary goal now for Sony. Nintendo is going to counter attack at some point, so having a strong focus on the PSP is critical. Nintendo rakes in a lot of profit from Gameboy sales, and the PSP can certainly make some serious profit for Sony.

Sony fabs don't need a PS3 to get maximum return on the investment. Shrinking the PS2 chips to a .65 nm process and producing PSP chips on .65 nm should be plenty of work for the fabs. The investment in CELL itself doesn't come across as that large. $400 million in R&D (or whatever the exact dollar amount) isn't back breaking. An extra 18 months of R&D on the PS3 would really be helpful to help the console live up to the hype it will generate.

We shall see what happens, but right now I think Panajevs observations on .45 nm SOI edram are signifigant.
 
Brimstone said:
The Playstation 2 is out pacing both the Gamecube and Xbox in terms of sales by a wide margin. The PS2 hasn't even dropped down to $99 yet and still strong consumer demand exists. Plenty of profit is to be had from the PS2 platform, why launch a PS3 or even talk about it?

Having the PSP become a big hit has to be the primary goal now for Sony. Nintendo is going to counter attack at some point, so having a strong focus on the PSP is critical. Nintendo rakes in a lot of profit from Gameboy sales, and the PSP can certainly make some serious profit for Sony.

Sony fabs don't need a PS3 to get maximum return on the investment. Shrinking the PS2 chips to a .65 nm process and producing PSP chips on .65 nm should be plenty of work for the fabs. The investment in CELL itself doesn't come across as that large. $400 million in R&D (or whatever the exact dollar amount) isn't back breaking. An extra 18 months of R&D on the PS3 would really be helpful to help the console live up to the hype it will generate.

We shall see what happens, but right now I think Panajevs observations on .45 nm SOI edram are signifigant.

I agree with you, but the CELL investment is bigger than what you think.

I think you missed the $2+ Billions bond they just took for Semiconductor R&D... CELL was one of the main items justifying it and previously they announced that over the next 3 years they were going to spend $4+ Billions on CELL R&D.

CELL will not be used only in PlayStation 3 and the R&D done for it can help cut saving costs in other products offered by Sony: Ken Kutaragi and his team members nhave been plannign the CELL strategy quite well IMHO.
 
The market is presssuring Sony to show CELL as investors want to see where all the money is going into and they will get a taste next year.

Didn't Kutaragi promised to demo cell by the end of March next year ?

I wonder, if he'll keeps his end of the bargain.
 
Brimstone said:
Sony fabs don't need a PS3 to get maximum return on the investment. Shrinking the PS2 chips to a .65 nm process and producing PSP chips on .65 nm should be plenty of work for the fabs. The investment in CELL itself doesn't come across as that large. $400 million in R&D (or whatever the exact dollar amount) isn't back breaking. An extra 18 months of R&D on the PS3 would really be helpful to help the console live up to the hype it will generate.

Panajev2001a said:
I agree with you, but the CELL investment is bigger than what you think.

I think you missed the $2+ Billions bond they just took for Semiconductor R&D... CELL was one of the main items justifying it and previously they announced that over the next 3 years they were going to spend $4+ Billions on CELL R&D.

Not only that, but there is (or should be) a real urgency to get CELL out as quickly as possible. Fabs are horrendously expensive, and since they depreciate at about 20% a year, you need to put out a lot of chips just to break even. A Goldman Sachs estimate put the revenue required to support a cutting edge 300mm fab at 3 billion USD/year. Sony can't let those fabs idle.

Also, with each passing year, the "traditional" PC architecture continues to ride up Moore's law, so the performance advantage of ASICs over PC will degrade. Since CELL is taped out, its logic is locked down - can't afford to sit on your hands while traditional GPU's improve.

So here's how its gonna happen (IMHO) - they get those fabs up as soon as they can (mid-2004 risk production, mid-2005 mass-production). Acculumlate enough for a launch, and launch. Sony is definitely not waiting for PS2 to wind down first - they're going to launch as soon they can. (They even might have to launch sooner than they would like to, if MS forces their hand)

I don't think it will be that hard to ramp up PS3 production - with limited amounts of chips rolling out in mid-2004, they can design the system and hand it off to their component suppliers upstream, and the assembly people downstream. The current PS2 supply-chain handles 2 million units/month. Sony could use the same network for PS3. It all depends on how fast Sony/Toshiba can get the yields up. The bottleneck is IC production.
 
Back
Top