Cell mass production plan for 2nd half of 2005

[PiNkY]

AFAIK it is not unusual to have fabs produce sample parts for a whole year until actual production start.

Edit: e.g. Intel demoed a working 7Ghz TCP-offload Risc ic @ 90nm node at the beginning of this year ( www.cs.utexas.edu/users/cart/trips/ publicity/UT_Austin_Rprnt.pdf ), with other sample products well before that.

 

[jvd]

I think we will have a summer 2006 launch in japan and a x mass 2006 launch in the usa and othe markets. I'm sure sony wants to have enough systems to meet demands as i'm sure they can sell 5 million consoles easy for xmass in the usa . Yes systems sell the best a few years down the road but its allways nice to have enough to meet demand when your product is the new kid on the block and the it thing to own.

 

[zidane1strife]

Was EE being 'mass produced' before the PS2 started selling, or was it being produced in a more moderate amounts at first, just to meet the smaller launch shipments?

No, it was mass produced from the beginning. If it is not mass produced you will only get a handful of chips

They only managed to accumulate about 980K+ units for their launch in japan, that was a year after they announced the system, and about a year after dev. units arrived(in summer of 1999, IIRC).

From mid-early march launch to october 26(IIRC), they only managed to accumulate 500K for the US launch. That is about 7-8 months, but true they also kept sendind some to japan.

Not sure, if that is considered mass production.... but I just hope they launch in 2005, at least in japan, which I'd say shouldn't be out of the question, if we look back and remember that the ps2 was meant for a late 1999 launch, but had to be delayed till march.... with mass production in fall 2005, all they need is to send out dev. units, which I still hope occurs in 2004.

ed

 

[Paul]

The sheer ignorance of many people shine in this thread.

The Sony plants going online in 2005 to mass produce Cell isn't new news, never was.

Meanwhile the TOSHIBA plant will be producing for the imminent PS3 Jap launch. With later the Sony coming in to help with the AMERICAN AND OTHER LAUNCHES.

Everyone knew this.

So what happened to those lawnchair experts in this forum that said CELL manufacturing would start 2004 for PS3 launch in 2005??? Seems like CELL and the rush to the holy grail of 65nm got pushed back after all.

....? The Toshiba plant mass producing Cell on 65 is going to be completed january. Mass production will start late summer 2004 from that plant. Toshiba says what it produces will be "used in SCEI's products"

I wonder what Panajev and Paul has to say...

We were right all along, and still are to this day. Right about Toshiba mass producing in 2004, and were right about the Sony fabs coming online in 2005 to help.

here were a couple of people here, I won't mention any names but they know who they are, who kept INSISTING that 65nm chip manufacturing would commence 1H 2004

It is. Nothing has changed for the Oita plant.

So now that it has been confirmed that mass production of 65nm chips won't start until 2H of 2005, can those unamed individuals please tell me what 65nm chips will be manufactured 1H 2004 and mass produced 2H 2004???

.....

Cell, amoung other 65 nm LSI.


PS3 will launch Fall 2005 in the US, quote this. Because In two years I'll be right.

 

[jvd]

Meanwhile the TOSHIBA plant will be producing for the imminent PS3 Jap launch. With later the Sony coming in to help with the AMERICAN AND OTHER LAUNCHES

Paul can you give me a link to where it says the toshiba plants will be making the chips the ps3 .

Also its end of fiscal year which goes past febuary if i'm not mistaken. At least if we are talking the same press releases that panja used to link too

 

[cthellis42]

It's been around: http://www.gamespot.com/ps2/news/news_6025378.html

In related news, Toshiba is also scheduled to build a new production line in its OTSS plant, which will produce 300mm wafers with 65-nanometer process. The total cost of the plant will be 200 billion yen, and Toshiba will invest 40 billion yen this year. The plant, which will start production in late April of 2004, will manufacture chips for Sony Computer Entertainment and other companies. Sony is planned to invest in Toshiba's plant as well. Toshiba said that it hopes to switch the production line to build chips with 45-nanometer process in the future.

One assumes if they're building anything CELL-wise for SCE, in the beginning it's only going one place.

 

[Panajev2001a]

Even as development work continues on the current generation of cutting-edge semiconductor chips, Sony's Manabe said the company plans to start installing equipment at a new plant that is creating a next-generation microprocessor code-named "cell".

"We expect to start mass production of 'cell' in the second-half of 2005," Manabe said. "For the time being, we plan on setting up a pilot line."

Sony plans to invest 500 billion yen over the next three years in semiconductors, including 200 billion yen for "cell", which will initially use 300 millimetre wafers and 65-nanometre circuitry.

Analysts expect the chip to power Sony's next-generation game console, but the company aims to make "cell" the global standard for consumer electronics in the high-speed Internet era.

The next-generation microprocessor is being developed with Toshiba Corp 6502.T and IBM IBM.N .

By midday, shares in Sony were down 2.51 percent at 3,890 yen, versus a 1.07 percent decline in the Nikkei 225 average.

jvd, never sell the skin of the bear before you made sure it is dead.

I am sure you know the difference between volume mass production and risk production ( pilot lines ) right ?

I will give you a tip, I think the 2005 date given is for the beginning of volume ( yelds sort of normalized ) mass-production.

You also know the difference between Oita #2 and Nagasaki #2 ? Two different plants which will be completed at different times.

How long I have been saying that mostly for economical reasons ( chip size being large at 65 nm ) they would try to push PlayStation 3's launch around Q4 2005 at least ?

The plans with the Oita #2 fab have not changed.

Details on CMOS5 ( their 65 nm process ) have not changed either: it is still one of the leading edge processes in the industry ( NEC's e-DRAM and SRAM cells are like 36% bigger in NEC's 65 nm manufacturing process ).

 

[Vince]

So what happened to those lawnchair experts in this forum that said CELL manufacturing would start 2004 for PS3 launch in 2005??? Seems like CELL and the rush to the holy grail of 65nm got pushed back after all.

As it's been discussed more times than I can remember, Cell is an architecture. It will have many incarnations that span Sony and Toshiba's electronics divisions. There are many manifestations that it will take, similar to how x86 or PPC is used all over the marketplace, across all price-points.

So, by this very statement we logically can see that it's possible for Cell production to begin in 2H 2004 (perhaps 90nm out of E. Fishkill) with the PS3 varient being produced for a 2005 release when the lithography allows for it's viability - eg. 65nm that co-insides with SCEI Nagasaki, OTSS expansion coming online, as well as XDR shipments.

I won't mention any names but they know who they are, who kept INSISTING that 65nm chip manufacturing would commence 1H 2004 and how SONY/Toshiba were ahead of everyone with regards to lithography. So now that it has been confirmed that mass production of 65nm chips won't start until 2H of 2005, can those unamed individuals please tell me what 65nm chips will be manufactured 1H 2004 and mass produced 2H 2004??? Was it the EE3 and GS3???

You do realize there is a difference between manufacturing and mass production. The very fact that you've stated them twice, as seperate terms, only illuminates the fact that you're just trolling instead of trying to see how this could work. As the article said, there will be a pilot line set up in the interm. Lets see what K. Kutaragi shows off before talking.

So, in conclusion - don't critisize said "lawnchair experts" when you, yourself, don't know what the f- is going on.


IMHO, it's looking ever more likely that PS3 will see a simultaneous launch in late 2005. I'll even go out on a limb and say that I'm around 95% sure of this now - Yeah, sounds about right.

PS. As for chap, can we get some effecive moderation? My Gawd... he's like a retarded 10year old on ritalin.

 

[Vince]

AFAIK it is not unusual to have fabs produce sample parts for a whole year until actual production start.

Exactly, and AFAIK that's what distinguishes what comes off a "pilot line" or borders on "risk production" and what constitutes "mass production."

Btw, during the numerous process technology discussions on this forum, it was more then once stated (if i remember correctly) that the unified 90nm EE+GS chip was already utilized in japanense PS2s. Now, it seems that Sony's move to 90nm production is yet to come (as of the 8th of octobre), quite in line with the rest of the industry.

I have stated this, and I thought the 50K series had it. Which would have coincided with the statements of a late spring production run. If I was mistaken, I am sorry.

Although, with the "competition" (eg. nVidia and ATI) on 130nm, the latter just moving to it, and the next move that won't happen untill 1H 2004 said to be to the obscure 110nm node - I'd say producing a 90nm IC with eDRAM is ahead, by a bit.

 

[nonamer]

Let's have a bet/poll. Say whether you think the PS3 will be out in 2005 or 2006. I say 2006.

Also, initial cost for the PS3 will be . Here's some cost estimates for the P4EE: http://www.amdzone.com/forums/messageview.cfm?catid=26&threadid=69639&STARTPAGE=2

Remember, this is a ~ 230 mm^2 die chip on bulk silicon on mature 130nm, and cost is about $90 at 100% yields. The PS3 will be two ~ 300 mm^2 chips on SOI at a relatively new 65nm process. Depending on yields, I would dare say that the cost of the PS3 will be $250 - 300 initially for just the CPU and GPU alone. Astoundingly expensive and in fact I don't think Sony will dare launch in 2005. Mid 2006 at the earliest. Possibly even late 2006.

 

[Megadrive1988]

early-to-mid 2006 in Japan. fall 2006 USA/Europe

 

[jvd]

Even as development work continues on the current generation of cutting-edge semiconductor chips, Sony's Manabe said the company plans to start installing equipment at a new plant that is creating a next-generation microprocessor code-named "cell".

"We expect to start mass production of 'cell' in the second-half of 2005," Manabe said. "For the time being, we plan on setting up a pilot line."

Sony plans to invest 500 billion yen over the next three years in semiconductors, including 200 billion yen for "cell", which will initially use 300 millimetre wafers and 65-nanometre circuitry.

Analysts expect the chip to power Sony's next-generation game console, but the company aims to make "cell" the global standard for consumer electronics in the high-speed Internet era.

The next-generation microprocessor is being developed with Toshiba Corp 6502.T and IBM IBM.N .

By midday, shares in Sony were down 2.51 percent at 3,890 yen, versus a 1.07 percent decline in the Nikkei 225 average.

jvd, never sell the skin of the bear before you made sure it is dead.

I am sure you know the difference between volume mass production and risk production ( pilot lines ) right ?

I will give you a tip, I think the 2005 date given is for the beginning of volume ( yelds sort of normalized ) mass-production.

You also know the difference between Oita #2 and Nagasaki #2 ? Two different plants which will be completed at different times.

How long I have been saying that mostly for economical reasons ( chip size being large at 65 nm ) they would try to push PlayStation 3's launch around Q4 2005 at least ?

The plans with the Oita #2 fab have not changed.

Details on CMOS5 ( their 65 nm process ) have not changed either: it is still one of the leading edge processes in the industry ( NEC's e-DRAM and SRAM cells are like 36% bigger in NEC's 65 nm manufacturing process ).

Yes panaj I have payed attention to your posts. I just have never heard anything offical that toshiba will be making the chips for the ps3 and untill I do it doesn't matter when the tosiba plant will come about. For all we know it will just be making psp chips .

 

[Johnny Awesome]

Agreed. It's foolish to think otherwise.

 

[Paul]

So Toshiba is going to be making Cell chips based on 65 nm tech which will be used in a PSP? Whoa! This is amazing.

Get real.

 

[jvd]

So Toshiba is going to be making Cell chips based on 65 nm tech which will be used in a PSP? Whoa! This is amazing.

Get real.

Paul i have yet to see a quote stating which chips toshiba will be making just that they are making chips for sony. I asked you for an offical anouncment on which chips toshiba was making in that plant and for what company. You reply with this .

 

[TheMightyPuck]

Fall 2006. Sony is so far ahead of the game there is no reason to go any earlier.

 

[Vince]

Hey, guys, what's the excuse going to be "if" Paul is proven right? I'm just curious. Will you admit to lacking the perception and sence Paul has or just let it slide? Or will you stumble into a DeammeatGA type condition and suffer from Dissocial Personality Disorder; the entire time claiming, "Ken Kutaragi changed the specs and fabs 'cause he wanted Paul to be right."

Because, objectivly Paul basically has every precedent and logical extrapolation from public (and not) info on his side.

 

[Vince]

Fall 2006. Sony is so far ahead of the game there is no reason to go any earlier.

I wouldn't be so sure of this. It's a nice, if somewhat superficial, picture - but ultimatly wrong IMHO. If you read what President Kunitake Ando and Nobuyuki Idei are working towards; both as an abstract vision and "in-play" praxis, what many of us are saying makes much sence.

Again, Cell is much bigger than just PlayStation3. And their vision of the 21st century Sony on it's 60th anniversary is much bigger than just Cell. Fundimental shifts in their products must happen by then, and they won't take place overnight. 2005 has more going for it than you can imagine.

 

[jvd]

Hey, guys, what's the excuse going to be "if" Paul is proven right? I'm just curious. Will you admit to lacking the perception and sence Paul has or just let it slide? Or will you stumble into a DeammeatGA type condition and suffer from Dissocial Personality Disorder; the entire time claiming, "Ken Kutaragi changed the specs and fabs 'cause he wanted Paul to be right."

Because, objectivly Paul basically has every precedent and logical extrapolation from public (and not) info on his side.

Hey i believe what paul siad can happen but so far he hasn't shown me any proof of it. I've asked for it twice. they could verywell be making cell chips for the ps3. But they can also be just as likely making chips for the psp at the toshiba plant .

 

[nonamer]

Fall 2006. Sony is so far ahead of the game there is no reason to go any earlier.

I wouldn't be so sure of this. It's a nice, if somewhat superficial, picture - but ultimatly wrong IMHO. If you read what President Kunitake Ando and Nobuyuki Idei are working towards; both as an abstract vision and "in-play" praxis, what many of us are saying makes much sence.

Again, Cell is much bigger than just PlayStation3. And their vision of the 21st century Sony on it's 60th anniversary is much bigger than just Cell. Fundimental shifts in their products must happen by then, and they won't take place overnight. 2005 has more going for it than you can imagine.

Vince, no offense but lose common sense when talking about Cell/PS3. Cell could be the second coming and still no reason to be out by 2005 (it works against it to an extent; get it right than get it on time). 65nm looks to be like a doozy, at least in the area of heat management. Leakage is going to be worse than in 90nm (which Intel is proving to be quiet bad), plus you're looking at a [two] really big chip(s). 2005 is, while concievable, looks to be extremely expensive and risky. My bet is that they'll wait till 2006 in order for yields to be good (an absolute must given the size of what they're producing) and for power consumption to come down, which means more work designing the Cell for better heat management and more delays. As it stands now, I would say that the entire next generation consoles will all be out in 2006 (with possible exception of the Xbox2 if they go with 90nm or 110nm, which would be disapointing but thats off-topic). Getting out in 2006 shouldn't be a problem, especially for heavyweight PS3. Whenever it's out will be the right time to be out if get what I'm saying.