TXB on Xbox2 = 51.2GB/s bandwidth??

[zurich]

Sure a dual core cpu for a console would result in a very small margin of profit, but what about the cost to Intel if the Playstation 3 and Cell perform well? Intel will have a challenge to its own architectures from IBM/Sony/Toshiba 5 years down the road. It's worthwhile to provide a powerful CPU to Microsoft that will go head to head with CELL. The more money IBM/Sony/Toshiba make on the PS3, the more money they have for future R&D, and the more oppritunities they have to spread CELL in to other markets.

Going head to head against CELL would be a very wise move for Intel. Why would Intel want to make it easy for IBM/Sony/Toshiba to introduce a new architecture that threatens all of Intels designs?

You raise a lot of good points, but I don't seem to remember Intel flinching at all over the Emotion Engine announcement back when. And given all of Sony's talk about extending its tech into other markets, one would think that Paranoid Intel (tm) would have done (or said) something about it.

Perhaps Intel is just calling STI's bluff?

edit: in 2005, Intel's 65nm lines will be precious and worth their weight twice over in gold. I think it would take ALOT for them to give up that space for Xbox production.. which is why I think if Intel DOES commit itself 110% to the XB2 cause, it'll be at 90nm.

 

[Gubbi]

I believe it's a foregone conclusion that Intel (among others) will utilize the significant space afforded to them by process improvements by increasing the die % devoted to the cache hierarchy aswell as integrating multiple processing "cores" at some future point.

I agree completely. Lowering miss rate is simple where you get most bang/die mm^2 at this point (that and multiple cores)

Unfortunatly, this doesn't negate the problems associated with superscalars and the x86 platform. An issue which has been discussed here in several pre-existing threads.

But as feature sizes decreases and speed increases more and more compute bound problems turns into memory subsystem bound problems which brings us back to more cache on chip; lowering the fraction of the Si devoted to dealing with x86 related stuff.

Cheers
Gubbi

 

[Brimstone]

Sure a dual core cpu for a console would result in a very small margin of profit, but what about the cost to Intel if the Playstation 3 and Cell perform well? Intel will have a challenge to its own architectures from IBM/Sony/Toshiba 5 years down the road. It's worthwhile to provide a powerful CPU to Microsoft that will go head to head with CELL. The more money IBM/Sony/Toshiba make on the PS3, the more money they have for future R&D, and the more oppritunities they have to spread CELL in to other markets.

Going head to head against CELL would be a very wise move for Intel. Why would Intel want to make it easy for IBM/Sony/Toshiba to introduce a new architecture that threatens all of Intels designs?

You raise a lot of good points, but I don't seem to remember Intel flinching at all over the Emotion Engine announcement back when. And given all of Sony's talk about extending its tech into other markets, one would think that Paranoid Intel (tm) would have done (or said) something about it.

Perhaps Intel is just calling STI's bluff?

edit: in 2005, Intel's 65nm lines will be precious and worth their weight twice over in gold. I think it would take ALOT for them to give up that space for Xbox production.. which is why I think if Intel DOES commit itself 110% to the XB2 cause, it'll be at 90nm.

With the Playstation 2 Sony wasn't teaming up with IBM. Intel and IBM compete against each other in many area's. Although it certainly isn't a blood feud like it is between IBM and Microsoft. Also last time Microsoft didn't have much time to plan out every aspect of the console and work with other partners.

All Intel needs to do is have the Xbox 2 CPU go to market first, instead of having a new CPU go to the PC market first and then trickle down to a console.

For example, Intel creates a new CPU and calls it Gemini, a dual core x86 CPU. Intel starts marketing Gemini just like they did with MMX and the Intel Inside campaign. Microsoft at the same time starts hyping Gemini as a revolutionary design that it impact all software design. The good news for Microsoft is that the Xbox 2 will be powered by a Gemini CPU and they market Intels technology with great zeal.

Six months after the CPU launches for the Xbox 2 console, Intel brings a more powerful PC version of Gemini to the PC desktop. Companies like Dell and Compaq start advertising their PC's with a Gemini Inside logo. Microsofts marketing of Gemini technology in the Xbox 2 combined with Intels efforts should create strong consumer demand in the PC market.

Of course to the the tech savy, they will know that all Gemini is a dual core x86 with some hyperthreading, but to the the average consumer, they will associate Gemini with powerful technology and something you need to have a good PC.

They may end up selling Gemini to Microsoft cheaply, but in otherways Intel wins big.

1) Cell technology is confronted by an Intel product making it harder for Cell technology to gain traction.

2) The marketing hype creates some incentive for consumers to upgrade their PC's with something that has Gemini. One of the big problems Intel faces is that consumers don't really need big fast processors. Most people just send email and surf the web. The good news is that some of the marketing hype comes from Microsoft pushing Gemini technology.

3) They make a little profit from Microsoft and continue to have a strong relationship with them.

 

[_phil_]

Are we doomed to download patches

 

[jvd]

In intel's playground its not just performance but compatablity (sp?) . That is why the athlon 64 is such a scary thing for them and thats why there is an extreme edition p4 coming out . Amd will have a chip that can run all x86 programs as fast if not faster than the p4 and then run 64bit apps that can be easly recompiled for the chip and of course new paths that will give even more performance. It is much scarier than a cell chip that can't run any of the software used in the pc sector.

 

[Megadrive1988]

What is the highest concivable bandwidth we could possibly (reasonably) expect for XBox 2 ?

 

[jvd]

What is the highest concivable bandwidth we could possibly (reasonably) expect for XBox 2 ?

Well if ms really wnats to push it they can prob get ram that does 75gbs .

Thats unless some new tech that i dunno about comes around . They might even have 32 megs or more of fast on die ram included in their version of the r500.