Toshiba, Sony close to 65nm sample production

[cybamerc]

cthellis42:

> If they DID launch worldwide in Q4 2004

2004 is out of the question. Minimal shipments in 2005 are possible but I don't think a world wide launch is likely.

 

[London Geezer]

If they do a ps2 on us, Japan will get the console in March, it will come out in America in September and in Europe in November. I think that's very similar to what they did at PS1 launch.

Now all we have to do is think of a year... 2005? 2006? Personally i think, if they push it, they could go for 2005 (march 2005 Japan, autumn for the rest of the world) but it will probably be PS2 all over again, shortages, murders in front of retailers, me awaiting months til i could get my hands on one, high risk of faulty machines, me bored at home with nothing to do waiting for my system to arrive, overpriced consoles, me ending up doing some study...

You know, usual stuff...

 

[passerby]

if they launch ANYWHERE in 2005 it would be Japan only, and a US launch somewhat afterwards

Agreement. Around November/December if 2005.

 

[marconelly!]

In the end, i would prefer a delayed-but-perfect-system than a rushed-but-flawed one...

Yeah, but the implications of such a delay would be that the technology used wouldn't be so 'hot' anymore. That is why I think people are hoping for an earlier release

 

[Paul]

A worldwide launch could just be North American and Japan with the rest waiting until Q1 06

Manabe said this Feburary of this year.

At a conference this week, Sony CTO Kenshi Manabe said: "We intend to launch the successor to PlayStation 2 ahead of our initial schedule, which was drawn up some years ago. This, we believe, will bring us ahead of Microsoft, which is planning a new console towards the end of 2005."

Oh and a new article.

Rob Fahey 10:52 11/12/2003
Industry-leading 65nm process to begin sample production in Q1 '04


Trial runs of the advanced semiconductor manufacturing process which will eventually create the much-vaunted Cell microprocessor are set to start at Toshiba's fabrication plant in March of next year.

According to a statement from Sony and Toshiba issued today, work on the 65nm chip production technology - which is more advanced than any system in commercial use today, as most companies are still coming to grips with the switch from 130nm down to 90nm processes - is proceeding to plan.

The nm (nanometre) measurement is important, because the smaller the size of the components that are used on a chip surface, the faster and less power-hungry the devices can be - a key consideration in the creation of Cell, which has been described as a "supercomputer on a chip".

Toshiba is expected to produce the first sample chips from its 65nm line in March and will ship them to customers for evaluation, but full production of the technology is not expected to ramp up until summer 2005 - just about in time to supply components for Sony's PlayStation 3 launch at the end of 2005, if that is indeed the plan.

The sample production will take place at Toshiba's Yokohama fabrication plant, but the commercial 65nm line will be in a new factory, which is currently being built in Oita prefecture and is expected to begin producing chips at 90nm in mid-2004 before being upgraded to the 65nm process further down the line.

Sony is also building its own fabrication plant for the 65nm Cell chips, with an investment estimated to be in the region of €5 billion being made in new plant in Nagasaki prefecture.

http://www.gamesindustry.biz/content_page.php?section_name=dev&aid=2702

 

[passerby]

Someone inquired about heat and power. Nothing is really known. As some kind of very very rough reference which is probably useless for estimates but still interesting, EE+GS at 90nm is supposed to eat just 8 Watts. I remember seeing the article at ruliweb.

 

[No-name]

Um, if PS3 launches late 2005 in Japan, then wouldn't that mean the North America does not get PS3 till fall of 2006 anyway, unless Sony released in the NA at the same time(fat chance)? Pretty long wait. Fall release is the most ideal time for a console in the N.A. and has been for half a decade. Late 2005 or early 2006 doesn't as much of a difference in the big picture as total release is concerned.

 

[cthellis42]

cthellis42:

> If they DID launch worldwide in Q4 2004

2004 is out of the question. Minimal shipments in 2005 are possible but I don't think a world wide launch is likely.

Brain fart on my part. I of course meant 2005.

Um, if PS3 launches late 2005 in Japan, then wouldn't that mean the North America does not get PS3 till fall of 2006 anyway,

I don't think they'd delay the 9 months they did with the PS2 from Japan to US launch--more like 3-6 months somewhere, as they have more resources to bring to bear for this one, and KNOW that there is competition from the Xbox breathing down their neck, which back in 2000 they knew there was no rush. Question is how much volume they can put out by the appropriate time frames.

 

[MrSingh]

2006 earliest. Place your bets everyone!

 

[hey69]

2005 is a bit stupid no? there are'nt even games for the unit that we know of. i hope E3 2004 will bring a lot of clearity

 

[Wunderchu]

That Toshiba would start mass producing Cell in the middle of next year.

Actually it was near the end of 2004, a window to late summer to October.

I'm fairly certain that when they say mass production could start April 05 they mean production with good yields and not risk production. Risk production could begin late 2004...

A previous Toshiba Oita PR stated that they would begin mass production Late summer to October.

Anyway, this confirms my guess that PS3 will have a worldwide launch of Fall 2005. Both the Toshiba and Sony 65 nm plants will be mass producing Broadband Engine and related 65 nm IC for upwards of 6 months prior to launch. On 300mm wafers I might add.

... however, .. :


[I am assuming that the manufacture of CELL will be a major (if not the major) factor in determining how soon the PS3 will launch]


1. - CELL is speculated to be a huge chip (according to rumors/speculation, CELL may be over 400 million transistors in size) ....... even with 300 mm wafers @ 65 nm , each CELL will still take up a lot of space

2. - remember that each PlayStation 3 will contain more than 1 CELL chip (again, from rumors/speculation each PlayStation 3 will contain as many as 8 CELL chips) ....... so, the PS2 required around 6 months of mass production to get enough consoles ready for launch .. but for each PS3 , as many as 8 CELL chips will be needed ... even with Toshiba (and maybe even IBM) helping Sony out with manufacture, they will still need to manufacture a lot of chips ...

 

[Paul]

PS3 is using Broadband Engine, it was rumored by everyone here that this was the Cell based CPU as found in the patent that would be used in PS3.

The Toshiba Rambus SCE contract proves that Toshiba and SCE are indeed building Broadband Engine and that it is going to use Rambus Redwood and Rambus XDR memory. Aka this configuration is Figure 6...

BE in the patent is comprised of 4 PE, each PE has 8 APU, thus 32 cores.

The final BE that Sony and Toshiba are building could be higher or lower in power though, we do not know.

When the media says "more than 1 cell" they are talking about PE's stuck together on a chip, Broadband Engine as seen in the patent has 4. They do not mean seperate physical chips.

 

[Wunderchu]

PS3 is using Broadband Engine, it was rumored by everyone here that this was the Cell based CPU as found in the patent that would be used in PS3.

The Toshiba Rambus SCE contract proves that Toshiba and SCE are indeed building Broadband Engine and that it is going to use Rambus Redwood and Rambus XDR memory. Aka this configuration is Figure 6...

BE in the patent is comprised of 4 PE, each PE has 8 APU, thus 32 cores.

The final BE that Sony and Toshiba are building could be higher or lower in power though, we do not know.

When the media says "more than 1 cell" they are talking about PE's stuck together on a chip, Broadband Engine as seen in the patent has 4. They do not mean seperate physical chips.

ohh... thanx for the insight

 

[V3]

They do not mean seperate physical chips.

Don't know about that. Broadband Engine could be on some sort MCM. Depend on how they tackle the problem. They'll just go to which ever is cheaper for them.

Toshiba is going start sample production early 2004, if all goes well, this could mean dev kit would shift soon after.

I think if PSP and PSX is successful as Sony hope for, PS3 might be a 2006 product.

 

[No-name]

cthellis42:

> If they DID launch worldwide in Q4 2004

2004 is out of the question. Minimal shipments in 2005 are possible but I don't think a world wide launch is likely.

Brain fart on my part. I of course meant 2005.

Um, if PS3 launches late 2005 in Japan, then wouldn't that mean the North America does not get PS3 till fall of 2006 anyway,

I don't think they'd delay the 9 months they did with the PS2 from Japan to US launch--more like 3-6 months somewhere, as they have more resources to bring to bear for this one, and KNOW that there is competition from the Xbox breathing down their neck, which back in 2000 they knew there was no rush. Question is how much volume they can put out by the appropriate time frames.

That would be nice to recieve PS3 in our area in a shorter amount of time, but just that looking at history, a new generation of machines now get release in the Fall of a year in North America regardless of it's initial Japanese release. It's the most ideal and tends to be the best time to launch. I don't see them releasing a console in the Spring or even the Summer in North America.

 

[nondescript]

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.

However, when and where they decide to launch probably depends more on what MS and Nintendo decide to do. IMHO, it's highly unlikely Sony will let Xbox2 have a Christmas season in the US all to itself. If MS launches Xbox2 in Christmas 2005, Sony will too, even if it can't put out enough units to meet expected demand. In any case, I expect Sony to make the US launch more urgent then in previous years. Since Xbox poses no serious threat in Japan, it can go for a smaller launch in Japan without worry and push up the US launch date or even simultaneously launch with more units in the US.

 

[London Geezer]

They do not mean seperate physical chips.

Don't know about that. Broadband Engine could be on some sort MCM. Depend on how they tackle the problem. They'll just go to which ever is cheaper for them.

Toshiba is going start sample production early 2004, if all goes well, this could mean dev kit would shift soon after.

I think if PSP and PSX is successful as Sony hope for, PS3 might be a 2006 product.

I thought it was common/accepted knowledge that many Cell cores are going to be on-die in one single chip. Not that many Cell chips are going to be solded on the board.

I, for one, am still looking forward to see the approach this architecture will take, and how it will be used...

 

[Paul]

Don't know about that. Broadband Engine could be on some sort MCM. Depend on how they tackle the problem. They'll just go to which ever is cheaper for them.

Possible, A few PE connected by Redwood? Interesting concept and idea, however I still say that it will be a one chip solution with multiple PE.

 

[V3]

I thought it was common/accepted knowledge that many Cell cores are going to be on-die in one single chip. Not that many Cell chips are going to be solded on the board.

Well, that was before Panajev2001a point out this

That suggest some kind of multichip module.

Possible, A few PE connected by Redwood? Interesting concept and idea, however I still say that it will be a one chip solution with multiple PE.

Well speculation is always interesting.

The speculation for one chip BE, was that the APU doesn't have a deep pipeline and its size is less than 300 mm2.

However its more likely that it will be, first to achieve its 32 GFLOPS, and to hide latency when they're actually processing stuff across network (even Kutaragi suggested this one).

Also, its not normally wise to go for a very large chip on a new process. They might have that module as a fallback in case things don't go well. But it could be, that's what they planned from the beginning.

BTW with a module like that, you would still only see one package on the board.

 

[London Geezer]

Well, that was before Panajev2001a point out this

That suggest some kind of multichip module.

Oh... Missed that!