Official: ATI in XBox Next

[Joe DeFuria]

PS. Joe do you not have a mind of your own? This is simple, I refuse to play your semantic game of backpeddling and not stating opinions while eating other people up based on linguistics. Your ideas have little bearing and are nonexistent. If you refuse to put up - then shut the heck up.

Vince, be a big boy, will you?

Why should I just repeat the same exact thing that Dave has already explained, if I agree with it? If you refuse to accept my opinion, then shut the heck up, or at least try to be a civil participant.

Dave Baumann is a great guy, but I'm not asking him - which I would have precisely done if I wanted his opinion.

Boo hoo. He gave it anyeay, didn't he? And that's my exact opinion as well.

I want to hear yours and what you think ATI is going to do.

See Dave's Post.

I think what ATI does hinges in large part on whatever Fab Microsoft has in mind to actually produce the chip. If MS is planning on TSMC or UMC, then I think ATI will get more directly involved in the actual place and route chip level design.

If MS is planning on going with a fab that ATI has little interest in for their other products, then I don't anticipate that ATI will be doing mu more than supplying software IP.

Is that OK with you?

Perhaps expand on the consequences (pro & con) of a 3rd party taking over a netlist for an advanced IC and running with it.

Pro: the 3rd party may be more familiar with a specific fab process than ATI, meaning that the 3rd party will have an easier time productizing the IP than ATI would.

Con: whenever you have multiple parties involved which assume some sort of responsibility, with Information flowing between them, you tend to have some more issues with managing the project and getting it done. There is a high degree of cooperation involved, and it doesn't always run as smoothly as expected.

This is exactly why I've argued in the past, (usually related to IMGTEC) that I feel the IP licensing model is FINE for the console market, but I do not like it for the PC market. Console designs flip over every 5 years or so, meaning you have more time to plan and sort things out. The PC race is break-neck, and having a more true fabless semi-con model is more beneficial.

I mean you're the ATI investor right? I mean, give us somethign to work with - your debating on merely my ideas and semantics.

Have fun. Now give us all a break and stop with the petty crap. Try debating with a non-combative tone, and stop being on the nVidia defensive as if you have some emotional tie to nvidia's well being.

But, some people are just jagoffs.

Indeed.

 

[Joe DeFuria]

ATI has the tools, experience, set-piece infastructure and relationship with TSMC/UM in place.

Correct...and if MS decides that TSMC/UMC is not in the cards to fab this chip?

To hand off the synthesis of a 60M+ Gate device to a 3rd party just doesn't sit well with me. Granted there are people who do this (eSilicon and , but I've never known them to do ICs with this level of complexity.

As I said in an earlier post...it doesn't sit well with me either for the PC space where things rapidly change. I have less issues with it for the space where turnover is much slower...like consoles, arcade technology...

 

[Joe DeFuria]

Alright, and this would make a great discussion at Fool.com. ATI's [sic] "buisnessmodel" is of no concern to us here. Just as IBM's, NEC's, Toshiba's or ATI's (Before you and XBox) were of any concern to us.

What is a concern in the Console Forum (Like, as opposed to the Investment Forum) is preformance and features.

Um, if we're speculating on why MS chose one vendor over another, you had BETTER consider not just performance and features, but everything else I mentioned.

This being a console forum doesn't make one iota of difference on the factors for choosing avendor.

This, as I proposed, is highly linked to lithography and the manufacturing abilities of the respective comapnies. We talk of IBM's success in SOI or Toshiba's success in eDRAM because it's reflective of PS3. We talk of NEC's achievements and Nintendo for a reason.

Lol...you also talk about "the financial report of the week"

We also talked of ATI and how we don't Nintendo's part to be absolutely bleeding-edge for a reason - ATI is a small part of this.

What do you mean? You don't expect Nintendo's part to be bleeding edge, in part because ATI is involved? On what basis? That ATI can't create a bleeding edge part?

The same holds of ATI/Microsoft and nVidia. If you can't see this, then you're just wasting space and limiting people's ability to suggest their own ideas because your arguing over semantics.

I have no idea what you're getting at. It seems you are saying something like:

"ATI has a history of not providing bleeding edge console tech, so you are just wasting space if you think X-Box2 will be any different."

Honest question: is that what you are trying to say, or am I reading that improperly?

I think Archie covered the DX10 lvl question I was thinking of. But, I have an OT question - how many games on the market are built utilizing the defacto standard?

Just wondering because I've seen a bunch of nVidia's Way it's ment.. stickers and haven't seen.. well... anything for ATI. Perhaps you can PM me the answer as this is OT. Thanks in advance.

"Way it's meant" is a marketing gimmick, or don't you see that?

The difference is between which card is the primary development platform, and which card is the secondary one.

In the DX8 era, nVidia was certainly the primary one. In the DX9 era, it's moving to ATI. See Half-life and Tomb Raider....

In short: ATI does not need at this time, to improve the characteristics of their core, just performance.

I smell a 3dfx parallel.

Correct. And as I explained, it's also an nVidia DX8 parallel. The strategy itself isn't an issue....it's how it's executed and to what degree you execute it.

Well, this is factually incorrect. nVidia had a NV20, then the NV25 refresh. Then the new NV30, refresh NV35. New NV40...

Well, you are factually incorrect.. nVidia had the NV20, then the NV20ti "refresh", then the NV25 refresh, then the NV25+ AGP8x refresh. 4 cycles with the NV2x core.

ATI is basically going R300, R350, R3xx. I see a difference, but it must not be appearent to those not with a vested interest in ATI. < shrug >

Shrug. ATI has R300, then R350, then R360 (which is likely to be similar to a NV20 ti refresh). Then R420, which while it is heavily based on R3xx, rumors suggest it's a step change in performance.

We'll see when they anounce NV4x and Loci, won't we?

Yes, we will.

Right...and that's a rather large assumption to make if MS is licensing technology, and not buying graphics chips.

I'm making no definitive assumptions. Just guesses...as we all are. Or am I not allowed to make my own guesses?

No it's not when you think about it objectivly. But, you still have yet to articulate your thoughts on ATI's agreement and it's short/long-term effects on preformance, quality, et al.

REad previous post, and be happy.

No, MS will not have to do "redesigning". MS's contracted engineering team will have to take ATI's core logic design and build a chip using it.

You know this how? You'd want this why?

I'm guessing this as much as you're guessing. They would WANT to do this if MS is able to secure some foundry that ATI has little specific experience with, and has little vested interest in beyond the x-box chip.

I can't understand why you don't accept and love Natoma's life choices either.

Wow...you really have some issues, Vince...

 

[zidane1strife]

Again, I have to ask why. Why is 65nm so important for you? They certainly may want 65 nm, but why have you decided the risks outweigh the rewards...and why does Microsoft agree with you?

Well, I'd say it's pretty important considering the ps3's gphx calcs are likely to have twice the amount of transistors of a 65nm gpu devoted to them, do to being split among two chips... there's some ps3 related info, from old press releases, in my sig...

 

[Joe DeFuria]

Well, I'd say it's pretty important considering the ps3's gphx calcs are likely to have twice the amount of transistors of a 65nm gpu devoted to them, do to being split among two chips...

These chips are so fundamentally different (assuming xbox-2 is more like a traditional GPU), that comparing transistor counts isn't particularly useful in and of itself.

What target clock speeds are each of these chips, including the X-Box, suppossed to be running at?

How much "work" can each chip complete in one clock cycle? How quickly does performance "degrade" as more and more features are "turned on?"

How expensive is each solution?

Again, all else being equal of course a larger transistor budget is more beneficial. I just get an awful taste in my mouth when people start comparing transistor counts of two completely different architectures and try and draw conclusions from it.

 

[V3]

These chips are so fundamentally different (assuming xbox-2 is more like a traditional GPU), that comparing transistor counts isn't particularly useful in and of itself.

Yep considering that SCE seems going to spend 750+ million transistors just on memory. They need 65nm process more than MS.

 

[zidane1strife]

Again, all else being equal of course a larger transistor budget is more beneficial. I just get an awful taste in my mouth when people start comparing transistor counts of two completely different architectures and try and draw conclusions from it.

I know I know, but what I'm trying to say here is that having a 2-3x advantage in terms of trans., and dealing with different solutions to a similar goal/prob... The one with significantly more trans. would have to be rather lacking in order for it to be significantly outdone by a much simpler/less complex( in terms of trans.), probably slower chip.

 

[Joe DeFuria]

I know I know, but what I'm trying to say here is that having a 2-3x advantage in terms of trans., and dealing with different solutions to a similar goal/prob... The one with significantly more trans. would have to be rather lacking in order for it to be significantly outdone by a much simpler/less complex( in terms of trans.), probably slower chip.

Not necessarily true at all. You have to look at the entire architecture, (chips and memory subsystems) not just the chips. For exanple, what if x-box 2 has a 256 bit G-DDR3 memory interface, sporting 40-50 GB/sec in external memory bandwidth to 256 MB external Ram?

How does that compare to PS3 specualtion?

 

[V3]

How does that compare to PS3 specualtion?

Segmented (Though they can address each other) with the CPU and GPU each having 64 MB using 1024 bit bus.

That's all that was hinted . It is also speculated to use the new Rambus memory.

 

[Megadrive1988]

Yep considering that SCE seems going to spend 750+ million transistors just on memory. They need 65nm process more than MS.

is that for the CPU's memory alone, or the CPU+GPU ?

 

[V3]

For the CPU alone.

The GPU, even if it has less SRAM, it has 4 seperate Image cache, so it will probably have more transistors than the CPU.

This is just looking at the patent mind you

 

[Vince]

Vince, be a big boy, will you?

Boo hoo

Is this what this argument has been reduced to? You talking like this and not offering anything to advance the argument? You're the only on here not advancing or challanging the arugment on technical or similar grounds. You propose little, back-up none of it - and just respond based on known debating tactics. I disgress...

Pro: the 3rd party may be more familiar with a specific fab process than ATI, meaning that the 3rd party will have an easier time productizing the IP than ATI would.

Con: whenever you have multiple parties involved which assume some sort of responsibility, with Information flowing between them, you tend to have some more issues with managing the project and getting it done. There is a high degree of cooperation involved, and it doesn't always run as smoothly as expected.

It's about time. Your comment has several fallacies (no wonder it took this long) that I'll lay out:

3rd parties aren't better equipt than a semiconductor company like ATI/nVidia to do the back-end work on an IC, nor have I seen the sucessful synthesis of a bleeding-edge IC from a 3rd party on the same level as a cuncurrent design by nVidia or Intel or IBM. If you can find some examples, thats great.. but it's doubtful.

EEDesign.com had several articles on how the development pipeline in sub-130nm parts need a high level of integration concerning how the design is written in RTL (remember this part of the top of my head) and several other pieces on this. Especially considering each Fab, each line, is different and the relationship held by a single entity like ATI with a fab like TSMC/UMC would carry thew the entire design cycle. Thus, your comment is wrong - but if you can find proof then please do.

I think what ATI does hinges in large part on whatever Fab Microsoft has in mind to actually produce the chip. If MS is planning on TSMC or UMC, then I think ATI will get more directly involved in the actual place and route chip level design.

If MS is planning on going with a fab that ATI has little interest in for their other products, then I don't anticipate that ATI will be doing mu more than supplying software IP.

Is that OK with you?

Fine with me, the competition too I assure you. Yet, the fact still stands that having the whole design pipeline within the confinds of a single top-tier semiconductor company is prefferable and that of such companies, ATI has been less aggressive in lithography advancement than it's major competitor. You can't change history...

Now give us all a break and stop with the petty crap. Try debating with a non-combative tone, and stop being on the nVidia defensive as if you have some emotional tie to nvidia's well being.

nVidia defensive? Have you learned nothing from this argument? Lithogrophy is everything, and I already stated that nVidia is better positioned to expliot this. You haven't opposed this in anyway, shown no objective numbers, histories, or precedece to show that I'm whong (while I have).

All you can do is post how I need a life and an on the 'nVidia defensive."

Want to have a poll and ask people if I'm more nVidia biased than you are ATI biased? I think anyone keeping up with this can see that I'm semiconductor biased, as opposed to your investment.

2005 is going to be an interesting time indeed.

 

[Vince]

Um, if we're speculating on why MS chose one vendor over another, you had BETTER consider not just performance and features, but everything else I mentioned.

Should we ask how many people here want to talk in depth about the superior corperate structure of ATI that lead to their choice rather than, say, a better solution from IMGTEC or VIA? I'm guessing nobody here cares, seems like everyone's at a 3D technology site for... um... 3D technology.

Fool.com

Lol...you also talk about "the financial report of the week

Better know as a "Software Sales Report" - hardly comperable to a suppliers corperate background. And the times people have posted Finacial reports (eg. Chap and Sony's one Q loss) we tried to get it removed.

What do you mean? You don't expect Nintendo's part to be bleeding edge, in part because ATI is involved? On what basis? That ATI can't create a bleeding edge part?

It's called precedence - we've seen it in GameCube.

I have no idea what you're getting at. It seems you are saying something like:

"ATI has a history of not providing bleeding edge console tech, so you are just wasting space if you think X-Box2 will be any different."

Honest question: is that what you are trying to say, or am I reading that improperly?

Honest Answer: I merely stated precedence as seen in past console designs aswell as past IC solutions on the PC that have utilized inferior lithography methods in comperason to nVidia (who has been pushing as far back as I remember).

Unlike you Joe, who seems to have no recourse in claiming ATI's absolute superiority after one cycle (in which the competition also faultered), I like to use historical precedence for my arguments. Unfortunatly, you just forget the past that doesn't work for you and focus on what does.

"Way it's meant" is a marketing gimmick, or don't you see that?

So, it's my fault that ATI has close to Zero marketing on this? I think not.

I'm guessing this as much as you're guessing. They would WANT to do this if MS is able to secure some foundry that ATI has little specific experience with, and has little vested interest in beyond the x-box chip.

Why? It's an invitation for disaster. Odds are ATI will hand off a completed GDSII tape to Microsoft. What your proposing is worse for Microsoft as it fractures the development pipeline and is worse for ATI as their name is attached to the end result.

My favorite part:

I can't understand why you don't accept and love Natoma's life choices either.

Wow...you really have some issues, Vince...

Yes, it must suck to realise just how convolutied your thoughs are towards ATI. You fight Natoma constintly based on how his views differ from yours and he can't understand why others don't accept his views. And then you turn around and state... What was that again?

"I can't understand why people don't like ATI" - JoeDefuria.

That's sig quality material right there.... going to have to hold onto this one.

 

[Vince]

Try debating with a non-combative tone, and stop being on the nVidia defensive as if you have some emotional tie to nvidia's well being.

I can't understand why people don't like ATI - Joe DeFuria

< scratches head >

 

[Vince]

How does that compare to PS3 specualtion?

http://biz.yahoo.com/bw/030710/95679_1.html

 

[PC-Engine]

What kind of speeds are we looking at in 2005 for GDDR3? 2GHz?

 

[Joe DeFuria]

Is this what this argument has been reduced to?

As if it wasn't reduced to this 4 pages ago?

You talking like this and not offering anything to advance the argument?

(See past direct responses to you)

It's about time. Your comment has several fallacies (no wonder it took this long) that I'll lay out:

3rd parties aren't better equipt than a semiconductor company like ATI/nVidia to do the back-end work on an IC, nor have I seen the sucessful synthesis of a bleeding-edge IC from a 3rd party on the same level as a cuncurrent design by nVidia or Intel or IBM. If you can find some examples, thats great.. but it's doubtful.

See Dreamcast and Arcade PowerVR derivitives.

Fine with me, the competition too I assure you. Yet, the fact still stands that having the whole design pipeline within the confinds of a single top-tier semiconductor company is prefferable and that of such companies, ATI has been less aggressive in lithography advancement than it's major competitor. You can't change history...

Yes, let's look at history, shall we?

There's a difference between just taking a blind risk and being aggressive with lithograpgy advancement, and successfully assesing the risk, and designing accordingly.

nVidia has at best a 50/50 rate with the past 4 core technologies.

TNT: Suppossed to ship on a more advanced process at 125 Mhz. Actually shipped on a less advanced process at 90 Mhz. (I think it was 0.35u and 0.25u, but don't remember exactly).

GeForce1: Successful introduction of new core...but on old and proven 0.25u lithography. Adavnced lithogragy (0.18) not utilized until GeForce2 GTS respin.

GeForce3: By all accounts: late. 0.15 wasn't "ready" as early as they thought it would be. 0.18u GeForce2 Ultra ships instead. Related: X-Box chip, target was 300 Mhz....shipped at what 225 Mhz?

NV30: You know the story.

What you see as nVidia being "aggressive", I see as being a failure to reasonably predict the state of the advanced process. Where you see nVidia being "aggressive" with lithograpghy, I see it as an over reliance on advanced lithography, the biggest failure of course being NV30.

In the same vein, ATI has shown that it does not need the same advanced lithography to compete with nVidia. Where nVidia's CEO (and 3DLabs, for that matter) was trumpeting how we "need" 0.13u for DX9 shaders, ATI was out proving them all wrong.

Does this mean that ATI can't target a highly advanced process? Not at all. It does mean that for PC Parts, ATI chose a different route that it felt was a better financial path.

Moral of the story: it's not just what process you use, it's how you use it.

nVidia defensive? Have you learned nothing from this argument? Lithogrophy is everything, and I already stated that nVidia is better positioned to expliot this.

Read above. If the R300-NV30 contest should prove to ANYONE, no, lithograpgphy is NOT everything. It's staring you right in the face for crying out loud.

And you have not shown how nVidia is better positioned to exploit more advanced lithography. What history shows is that nVidia takes more risks in the department, sure. Delivery is a whole nother story however, as I showed above.

If nVidia was in a better position to "exploit" 0.13, then surely NV34 would be dominating RV350. These are two parts, both targeted on 0.13, both targeted at the same time frame, both targeted at the same price demographic. In a BEST case for nVidia, you might argue they are roughly on par. At worst for nVidia (general concensus), RV360 is the superior core.

How did nVidia's "better position to exploit" 0.13 help them in that race?

You haven't opposed this in anyway, shown no objective numbers, histories, or precedece to show that I'm whong (while I have).

Sigh...read above.

All you can do is post how I need a life and an on the 'nVidia defensive.

Lol! Oh, the hypocricy!

Want to have a poll and ask people if I'm more nVidia biased than you are ATI biased?

At the same time, want to have a poll about who's more childish and condecending?

I think anyone keeping up with this can see that I'm semiconductor biased, as opposed to your investment.

I think anyone keeping up with this realizes that technology and economics are both highly intertwined when choosing a vendor.

I think anyone who thinks that nVidia has a more or less "proven track record" on more advanced semiconductor processes, isn't seeing the forest through the trees.

 

[Simon F]

Sorry to everyone else for the off-topic post but...

Not to go too much off topic but he's right, "laws of physics" might be a wrong term to use but i understand what he means, planes rely on pressure created by a vortex along the wing's surface for the atmosphere to "suck up" the plane in the air. Its a totally different approach than a sailboat who just rely on its sail to catch the wind.

You clearly are thinking the way the old square-rigged sailing vessel makers thought. A sailing boat with a triangular sail and boom uses much of the same principles as an aeroplane wing. That is what helps it to sail almost into the wind. AFAIU (and I've done a bit of sailing) when tacking into the breeze, the wind combined with the battons shapes the sail it into an airfoil giving it "lift".

 

[Joe DeFuria]

Should we ask how many people here want to talk in depth about the superior corperate structure of ATI that lead to their choice rather than, say, a better solution from IMGTEC or VIA? I'm guessing nobody here cares, seems like everyone's at a 3D technology site for... um... 3D technology.

In fact, the corporate structure / business model of IMGTEC has in fact been discussed on these boards quite in depth before. Seems like everyone here has all kinds of interests related to 3D technology, the companies that supply them, and the games that result using it.

But please, pigeon-hole and discussion as you see fit.

Better know as a "Software Sales Report" - hardly comperable to a suppliers corperate background. And the times people have posted Finacial reports (eg. Chap and Sony's one Q loss) we tried to get it removed.

But you didn't apparently. Right. Moving on...

What do you mean? You don't expect Nintendo's part to be bleeding edge, in part because ATI is involved? On what basis? That ATI can't create a bleeding edge part?

It's called precedence - we've seen it in GameCube.

And we've also seen GameCube continually sell at a lower price, and most believe it also sells for a profit at a lower price.

Because one customer (Nintendo), chooses one set of requirements, and ATI meets them, doesn't mean ATI can't meet a different set of requirements.

It might be different if ATI hasn't proven with R300 that ATI can in fact, produce bleeding edge tech. In other words, there's NO evidence to suggest that ATI can't do anything other than meet the customer's needs.

Honest Answer: I merely stated precedence as seen in past console designs aswell as past IC solutions on the PC that have utilized inferior lithography methods in comperason to nVidia (who has been pushing as far back as I remember).

And I am stating that just utilizing "superior" lithography doesn't mean a hill of beans. See last post. When you need superior lithography to compete with a product on "inferior" lithogragy, that seems more like a crutch, than a feather in your cap.

Unlike you Joe, who seems to have no recourse in claiming ATI's absolute superiority after one cycle (in which the competition also faultered), I like to use historical precedence for my arguments. Unfortunatly, you just forget the past that doesn't work for you and focus on what does.

See past posts on the historical "success" of nVidia pushing lithography. And BTW, you can refrain from putting words in my mouth. ATI does indeed have the superior core in the R300 vs. NV30. I never said that gives them absolute superiority or anything like it.

But you can continue on with your tactics of trying to tear me down on statements and implications that I never made, or you can try and actually read what I write.

So, it's my fault that ATI has close to Zero marketing on this? I think not.

No, and I never said that. The point is, it IS marketing, and nothing to do with developer's choice of primary platform.

Why? It's an invitation for disaster. Odds are ATI will hand off a completed GDSII tape to Microsoft. What your proposing is worse for Microsoft as it fractures the development pipeline and is worse for ATI as their name is attached to the end result.

Sure, just throw out the risks, without looking at potential rewards. What I am proposing requires less investment and resource drain for ATI, and more flexibility for Microsoft.

Yes, it must suck to realise just how convolutied your thoughs are towards ATI.

No, it only sucks having to deal with people like you who resort to putting words in my mouth in order to try and toss an insult or two.

You fight Natoma constintly based on how his views differ from yours and he can't understand why others don't accept his views. And then you turn around and state... What was that again?

It also sucks having to deal with someone who drags a moral argument over homosexuality into a debate about console 3D technology partnerships.

And then, in a blatant display of hypocricy, imply that I have some sort of emotional interest in this above and beyond what you do? Truly laughable.

"I can't understand why people don't like ATI" - JoeDefuria.

That's sig quality material right there.... going to have to hold onto this one.

Please do. I can't understand why someone wouldn't like ATI, who is delivering superior technology, and who hasn't resorted to nearly the same level of overt PR and marketing crap that nVidia has over the past years. (Latest example being the 3DMark fiasco).

That doesn't mean I give a rat's ass what people think....I just don't understand it.

Of course, you fail to highlight the sentence that quote was in response to:

I can understand why people like ATI, who likes nVidia?

So, Vince...what do you care if people like nVidia or not?

 

[Fox5]

So how about if Microsoft gets Nvidia to fab ATi's chips?

Anyhow, Nvidia may have the superior technology, but I believe it is also currently larger, hotter, louder, and maybe even drains more power, and microsoft wants the xbox 2 not to be high end pc like in appearence.