Xbox 2 in 2006?

[notAFanB]

No.

A 733 celery might not be the hottest CPU back then, but for a game console(versus the competition) coupled with the latest GPU, a 8GB HDD, Network Card, 5.1 Audio and Nforce Mobo, it is excellent, more than enough. I dont see why the usual people are laughing at MS choice for a 733 celery. They are giving you an unbeatable deal at just 299.

Would i want a faster CPU or GPU? Sure i would, but looking at all conditions available, i am pleased with what Xbox has given me. Cant say the same for some other systems.

then try to avoid it (joking or not I suppose) whose claiming such a thing. for a console the XCPU is fine but if (note discussing if) we were to move the Xbox into the destop arena (as a toy). the drawback of MS decision become a factor of concern.

no one is 'laughing' at MS, they made a design decision to go with the celeron733 and it paid off for them so far.

Note: still waitingfor your sources, care to elaborate.

 

[function]

As people have already said, price is the key issue here. The Xbox is by far the most expensive console to manufacture, and Microsoft are still reported to be taking a significant loss on each one nearly 2 years after its release. And I don't think anyone doubts that for their money they've produced the most advanced and well equipped games machine. Calling any company cheap ass under these circumstances seems a little unfair.

And while it's true that no "hardcore PC gamer" would build a PC with a Celeron 733 at it's core (and again, the Xbox isn't using a straight Celeron), no "hardcore gamer" would use only 64MB of RAM or an 8GB HD either. And that's because no hardcore gamer has to design and manufacture a (relatively) small, cool and quiet console to sell for $299 and lose money on each one.

A more expensive CPU would undoubtedly have made the Xbox more powerful, but which other areas of the machine should they have cut back on to balance that cost? The GPU? And how much bigger, hotter and louder should they have been prepared to make the machine? Just asking some rhetorical questions.

 

[chaphack]

What Function said. Good job me friend. 8)

 

[chaphack]

then try to avoid it (joking or not I suppose) whose claiming such a thing. for a console the XCPU is fine but if (note discussing if) we were to move the Xbox into the destop arena (as a toy). the drawback of MS decision become a factor of concern.

no one is 'laughing' at MS, they made a design decision to go with the celeron733 and it paid off for them so far.

Note: still waitingfor your sources, care to elaborate.

What the heck are you talking about? I think you better re read this thread before adding more comments....

 

[notAFanB]

sorry rambling again.

We shall see....we shall see
I have my sources.

regarding the above care to clarify?

 

[mat]

And I'm sure you already know it's not a standard Celeron (or P3) in there anyway.

Actually, it is a standard (mobile) Celeron.
Its this one

 

[AzBat]

And I'm sure you already know it's not a standard Celeron (or P3) in there anyway.

Actually, it is a standard (mobile) Celeron.
Its this one

Good link. Here's another...

http://www.vanshardware.com/article..._More_On_Xbox_CPU/011206_More_On_Xbox_CPU.htm

Tommy McClain

 

[Saem]

No different from games which run all T&L on the graphics card on the PC, where it didnt beat the Duron clock for clock either.

I can't agree with this.

The difference with the XBox is that it's more than static T&L AFAIK, most games that use T&L functionality of hardware are still doing the static stuff.

 

[function]

Then I stand corrected! A mobile Celeron 733 it is.

It looks like this processor is in Intel's ultra low voltage range now (although with a 100mhz bus and 256K of L2 cache): http://developer.intel.com/design/mobile/celeron/index.htm. Perhaps Microsoft were planning on a low power version of the Xbox CPU to produce the much rumoured smaller version of the Xbox. I suppose we'll know if such a thing appears this year.

 

[Sonic]

Saem, are you stating that most Xbox games are onl using the vertex shaders for static geometry? Or did you mean that about PC games, or both?

 

[Saem]

Saem, are you stating that most Xbox games are onl using the vertex shaders for static geometry? Or did you mean that about PC games, or both?

Well, I'm not really talking about XBox games, a bit more on the theoretical side saying this is possible and likely what's happening.

What I'm saying is that on PC games it's all DX7 T&L, static geometry.

On the XBox, this isn't the case.

 

[nonamer]

So we'd be looking at:

PS3
2005 March/Q1 Japan launch
2005 November/Q4 North America launch
~sometime 1H 2006 - PAL launch

I don't think any console can safely launch as early as Q1 2005. AFAIK, a chip needs like 6 months or so of testing before they can actually release it. If the Oita fab is ready in mid-2004, then it could start producing the PS3 chips no earlier than the very end of 2004. Give a few months need to produce the chips in quantity, March will pushing it extremely, and it would be super rushing things. One single hiccup and March will be impossible. My bet is that Sony will wait and make sure everything is okay before launching, which is probably no earlier than late 2005. Just my $0.02

 

[Paul]

If the Oita fab is ready in mid-2004, then it could start producing the PS3 chips no earlier than the very end of 2004.

Kutagari has stated that they will demo Cell by the end of this fiscal year, probably at GDC. They are obviously testing the chip right now.

 

[nonamer]

If the Oita fab is ready in mid-2004, then it could start producing the PS3 chips no earlier than the very end of 2004.

Kutagari has stated that they will demo Cell by the end of this fiscal year, probably at GDC. They are obviously testing the chip right now.

Yes, but probably at 130nm or 90nm. The 65nm process doesn't exist yet.

 

[Paul]

There are no plans of a 130nm or 90nm Cell, only 65.

They have already started to construct the Oita plant, and it's going into mass production spring 2004. This doesn't mean they can't get 1-2 Cell chips out of it for a feb-march demo session. Since the thing is going to be completed by January.

 

[nonamer]

There are no plans of a 130nm or 90nm Cell, only 65.

They have already started to construct the Oita plant, and it's going into mass production spring 2004. This doesn't mean they can't get 1-2 Cell chips out of it for a feb-march demo session. Since the thing is going to be completed by January.

Then they don't have Cell now.

First news of Cell was that it'll be on a 100nm process, which probably means 90 now. I doubt Cell could be only on 65, especially if this is suppose to be on a lot of things, not just PS3. A 65nm-only Cell probably would be too expensive for too long for this to happen. And they certainly would have it for a feb-march demo in 2004. It takes months of testing the silicon before they have anything working yet, and the fab in Oita will probably produce mainly DRAM or SRAM cells at first, nothing really complex AFAIK.

 

[Paul]

Sony has NO plans of a Cell other than .65 nm.

Official slides indicate this.

And nope, Cell was first announced to be smaller than 100nm.

Code-named "Cell," the new microchips will employ the world's most advanced research technologies and chip-making techniques, including copper wires, silicon-on-insulator transistors and low-K dielectric insulation, with features smaller than 0.10 microns -- 1,000 times thinner than a human hair.

http://www.ibm.com/news/2001/03/12.phtml

If Kutagari says expect a prototype by the end of this -fiscal- year than there is no reason not to believe him. Also no reason not to belive that it will not be on .65 nm as Sony has no plans for different micron Cell's.

Panajev knows more about the Fab's than I do.

 

[Vince]

Sony has NO plans of a Cell other than .65 nm.

This is wrong my friend. AFAIK, the initial cell incarnations will be using 90nm SOI and could (should?) appear in the late 2004 timeframe - this is what STI-Austin is presently doing. My indication is that SCE/Toshiba will then take the joint STI architecture, move it to 65nm and produce it for PS3 in the 2005 timeframe.

 

[Panajev2001a]

Of course implementations of Cell can be done in 90 nm as well as 65 nm

Cell for PlayStation 3 is intended for 65 nm ( 90 nm would yeld a chip which is WAY too big and expensive )... Oita #2 and Nagasaki #2 both have quite big 65 nm lines and Oita IS scheduled to enter VOLUME MASS PRODUCTION by mid 2004 which means that quite a bit of wafers are prduced and working chips are made before that... the date of completion for the fab is late January 2004...

 

[Panajev2001a]

Sony has NO plans of a Cell other than .65 nm.

This is wrong my friend. AFAIK, the initial cell incarnations will be using 90nm SOI and could (should?) appear in the late 2004 timeframe - this is what STI-Austin is presently doing. My indication is that SCE/Toshiba will then take the joint STI architecture, move it to 65nm and produce it for PS3 in the 2005 timeframe.

Again, I agree with Vince: we should see Cell implementations ( that is finished products ) suing 90 nm SOI Cell chips...

A 1 TFLOPS class Cell processor with a good amount of e-DRAM with the bandwidth to keep the processor well fed should be on 65 nm...

BTW, in Nagasaki#2 Sony will manufacture both the PSP as well as Cell chips using 65 nm technologies: PSP will use 90 nm technology and will be made in the second floor of that fab which also houses the PSX and the EE+GS@90 nm production...