Microsoft moves into chip world with Xbox Next

chaphack said:
Sony, Toshiba and IBM, however, have been working on Cell since 2001, and it won't come out until at least late 2005.
Hmm....US or Japan...

Likely Japan, though I would estimate a shorter launch delay in NA and Europe than before, since they know the others are hot on their heels.
 
MS may have a lot of money, but they can't afford to spend it all. They already blew like what, $2 billion on the xbox?(probably more, since I think sega was $2 billion in debt after the saturn) Right with IBM they maybe nearly spending that amount, perhaps they want something like a 256 gigaflop cpu from IBM(money doesn't grow on trees, high tech chips do) and just use the latest off the shelf part from ATi.
 
MS may have a lot of money, but they can't afford to spend it all.

I beg to diagree, if they are still sreious about making inroads in this market, they'll fight with everything they've got.
 
chaphack said:
Any info on how much they are plunging into Ati? A lot more money MS has. :oops:

Money can buy you only so much time and 2 extra years of R&D are difficult to match for anyone... do not expect something inferior, but do not expect something superior either specs wise.
 
Panajev2001a said:
chaphack said:
Any info on how much they are plunging into Ati? A lot more money MS has. :oops:

Money can buy you only so much time and 2 extra years of R&D are difficult to match for anyone... do not expect something inferior, but do not expect something superior either specs wise.

you're assuming that they start from scratch rather than building from salvegable design from XXX. it's not like they are operating in a vacum in terms of R&D.
 
any ideas on what aspect is Sony R&Ding on? concentrato on making the most healthiest, connectivity-est general purpose cpu for their connected living room dreams or one really kickass dedicated 3D gaming system? it could be both, hmmm...? launch Xgames might look better than Pgames if the traditional vs radical comes true. interesting...

anyone missed this
By switching from using relatively standard parts to more customised silicon, the company can better optimise its game console, due in 2005. At the same time, the move potentially gives the company a toehold in a completely new market.
like i suspected, MS wants to beat Sony to the US market. winter 2005!

Chairman Bill Gates is expected to discuss Xbox next at the CES show in Las Vegas this January, Doherty said.
Thats fast. Hmmm..
 
notAFanB said:
you're assuming that they start from scratch rather than building from salvegable design from XXX. it's not like they are operating in a vacum in terms of R&D.


hmmm...thats right. MS might take existing IBM research and force accelerate it with their hard cash and gulping down the initial cost. hmmm..? like i said, MS will never allow PS3 to overpower them. If you can do 5 years supreme physics calculations, so are we going reach for that! hmmm...
 
Still no word on backward compatibility (whether there will be any at all that is)?

DaveBaumann said:
I'd still say this is going to have fundamental similarities with the contempary DirectX functionality.
It's likely in their best interest to keep it that way - given where majority of their 3rd party support came from so far.
 
chaphack said:
notAFanB said:
you're assuming that they start from scratch rather than building from salvegable design from XXX. it's not like they are operating in a vacum in terms of R&D.


hmmm...thats right. MS might take existing IBM research and force accelerate it with their hard cash and gulping down the initial cost. hmmm..?

I prefer the term 'realise' to any designs IBM have on the table myself.

force acelerate is a magic term which has no meaning.
 
Paul said:
MS will never allow PS3 to overpower them.

Tell me Chap, what is a better process; the 65 nm node or the 90/130 and 110.

Telling you Paul, how the fuck would i know!? :LOL: well, size wise, the smaller the better i guess, but the performance? hmmm...hell, is that even any confirmation that those are to be the respective die sizes?

OH HALT everyone! Hmmm...didnt Ati larger R300 core outperform Nvidia smaller NV30 core? 3D graphiX goes to Ati today, yeay?
 
Chap please refrain from using profanity, thanks

regarding fab process, it's pretty important since your effectively squeezing more logic (read functional unit) into the equivalent space.

your nVidea comparison is used often but misunderstood, it's alot more complicated than the die process (I suppose you *could* put it down to incompetence, but let's not start a flame war here).

for a fixed unchanging architecture, once the design is locked donw the smaller the better.
 
Telling you Paul, how the fuck would i know!? well, size wise, the smaller the better i guess, but the performance? hmmm...hell, is that even any confirmation that those are to be the respective die sizes?

Smaller transistors means more you can pack on a die, more transistors means more gates, meaning higher ability of the IC, meaning a higher specification.

Sony is using 65 for both their CPU and GPU IC, X2 CPU will use either 130 or 90, and the GPU 110.

Comparing NV30 and R300 to this is just wrong. Cannot compare it to the console world Chap, it's like night and day.

Oh yea and R300 had 110 Million transistors, NV30 125. Nvidia didn't push 130 for shit, not happening with Sony and MS. Both will push the lithography to the extremes probably to the point of bad yields.
 
notAfanB,
happens once with the PS2. "incompetent" if you will. Sony fannys like to call it "locked down too early". ;) Cell is, what mulit core maddness, i heard its even harder to reach high efficiency..hmmm..? when was the patent filed? yeay know the one with BE and VS thingie? i think, unless, Sony wants to conspire to expire MS( :LOL: ), it be good for billy and gang to study it well!


Paul,
Too much assumption. and lithography is lithography(that the right term Vince? :LOL: ), console or pc dont make too much difference, hardware is created to run the 1s and 0s, yeay? it be good, it be bad. now if you are talking bout the software making use of the hardware, then, yeay guess console > pc.

hmmm...with less transistors, R300 can still outwork a NV30. cool. if you can assume that Sony will get it all right, it be fair that i can also assume the possibility that Ati might once again create the more elgant design.

Hmmm..i say... :oops:
 
Faf,

It's likely in their best interest to keep it that way - given where majority of their 3rd party support came from so far.

Come on man... you know the "majority" of 3rd party support on xbox didn't come from the PC world. it's a much smaller percentage than half... probably closer to 10-15%
 
Yuppers. It is more like Western vs Japanese. Since most western dev work on PC, no harm done with a quick and dirty port about.

BUT i think backward comp is way way way over rated. apart from playing around with the filter settings, i have NEVER play thru any ~20fps(exaggeration) Pixelstation1 "3d" games! :LOL:
 
I'm not assuming anything, Sony is going to be working with 65 and MS a far more outdated process.

Again comparing dedicated IC components(in a console) to that of a PC GPU is just wrong.

NV30 has a game optimized to make it run better than if the same game were run on the R300, is the NV30 more powerfull?

Oh yea and being 'elegant' means nothing when your talking about the max specification game. A group of inexperienced kids can win against 4 military commando's if the kids have guns and the commando's knives.
 
Back
Top