Cell mass production plan for 2nd half of 2005
- Thread starter V3
- Start date
Panajev2001a
Veteran
Nonamer, I partially agree with you and partially with Vince.
I expect a Japanese launch around either Q4 2005 or Q1-Q2 2006.
Why do I say that ?
Because I want to make sure that the Japanese launch would be as close as possible to the availability of the 45 nm SOI process ( with capacitor-less e-DRAM cells ) SCE, Toshiba and IBM are working on ( SCE and Toshiba are leading on this new manufacturing process it seems ).
Why ?
Because I do expect the chips to be expensive using 65 nm technology ( due to sheer size ~280 mm^2 or more and due to initial non optimal yelds ) and SCE would take the loss, but it would only do that for a limited period.
Of course this limited period would also include yelds for 65 nm increasing which would give SCE some more time.
I expect 45 nm to be ready by Q2-Q3 2007 approximately, but I do hope that early Q2 will be the effective date.
I expect a Japanese launch around either Q4 2005 or Q1-Q2 2006.
Why do I say that ?
Because I want to make sure that the Japanese launch would be as close as possible to the availability of the 45 nm SOI process ( with capacitor-less e-DRAM cells ) SCE, Toshiba and IBM are working on ( SCE and Toshiba are leading on this new manufacturing process it seems ).
Why ?
Because I do expect the chips to be expensive using 65 nm technology ( due to sheer size ~280 mm^2 or more and due to initial non optimal yelds ) and SCE would take the loss, but it would only do that for a limited period.
Of course this limited period would also include yelds for 65 nm increasing which would give SCE some more time.
I expect 45 nm to be ready by Q2-Q3 2007 approximately, but I do hope that early Q2 will be the effective date.
jvd said:
PSP doesn't fall under any inter-corperate production agreement or alliance, did you even check?. As in, unlike Cell which is bound under not only the 2001 STI agreement for architectural development and the seperate 2001 agreement on SOI technology transfer, but also the subsequent 2002 agreement on cooperative process development. And most importantly, the extension of OTSS which has seen Sony invest in Toshiba's 65nm fab under the same corperate movement that had SCEI begin work on Nagasaki's 65nm lines, which we know are Cell.
This line of argument is akin to your earlier one in which many venomously opposed the idea that Cell would be in PS3 because Sony said it was a "Processor for the Broadband Age not intended for PS3" or some BS like that. Just like then, you will be proven wrong again.
Nonamer, with all due respect everything you said amounted to your last sentance, "Whenever it's out will be the right time to be out if get what I'm saying." Which is about as decisive as one of JoeDeFuria's posts on these topics. I don't need to get into another prolonged debate over this. I'm confident time will prove me right, I've previously stated why over a plathora of threads.
Panajev2001a
Veteran
Man, late 2005 is not too late 
PSP doesn't fall under any inter-corperate agreement or allience for your information. As in, unlike [/i]Cell[/i] which is bound under not only the 2001 STI agreement for architectural development and the seperate 2001 agreement on SOI technology transfer, but also the subsequent 2002 agreement on cooperative process development. And most importantly, the extension of OTSS which has seen Sony invest in Toshiba's 65nm fab under the same corperate movement that had SCEI begin work on Nagasaki's 65nm lines, which we know are Cell.
This line of argument is akin to your earlier one in which many venomously opposed the idea that Cell would be in PS3 because Sony said it was a "Processor for the Broadband Age not intended for PS3" or some BS like that. Just like then, you will be proven wrong again.
don't have a clue as what you are talking about. I have only said the cell chip in the ps3 will not be a 1 tflop chpi. I will be right about that . As for your first part The psp chips need to be built some where. .65 will be the best for heat and cost. Toshiba can run the psp chips through thier plants and tweak the line. Or of course toshiba can be making cell chips for thier own products. Like I said show an announcement where toshiba or sony say the thosiba plant will make the cell chips that are to be used in the ps3. Till then you are wrong .
This line of argument is akin to your earlier one in which many venomously opposed the idea that Cell would be in PS3 because Sony said it was a "Processor for the Broadband Age not intended for PS3" or some BS like that. Just like then, you will be proven wrong again.
don't have a clue as what you are talking about. I have only said the cell chip in the ps3 will not be a 1 tflop chpi. I will be right about that . As for your first part The psp chips need to be built some where. .65 will be the best for heat and cost. Toshiba can run the psp chips through thier plants and tweak the line. Or of course toshiba can be making cell chips for thier own products. Like I said show an announcement where toshiba or sony say the thosiba plant will make the cell chips that are to be used in the ps3. Till then you are wrong .
Vince said:
Vince, I don't think you "got" what I'm saying. No matter when the PS3 will be out, the timeframe it came out will be considered the timeframe that the next consoles are supposed to be out and all other consoles will be judged by that timeframe. For instance in the PS2's case the Dreamcast was considered early while GC and XB were considered late simply because the Dreamcast came before the PS2 and the others afterwards. The same case will be for the PS3. And by the way that's not the main thrust of my post, which is that it isn't absolutely necessary for the PS3 to be out by 2005, and in fact economic concerns will, IMO, force it to be out in 2006. I don't why you're attacking my personally, especially comparing me to a guy like DeFuria who I have next to nothing in common. I'm not going to take this for pages and pages, just pointing out my beliefs, Okay?
jvd said:
Problem perhaps my friend? Think it's a problem you should resolve before posting what you did at the bottom about PSP?
Hmm...
Nonamer said:
Excellent point. I sincerly appologize. Although, I disagree about the 2005 point. There is signifcant intrinsic benefit for Sony to launch PS3 as early as possible for several reasons.
marconelly!
Veteran
Just out of curiosity, what do you suggest Sony does with those expensive chips they would be making and not willing to put on the market at lower price? Let them sit in a storage for a year (losing even more money) It just doesn't really make sense to me :?
marconelly! said:
They're not going to let those chips sit in a warehouse for a year; they won't exist. In other words, STI is not going to mass make any Cell chips until yields are at an absolution maximum because it would be too expensive otherwise. Of course, this is assuming that costs are VERY high, so if the cost are lower then there will less of a concern on costs per unit and more on volume. Regardless, I think that yields and costs will be problematic enough to delay the PS3 til 2006, where yields are at least high enough, if the max is not necessary.
Another question for the Sonyers, say this 1TFLOPS number floatign around...is it that impressive? I mean, whats the FLOPS ratings of current top of the line PC hardware or at least teh bestest GPU( i heard FXs are about 200GFLOPS -> 5X lesser FLOPS?)
1TFLOPS...does that take into consideration how rendering is to be done? For instance, say Sony's preferred "polygon madness" approach, does it needs more FLOPS to acheive similar results as ATI "shaders" approach, or simply the more FLOPS the merrier?
Dont get me wrong, i am sure if Sony really loves the 1TFLOPS mantra, PS3 will have 1TFLOPS, one way or another.
BUT I think we should at least, have a yankity comparison before going gaga over 1TFLOPS *chuckles* ? 2006 is a long way off for more advancement with the other 3D guys.
1TFLOPS...does that take into consideration how rendering is to be done? For instance, say Sony's preferred "polygon madness" approach, does it needs more FLOPS to acheive similar results as ATI "shaders" approach, or simply the more FLOPS the merrier?
Dont get me wrong, i am sure if Sony really loves the 1TFLOPS mantra, PS3 will have 1TFLOPS, one way or another.
BUT I think we should at least, have a yankity comparison before going gaga over 1TFLOPS *chuckles* ? 2006 is a long way off for more advancement with the other 3D guys.
chaphack said:
we discussed this some time ago, and the answer was: "? :? ! :| ?
!".basically, NVFlops were discusses as opposed to "otherFlops" (which are usually strictly related to polygon counts)... the 200GFlops Nvidia and ATI are talking about are related to other things than the, for example, 6.2GFlops the EE is theoretically capable of doing... and we still dont know whether this 1Tflop is a NVflop rating or a traditional one.
if it's a 1Tera-NVflops then that means the graphic part of PS3 could be 5 times as powerful as current PC cards (which is a lot)
if it's a "traditional" rating, then it means that PS3 would be like hundreds of times the power of current hardware, and definately VERY different from anything available now, so much so that any comparison would just be out of place....
if i got this whole thing wrong, please do correct me...
Almost everyone so far has already said that what something "does" in the simplest of terms bears no discernable correlation to what the console can produce gaming-wise, or rather... what developers can GET the machine to do. PS2 bears no real correlation, Xbox doesn't... (In a way even the DC didn't because they announced numbers too LOW! )
The proof of the pudding is in the eating, and we're not even going to get a SNIFF until next E3, so in the meanwhile arguing about whether the PS3 will DEFINITELY get to 1TF, or whether it will "merely" be half that or less... it all makes no difference because we STILL wouldn't know what the number MEANS until we see the whole package--and see the results.
Personally, I figure they'll hit that number in SOME manner, but even if they announce it TOMORROW I still won't care at all because we won't know what it means until we start seeing results. We won't know how to put it in perspective until we can see the whole machine--not one number. From there we can continue to extrapolate and go through all the usual motions, but right now it doesn't make SHIT difference.
I will be happy when I start seeing screenshots and gameplay videos, and have hot little titles in my own hands. THAT is when I start to judge what the "next generation" is bringing us.
----
Meanwhile, on the OTSS plant, for a section co-invested with Sony (probably among others) and bringing 65-nm chips to light in 2004, I can't think of anything ELSE that refers to but CELL. (Especially considering it's mentioned in the midst of articles entirely ABOUT that processor.)
)The proof of the pudding is in the eating, and we're not even going to get a SNIFF until next E3, so in the meanwhile arguing about whether the PS3 will DEFINITELY get to 1TF, or whether it will "merely" be half that or less... it all makes no difference because we STILL wouldn't know what the number MEANS until we see the whole package--and see the results.
Personally, I figure they'll hit that number in SOME manner, but even if they announce it TOMORROW I still won't care at all because we won't know what it means until we start seeing results. We won't know how to put it in perspective until we can see the whole machine--not one number. From there we can continue to extrapolate and go through all the usual motions, but right now it doesn't make SHIT difference.
I will be happy when I start seeing screenshots and gameplay videos, and have hot little titles in my own hands. THAT is when I start to judge what the "next generation" is bringing us.
----
Meanwhile, on the OTSS plant, for a section co-invested with Sony (probably among others) and bringing 65-nm chips to light in 2004, I can't think of anything ELSE that refers to but CELL. (Especially considering it's mentioned in the midst of articles entirely ABOUT that processor.)
Similar threads
