So... what will G80/R600 be like?

[_xxx_]

I think it's time to officially begin speculating :)

I say both will have unified shaders, decoupled tex units, advanced memory controller and scheduler along the lines of R520 and at least 24 "pipes". HDR+AA and correct AF will be there as well. G80 will of course feature multiple clock domains, R600 maybe as well.

I also think we'll see a replay of this year (business-wise). G80 in late summer, R600 a few months later. I do expect ATI to have parts which are actually available this time around, but nV doing the same trick they did this year.

All just IMHO. What do you people think?

EDIT: to make it perfect, I think they'll both feature some physics calculations capabilities, too :razz:

[AlphaWolf]

If G80 has unified shaders I'd expect it to be after R600. They seemed to be fighting the notion while ATI was pushing for it.

[Matasar]

R600 supposed to hit late Q4 2006 or early 2007 ?

[Karma Police]

I think it's time to officially begin speculating :)

I say both will have unified shaders, decoupled tex units, advanced memory controller and scheduler along the lines of R520 and at least 24 "pipes". HDR+AA and correct AF will be there as well. G80 will of course feature multiple clock domains, R600 maybe as well.

I also think we'll see a replay of this year (business-wise). G80 in late summer, R600 a few months later. I do expect ATI to have parts which are actually available this time around, but nV doing the same trick they did this year.

All just IMHO. What do you people think?

Triple-slot cooling!!!!!!!

[_xxx_]

If G80 has unified shaders I'd expect it to be after R600. They seemed to be fighting the notion while ATI was pushing for it.

Nah, that's just politics like ATI downplaying SM3.0 until they had SM3.0 parts themselves, where it suddenly became the best thing since sliced bread.

[Fodder]

Are we still expecting G80 to be a multi-core (and not just duplicates) GPU?

[Kaizer]

Nah, that's just politics like ATI downplaying SM3.0 until they had SM3.0 parts themselves, where it suddenly became the best thing since sliced bread.

As far as I can remember, ATi said they wouldn't do 3.0 until it could be done right, whereas nV didn't see the usefullness of unified shaders just yet (or something of the sort). In too much of a hurry, though.

_
K

[N00b]

R600 supposed to hit late Q4 2006 or early 2007 ?Since it will most probably be a DX 10 part and DX 10 will be introduced with Vista, it will be available shortly before the Vista launch (probably at least one or two month earlier). Vista is supposed to launch mid-2006, but will probably be a bit late since Beta 2 has slipped to january or february.

[_xxx_]

As far as I can remember, ATi said they wouldn't do 3.0 until it could be done right, whereas nV didn't see the usefullness of unified shaders just yet (or something of the sort). In too much of a hurry, though.

_
K

Well yeah, it's valid for both sides. But frankly, do you think nV did SM3.0 wrong in some way? Seeing the current bunch of benchmarks, I somehow can't see nV's SM3 being useless, neither I think that unified shaders are useless.

[Ailuros]

From what it looks like unified shader cores cannot be avoided in the longrun; whether NVIDIA will arrive with such a sollution at least for it's first DX10 incarnation is another story. I personally wouldn't bet my money on it.

[Radeon600]

Aw, well I love to speculate also :)


WGF - 2.0
Unified Approach
GDDR-4 (Same Ring Bus Technology)

[DemoCoder]

I wouldn't trust what NV says in public PR as a guide to what they are doing internally. I think it's a smokescreen.

[Dave Baumann]

An analyst contaced me a little while back saying that some the Wall Street guys that are in contact with them are now of the opinion that G80 will be unified. I suspect that this could well be the case now, but I don't think it was the initial plan.

[Ailuros]

Interesting....

[Jawed]

:drool: <-insert drool smiley

Jawed

[CMAN]

I'm waiting for the ExGPU. My vision of this will be a crazy CrossFire/SLI config where it connects to a master card through an external cable (ala CrossFire) and is a self contained rendering box. External GPU! With all the one upmanship recently, I wouldn't be surprised.

Are we expecting more of a performance based upgrade in R600 and G80 or more emphasis on features?

[nAo]

An analyst contaced me a little while back saying that some the Wall Street guys that are in contact with them are now of the opinion that G80 will be unified. I suspect that this could well be the case now, but I don't think it was the initial plan. Even if we know it could mean anything we should remember nvidia has actually patented some unified shading tech..

[Matasar]

Someone said that R600 cant be released until 1 year after xbox launch because they signed some contract with MS ?

[DemoCoder]

I think it was probably in the roadmap. But I think NVidia PR is divorced from NVidia engineering.

For example, I don't believe NVidia "threw in" a 256-bit bus to the NV35 at the last moment after being "surprised" by the R300. I think they had investigated all the options they had available in terms of # of pipelines and bus size, and I think a 256-bit bus was in the internal roadmap, it was just a question of if they could deliver it in time. They probably deemed the extra complexity as making an already delayed design even further out. Once the decision was made to launch with 128-bits (and 4 ROPs) Just like ATI did with the R400/R420/R520/Xenos, they settled on what was fastest they could deliver to market, by taking baby steps. However, in NVidia's case, their decision to go with a paired down bus for the initial launch of the architecture put them at a PR disadvantage, so that's why you saw all that FUD about "we don't need a 128-bit bus", and then 1 quarter later, they launched a 256-bit bus.

I think NV has been badmouthing unified for fear that ATI would beat them to market with a unified architecture, but I don't think that internally they are forgoing a unified architecture. I think it's in the roadmap, has always been there, the only difference is, the degree to which the roadmap has been compressed by pressure from ATI and the hype around unified.

[obobski]

ever wonder what nVidia did/does with all the 3dfx tech they now own? i've often wondered how different nVidia's products would be w/o 3dfx, if they didn't own all the IP

idk, just thinking maybe we might see a surfacing of some un-used/un-shown 3dfx technology, combined with more engineering, sort of like nVidia seriously showing their teeth, not just competition, but ATi has enough momentum to maintain competition imo

still, i'm wondering what ever happened to tiled gfx (ignoring CrossFire's use of tiles)
the Kyro II used it, 3dfx's roadmap showed a product like 4-5 gens out from Rampage, "Mojo" to use tiled graphics, but nothing else has been published on that...and btw, for what it had in terms of hardware Kyro II was amazing...


but given 3dfx's use of tiles for memory, possibly a rendering solution wherein a small portion of the screen has it's own RAM address range, and portion of rendering power, then again who knows?

as far as G80/R600...i'm thinking GDDR4 for ATi, and XDR or XDR II for G80, with a bias towards XDR II, along with DX10 support, maybe higher FP precision, maybe not, better CrossFire modes for ATi, and possibly version of SLI that allow non-50/50 SMP (ASMP rendering, but w/o tiles?)

[Joe DeFuria]

I think it's in the roadmap, has always been there, the only difference is, the degree to which the roadmap has been compressed by pressure from ATI and the hype around unified.

I think it's a little different...I believe the pressure for NV to move to unified "earlier than they would have preferred" is likely due to the final form that DX10 is taking. In other words, DX10 likely pushed chip economics toward a unified solution. ATI was probably more successful in lobbying MS to get DX10 more "unfied friendly", than nVidia was in keeping DX10 "unified averse."

On a related note, I'm sticking to my guns that Vista won't be available to 2007...and that's when the first G80 (assuming unified) / R600 parts will also appear. Nothing but a pure hunch on that.

[DemoCoder]

Even in DX9, VS3.0 and PS3.0 instruction sets were converging. I think regardless of DX10, chip economics are already pushing for a unified approach. Putting so many vertex shaders on the die wastes space (plus now we have geometry shaders), and ILP has been proven not to scale, so if you're going to put as many ALUs on the chip as budget will allow, you need TLP, and TLP+huge numbers of ALUs cry out for unified. I don't think the NV engineers are dumbasses who aren't familar with all of the various architectural advantages and designs out there, just like Intel was blind to dual core, or TLP, they just tried to squeeze as much out of ILP as they possibly could instead of abandoning ship earlier.

I think NVidia PR operates on the "if a competitor is going to beat us to market with X, than downplay important of X" algorithm.

[Joe DeFuria]

Even in DX9, VS3.0 and PS3.0 instruction sets were converging.

Right...just not enough in the PC space to push vendors toward a unified hardware environment.

I don't think the NV engineers are dumbasses who aren't familar with all of the various architectural advantages and designs out there...

?? Neither do I...I certainly was not implying that.

I think NVidia PR operates on the "if a competitor is going to beat us to market with X, than downplay important of X" algorithm.

Sure...as does everyone.

[JoshMST]

I think NVidia PR operates on the "if a competitor is going to beat us to market with X, than downplay important of X" algorithm.

NVIDIA isn't the only one... I remember Intel saying "We don't see a need for 64 bits on the desktop anytime soon" then Yamhill got leaked and now Pentium 4's are X86-64. I think it is pretty common everywhere in every marketplace.

[Dave Baumann]

I think it was probably in the roadmap. But I think NVidia PR is divorced from NVidia engineering.

Well, I can't ever recall NV's PR making any comment about it. The commentry and reaction has stemmed from Kirk and he's not divorced from engineering.

[Geo]

Are we still expecting G80 to be a multi-core (and not just duplicates) GPU?

:shock: Don't take my sig as the Conventional Wisdom! It's labeled "Wild-eyed Flyer" for a reason, y'know!

[mapel110]

:shock: Don't take my sig as the Conventional Wisdom! It's labeled "Wild-eyed Flyer" for a reason, y'know!
What means Wild-eyed?

R600
~450-500 Mio Transistors
Unifeid Architekture
256bit Memory Interface
FP16 Filtering reloaded
Vertextexturing reloaded
No new AA oder AF Modes

G80
nealy same transistorcount
Not unified, 32 Pipes
256bit Memory Interface
maybe better MultisamplingAA and no better AF. I think the design is too far in the production to make this change. (dont think, nvidia knew about R520 HQ-AF early enough)

That are my thoughts till now.

[Turtle 1]

Thats correct Dave when nvidia engineers were talking about R600 Vs. G80 the nvidia engineers were saying no need for unified drivers yet. So now all of the sudden nvidia's G80 is a unified Arch. I don't believe that for 1 second.
If its true than the G80:cool will arrive way after the R600 at least 6months. :cool:

[SugarCoat]

there is no G80 ;)

[Geo]

What means Wild-eyed?



Wild-eyed = Well, wild eyes. Unshaven. Dishelveled hair. Possibly a little drool. Think the type of guy you see on the corner in some downtowns preaching animatedly to the air as disinterested folks walk by.
Flyer = highly speculative. Think internet day trader during the .com boom.

I may have to rethink this whole sig business. :lol:

[Jawed]

there is no G80 ;)
Ah yes, the infamous NV50 (G80) has been cancelled rumour.

Jawed

[DemoCoder]

Well, I can't ever recall NV's PR making any comment about it. The commentry and reaction has stemmed from Kirk and he's not divorced from engineering.

Divorced in the sense that what they say in public has no bearing as to what is in their roadmap. Kirk was out there saying 256-bit buses weren't needed, yet the NV35 arrived 1 quarter later with 256-bit. So are we to believe that Kirk had an epiphany the day after he said that, and NVidia magically rushed a 256-bit NV35 architecture into production in such a short period of time, or that NVidia had planned on 256-bit bus all along, and Kirk was just broadcasting NVidia PR talking points for damage control? I rather think that the NV35 was what the NV30 was originally supposed to be, but because of time constraints, delays, problems with their manufacturing process, the NV30 hackjob became neccessary.

NVidia has some patents on unified technology. That means atleast one of their teams has been working on it either at the architectural design or implementation stage. Thus, Kirk badmouthing unified shading I think has little bearing as to how much internal effort Nvidia is placing on it in their roadmap. In fact, given the long timespan unified is taking to get to market, it would be bad strategy for PR comments to accurately reflect what you are working on.

I don't believe in a late-stage alteration of the G80 to be unified. Unified requires large architectural changes. You don't just take a traditional design like 32-pipe G70 and stuff a unified architecture into it. At this stage, changing a non-unified G80 to be unified would be like throwing away your design and starting over. This would be pretty bad from the standpoint of trying to hit the Vista-timeline unless you are 100% sure of a Vista delay.

So, if the G80 does turn out to be unified and is delivered in 2006, then I believe either one of two situations exist: Either a) the G80 always was going unified (or the decision to change was made a long time ago) or b) the G80 has been scrapped, and a G90 unified core has been renamed to be the G80.

Scenario c) which is NVidia burned the midnight oil to refactor the G80 into a unified design two to four quarters before launch, I think is highly unlikely. Especially since they also have ot deal with a process change as well.

[AlphaWolf]

Scenario c) which is NVidia burned the midnight oil to refactor the G80 into a unified design two to four quarters before launch, I think is highly unlikely. Especially since they also have ot deal with a process change as well.

That would depend on when you expect it and what process you expect it to be.

[hkultala]

I expect R600 to be quite like "doubled xenos",
16 pipelines/ROPs, 96 shader ALU's ,
32 + 32 TMU's.

though not sure about the eDRAM; there is the problem of having enough it for big resolutions without having to draw the screen in many parts ( which would require storing the vertex information etc.. )

[SugarCoat]

Ah yes, the infamous NV50 (G80) has been cancelled rumour.

Jawed


well i said this before but i guess i can do it again.

Nvidia is against Unified architecture citing that they wont use it until it shows benefit, and will problably continue to operate on highly programmable pipelines instead. This can benefit both in cost and dificulty to produce as well as overall speed (perhaps). I do not think they will launch their DX10 compatable products without a Unified part however. I expect them to very much launch either/or meduim/low end parts based on unified architecture. Both to get a feel for production and to aid in driver maturation for a flagship Unified part which may or may not come sometime in 2007 so that they arent releasing products that prove to be immature due to drivers. Lets face it, a retail product release helps both companies mature drivers far more then in house driver production. People can say company X has had so much time to do drivers that when it launches it will already be top notch, but i have never seen that as the case. The best performance drivers seem to come between 3-6 months after a launch and driver performance imrovments, both insignificant, and significant, continue through the products cycle.

The NV50 core has been in production for Vista and DX10 for quite a long while. Almost 2 years by my judgment. Both companies have had access to the ever changing DX10 API for well over a year. I see the "G" code named cores, as the stop gap between the NV40 and NV50. Think of it as hey look we've had this core on the burner for awhile but Microsoft keeps changing things as well as pushing release dates, we need to do something about our product inbetween then and now or we'll be infringing on codenames. Obviously Nvidia wont change their time table or core succession, so enter the G70 and departure for the time being of the NV codename. If there is infact a G80, i very much suspect it to be launched early or mid next year, and most certainly prior to their first DX10 part. And as soon as that DX10 part is introduced, i think we'll see Nvidia go back to NV codenames. This is literally, the best and most logical reason i can come up with for any reason of their code name departure from what they have been using for the last 5+ years.

They will keep riding this tech, NV40 derivative, modifying it through-out, keeping essentially the same SM3.0 technologies, until Vista launches (now late 2006/early 2007). Once that happens, we should see a very matured and substantially impressive/complex core, technology wise from them, as i think they have been working on it(NV50) for quite a long time.

R600 im sure will be its own wonder. Although even in its launch time table i dont think it will use imbedded DRam. Costs too much and will cause problems in games. I believe it takes specific coding to use it.

Thats my theory in a nutshell.

[Turtle 1]

Sounds to me like you got it pretty good . the EDram though thats the one I have been thinking about for the R600. will they or won't they?:roll:

[Fodder]

:shock: Don't take my sig as the Conventional Wisdom! It's labeled "Wild-eyed Flyer" for a reason, y'know!I'm referring to an old rumour; there was talk a while back of pixel and vertex ops being separated onto different cores, though I guess unified shaders neuter that one. :)
I don't believe in a late-stage alteration of the G80 to be unified.Perhaps, like NV40, there are two G80s. :lol:
This is literally, the best and most logical reason i can come up with for any reason of their code name departure from what they have been using for the last 5+ years.I figure the dragging out of NV4x simply gave them an appropriate opportunity to re-jig their naming scheme. G for graphics, C for chipsets ... what are the codenumbers for their handheld parts?

[DegustatoR]

Wasn't Kirk talking about now? As if unified approach isn't optimal for now? When did he say this anyway? A year ago?

I think it's possible that he meant DX9 SM3 generation.

[AlphaWolf]

Wasn't Kirk talking about now? As if unified approach isn't optimal for now? When did he say this anyway? A year ago?

I think it's possible that he meant DX9 SM3 generation.

"We will do a unified architecture in hardware when it makes sense. When it's possible to make the hardware work faster unified, then of course we will. It will be easier to build in the future, but for the meantime, there's plenty of mileage left in this architecture."

Taken from bit-tech (http://www.bit-tech.net/bits/2005/07/11/nvidia_rsx_interview/4.html) in july of this year.

[DegustatoR]

Well, i think it's pretty clear he was talking about DX9 SM3 and NV4x/G7x specifically. And he's even saying that "It (unified architecture) will be easier to build in the future".

[Dave Baumann]

I think this (http://www.extremetech.com/article2/0,1697,1745060,00.asp) is the one pertinent to future architectures.

[Geo]

Wasn't Kirk talking about now? As if unified approach isn't optimal for now? When did he say this anyway? A year ago?

I think it's possible that he meant DX9 SM3 generation.

This doesn't pass the giggle test for me tho. Context is all, and that context would be "duh!". Everybody and his brother knows you are talking about Vista and dx10 for that conversation.

I'd find intentional misdirection an easier to swallow answer than "oh, you didn't mean SM3?".

[AlphaWolf]

This doesn't pass the giggle test for me tho. Context is all, and that context would be "duh!". Everybody and his brother knows you are talking about Vista and dx10 for that conversation.

I'd find intentional misdirection an easier to swallow answer than "oh, you didn't mean SM3?".

Well the question bit-tech asked before the Kirk quote I posted was, "So what about the future?"

I guess you could question how far in the future he took that to mean, but I would think he meant beyond G7x.

[psurge]

You could argue that Kirks statement was more about RSX vs. Xenos than upcoming PC products though... (not that I have any idea of what they are going to end up producing)

[Turtle 1]

You could argue that Kirks statement was more about RSX vs. Xenos than upcoming PC products though... (not that I have any idea of what they are going to end up producing)
I agree with you on that 100%

[DemoCoder]

I think this (http://www.extremetech.com/article2/0,1697,1745060,00.asp) is the one pertinent to future architectures.

I find it hard to believe that Kirk is unaware that texture units may be decoupled from pixel shading ALUs and that while a pixel ALU may be doing vertex work, it's texture units won't neccessarily be "idle".

I could believe that specialized geometry shading units would still be more efficient than reusing pixel shader ALUs due to memory access pattern differences.

[_xxx_]

ever wonder what nVidia did/does with all the 3dfx tech they now own?

Threw it away since it's 5 year old tech now? Everything from 3dfx IP is pretty outdated nowadays.

[Kaotik]

Threw it away since it's 5 year old tech now? Everything from 3dfx IP is pretty outdated nowadays.

I think the real question is - did they ever really use it (okay, they did, at least IIRC it was mentioned somewhere that they used some video engine they got on the 3dfx tech asset buyout) - or did they ever really started any research for future based on 3dfx tech

[Reverend]

Well, I can't ever recall NV's PR making any comment about it. The commentry and reaction has stemmed from Kirk and he's not divorced from engineering.
No he isn't but comments by him for the public are basically NV PR-approved. Or, in better words, has to be.

As for the original topic, these two products will be what Vista wants. Know that (with finality) and we have the answer :) . I will start another thread regarding an interesting related and side topic, that of unified shaders (I have a lot to say about this!).

[Reverend]

I think the real question is - did they ever really use it (okay, they did, at least IIRC it was mentioned somewhere that they used some video engine they got on the 3dfx tech asset buyout) - or did they ever really started any research for future based on 3dfx tech
Who cares. Compare then (in 3dfx's time) and now, those tech are basically outdated (by virtue of other technologies beyond those in 3D).

And you're not talking about Gigapixel are you? Because that isn't "3dfx tech" strictly speaking.

[wireframe]

I will start another thread regarding an interesting related and side topic, that of unified shaders (I have a lot to say about this!).
I hope this will include the differentiation between a unified shader API (software) and a unified shader architecture (hardware). I have been itching to have this clarified to me.

[Junkstyle]

Nvidia has to shrink the die. ATI already took the hit and it was nearly a total disaster with the R520. This cant be an easy thing so I'm guessing Nvidia will/could have similar problems.

[nAo]

According PS3 public roadmap at this time NVIDIA has already produced a high end GPU using a 90 nm process (from a fab they never used before).

[_xxx_]

I don't believe in a late-stage alteration of the G80 to be unified.

What makes you think that G80 is the same G80 that was there a few months ago? NV already switched names before, that means nothing IMHO.

Just saying, what we will eventually know as G80 may as well have been, dunno, G87.5 untill yesterday.

[Jawed]

I think this (http://www.extremetech.com/article2/0,1697,1745060,00.asp) is the one pertinent to future architectures.
The key thing, I think, is that he "bottles-up" a shader engine (pixel or vertex) with the features associated with either, e.g. rasterisation, interpolation.

If you separate these kinds of components and keep the shader engine "pure" as just an "ALU engine" then his argument falls on its face.

Well, that's how it seems to me. The block diagram for Xenos seems to back this up.

http://pcweb.mycom.co.jp/articles/2005/09/09/cedec1/images/012l.jpg

The PowerVR SGX also makes a mockery of his stance.

Jawed

[Jawed]

I find it hard to believe that Kirk is unaware that texture units may be decoupled from pixel shading ALUs and that while a pixel ALU may be doing vertex work, it's texture units won't neccessarily be "idle".

I could believe that specialized geometry shading units would still be more efficient than reusing pixel shader ALUs due to memory access pattern differences.
I think NVidia's backwardness on this whole topic stems from a lack of foresight on scheduling.

The Xenos scheduler is not a trivial bit of gear. But with it, the whole USA falls into place. Without it, it just looks like a minefield.

The argument over whether a GS should be a dedicated piece of hardware or can re-use the shader engine (e.g. in collaboration with a primitive assembler) is certainly ripe for discussion...

Jawed

[satein]

I think NVidia's backwardness on this whole topic stems from a lack of foresight on scheduling.

The Xenos scheduler is not a trivial bit of gear. But with it, the whole USA falls into place. Without it, it just looks like a minefield.

The argument over whether a GS should be a dedicated piece of hardware or can re-use the shader engine (e.g. in collaboration with a primitive assembler) is certainly ripe for discussion...

Jawed
Oop... it's a nice short wording :cool: ... USA... Unified Shader Architectures, not United State of America. I knew now that ATi think big enough :eek:.
Sorry for OT.

[wireframe]

I think NVidia's backwardness on this whole topic stems from a lack of foresight on scheduling.

I find that very hard to believe. The "backwardness on the whole topic," that is. A unified shader architecture is quite logical. It is not so much a performance enhancer as maintaining scaling feasibility. You can bet your last quid that scheduling would be the hot topic. I think scheduling is the hot topic even without a USA.

Scheduling on graphics hardware is an obvious place for improvement as programmability grows. However, it's not like these boys are sailing into unchartered territory. They have decades of research in the CPU field to draw from. I find it very unlikely that they will come up with something startlingly new.

Perhaps your love for Xenos/C1 has blinded you?

[Jawed]

Yes, definitely.

Xenos is the bees-knees. Where's that :heart: smiley?

Jawed

[nAo]

Don't worry, there's no need for a new smiley, we already knew you're in love :wink:

[CMAN]

Nvidia has to shrink the die. ATI already took the hit and it was nearly a total disaster with the R520. This cant be an easy thing so I'm guessing Nvidia will/could have similar problems.

ATI didn't have a problem with the die shrink they had a problem with the library they were using to design the chip.

[DemoCoder]

Was that the same thing NV claimed with the NV30? Hopefully the libraries are debugged now. :)

[nAo]

But everytime you switch process you have new libraries to debug :razz:

[Megadrive1988]

well i said this before but i guess i can do it again.

Nvidia is against Unified architecture citing that they wont use it until it shows benefit, and will problably continue to operate on highly programmable pipelines instead. This can benefit both in cost and dificulty to produce as well as overall speed (perhaps). I do not think they will launch their DX10 compatable products without a Unified part however. I expect them to very much launch either/or meduim/low end parts based on unified architecture. Both to get a feel for production and to aid in driver maturation for a flagship Unified part which may or may not come sometime in 2007 so that they arent releasing products that prove to be immature due to drivers. Lets face it, a retail product release helps both companies mature drivers far more then in house driver production. People can say company X has had so much time to do drivers that when it launches it will already be top notch, but i have never seen that as the case. The best performance drivers seem to come between 3-6 months after a launch and driver performance imrovments, both insignificant, and significant, continue through the products cycle.

The NV50 core has been in production for Vista and DX10 for quite a long while. Almost 2 years by my judgment. Both companies have had access to the ever changing DX10 API for well over a year. I see the "G" code named cores, as the stop gap between the NV40 and NV50. Think of it as hey look we've had this core on the burner for awhile but Microsoft keeps changing things as well as pushing release dates, we need to do something about our product inbetween then and now or we'll be infringing on codenames. Obviously Nvidia wont change their time table or core succession, so enter the G70 and departure for the time being of the NV codename. If there is infact a G80, i very much suspect it to be launched early or mid next year, and most certainly prior to their first DX10 part. And as soon as that DX10 part is introduced, i think we'll see Nvidia go back to NV codenames. This is literally, the best and most logical reason i can come up with for any reason of their code name departure from what they have been using for the last 5+ years.

They will keep riding this tech, NV40 derivative, modifying it through-out, keeping essentially the same SM3.0 technologies, until Vista launches (now late 2006/early 2007). Once that happens, we should see a very matured and substantially impressive/complex core, technology wise from them, as i think they have been working on it(NV50) for quite a long time.

R600 im sure will be its own wonder. Although even in its launch time table i dont think it will use imbedded DRam. Costs too much and will cause problems in games. I believe it takes specific coding to use it.

Thats my theory in a nutshell.


so you're thinking the G80 is not the NV50 as many have thought, but rather, G80 is another NV4x architecture like G70. I've often thought that might be the case.

[Chalnoth]

so you're thinking the G80 is not the NV50 as many have thought, but rather, G80 is another NV4x architecture like G70. I've often thought that might be the case.
That's possible. It might be somewhat better to talk about the NV50 vs. R600, then. A possible 90nm version of the G70 may be called the G80, as opposed to G75.

I fully expect to see the NV50 (or its equivalent) to be released late next year, at about the same time as Vista. ATI will likely do something similar.

But from nVidia's statements, it currently seems very unlikely that the NV50 will be a unified architecture. I do, however, have high hopes that it will have good branching performance, MSAA on FP16 framebuffers, higher-quality anisotropic filtering, and, of course, full SM4 compliance.

[DegustatoR]

I think this (http://www.extremetech.com/article2/0,1697,1745060,00.asp) is the one pertinent to future architectures.
Bingo! That's what i remember as being a year old :)

This doesn't pass the giggle test for me tho. Context is all, and that context would be "duh!". Everybody and his brother knows you are talking about Vista and dx10 for that conversation.

I'd find intentional misdirection an easier to swallow answer than "oh, you didn't mean SM3?".
Well....

Unified Shader architectures

...skip...

"Debating unified against separate shader architecture is not really the important question. The strategy is simply to make the vertex and pixel pipelines go fast. The tactic is how you build an architecture to execute that strategy. We're just trying to work out what is the most efficient way.

"It's far harder to design a unified processor - it has to do, by design, twice as much. Another word for 'unified' is 'shared', and another word for 'shared' is 'competing'. It's a challenge to create a chip that does load balancing and performance prediction. It's extremely important, especially in a console architecture, for the performance to be predicable. With all that balancing, it's difficult to make the performance predictable. I've even heard that some developers dislike the unified pipe, and will be handling vertex pipeline calculations on the Xbox 360's triple-core CPU."

"Right now, I think the 7800 is doing pretty well for a discrete architecture?

So what about the future?

"We will do a unified architecture in hardware when it makes sense. When it's possible to make the hardware work faster unified, then of course we will. It will be easier to build in the future, but for the meantime, there's plenty of mileage left in this architecture."

To me it looks like he was talking about present implementations of USA :) versus their present architecture. He's clearly saying that "7800 is doing pretty well" and then continues with "for the meantime, there's plenty of mileage left in this architecture" (RSX, G70-512M, G7x, 90nm etc -- there are pretty much of future in G7x even now) and he says "We will do a unified architecture in hardware when it makes sense".

When will it make sense? When Vista comes out. When ATI will have USA in the PC scene. So IMO G8x is very possibly a USA right from the beginning of development.

[Geo]

I hope this will include the differentiation between a unified shader API (software) and a unified shader architecture (hardware). I have been itching to have this clarified to me.

Really? What's unclear about "the differentiation" between them in your mind?

Edit: Tho I'm interested to hear Rev have a go at it and such other technical points he'd like to thrown down on in this area.

[Chalnoth]

When will it make sense? When Vista comes out. When ATI will have USA in the PC scene. So IMO G8x is very possibly a USA right from the beginning of development.
Not necessarily. As David Kirk noted, it is an implementation detail. It's performance in real games that is going to be important.

Part of me really hopes that nVidia will go for a unified architecture, just because the idea of a unified architecture is so simple and beautiful. Part of me doesn't just so that we can see a good showdown between a more traditional architecture and a unified one.

[Pete]

According PS3 public roadmap at this time NVIDIA has already produced a high end GPU using a 90 nm process (from a fab they never used before).Is that something other than RSX, and how will that translate to their usual PC-space fabs?

[Geo]

Is that something other than RSX, and how will that translate to their usual PC-space fabs?

Well, "PS3 roadmap" might suggest the answer to the first. . . Or are you being as subtle as a Yendi on that one? (blatant digi shoutout).

[Pete]

Heh, OK, assuming it's RSX, I thought different fabs have different libraries or something, so I'm honestly curious how NV producing RSX at 90nm translates into future parts on different fabs. At this point, ATI seems to be a bit ahead of the 90nm game

NV has had the option to keep their 90nm work quiet, considering how well the GF6 and now GF7 are doing, so they may surprise me by jumping in without missing a beat.

[Skrying]

I personally think G80 will be very similair to G70 at the hardware level, but have things that make it appear and function similair to a USA chip in Vista/whatever software is being used. Is that even possible?

I think R600 is fully USA though. It'll be a Xenos and R580 hybrid, that's mainly the ring bus memory controller from my point of view. With lots of tweaks and be fully ready for whatever feature set is the main one at the time and possible a little bit in the future.

[Chalnoth]

SM4 still has vertex and pixel shaders. The only difference is that they're more capable and the instruction sets are the same. This lends itself to a unified architecture, but doesn't necessitate one.

[Geo]

As for the original topic, these two products will be what Vista wants. Know that (with finality) and we have the answer :) . I will start another thread regarding an interesting related and side topic, that of unified shaders (I have a lot to say about this!).

Well? Inquiring minds want to know!

[aaronspink]

Well, I can't ever recall NV's PR making any comment about it. The commentry and reaction has stemmed from Kirk and he's not divorced from engineering.

I have severe doubts that Mr. Kirk has written or looked at a single line of RTL within the last 3-4 years. He's in management, regardless of what his title out PR persona is.

The rules are pretty simple, if you have more then 1000 people, and you are talking to the public, then you are PR.

Aaron Spink
speaking for myself inc.

[SugarCoat]

That's possible. It might be somewhat better to talk about the NV50 vs. R600, then. A possible 90nm version of the G70 may be called the G80, as opposed to G75.

I fully expect to see the NV50 (or its equivalent) to be released late next year, at about the same time as Vista. ATI will likely do something similar.

But from nVidia's statements, it currently seems very unlikely that the NV50 will be a unified architecture. I do, however, have high hopes that it will have good branching performance, MSAA on FP16 framebuffers, higher-quality anisotropic filtering, and, of course, full SM4 compliance.

Considering the age of the actual G70 chip i tend to believe that their next flagship product with significant improvments will be the G80. Anything relating to the G7x series should be minorly modified or on a new process as well as mid-low range series. However, combinging my theory with popular theory creates a strange launch schedule unlike what we've seen before if Vista launches on time. 3 high end products between january 06 and january 07.

G7x die shrink with speed increases, which i dont think will ever exist. They'll shrink the die if they feel they have enough time to benefit cost wise from switching production to 90nm. Although i must use harsh fact and point to how fast they shut down NV40 production do to cost and the G70 supersession. So i dont expect to see this core on 90nm unless it launches before christmas time. But architecture differences as well as 30-40% higher clocks to that? I think thats pushing the realistic spectrum. Dont forget both companies are more then willing to delay launching to stock product for hard launching in quantity. That means both the R580 and the 90nm G7x (assuming its 90nm, had its architecture changed significantly, and has much higher clocks, have to be in full swing now or very soon if they're going to launch as early as people keep saying (early next year?). Something which i just dont see. I think we'll know alot more about plans in Febuary.

G80 with significant improvments to the architecture, 90nm, what everyone expects from the G7x. Still remaining SM3.0. I'd expect this to be the real R580 competitor for spring/early summer. I'd also expect significant advanced shader enhancments from Nvidia here as well although perhaps not quite up to par with the R580.

NV50 speculate what you will, perhaps 80nm, real monster fully DX10 compatable. Winter launch time frame just before holidays. IF MS can stick to a schedule. I'd expect many improvments to come like free AA/AF from both ATI and Nvidia by the time the NV50 and R600 are launched.





I personally think G80 will be very similair to G70 at the hardware level, but have things that make it appear and function similair to a USA chip in Vista/whatever software is being used. Is that even possible?

I think R600 is fully USA though. It'll be a Xenos and R580 hybrid, that's mainly the ring bus memory controller from my point of view. With lots of tweaks and be fully ready for whatever feature set is the main one at the time and possible a little bit in the future.


The new memory controller that we see on the R520 is beyond its time. I think its fair to actually think that the memory controller was designed exactly for cores like the R600 and simply, well not so simply, implimented into the R520 as well. This memory controller is made to last well into GDDR4. And an ATI employee recently stated that we'd see some serious performance advantages with this memory controller once they can mass produce cards with GDDR4 which i believe it was really designed for.

[Chalnoth]

Considering the age of the actual G70 chip i tend to believe that their next flagship product with significant improvments will be the G80. Anything relating to the G7x series should be minorly modified or on a new process as well as mid-low range series.
The whole reason for the post you are replying to is that a part named "G80" may actually be G70-based. With a new naming scheme comes uncertainty in how future products are named.

[CarstenS]

My guess is that if NV is keeping to re-using codenames like not calling the 7800GTX-512 an G75 and the like, they'd have enough numbers left until 80 to really keep the G80-label for their first WGF2.0-/USA-GPU. I don't dare to speculate if these two possibilities will coincide or if their first WGF2.0-GPU is another Chip than their first USA-GPU.

I guess, they'll start with trying to decouple the TMU and eventually the texture adress generator from the ALU. At least that would make the most sense considering their already vastly superior math-power and their vastly inferior efficiency using that advantage in existing games, which still heavily rely on texture-fetches and high-quality texture filtering.

[Chalnoth]

The label "G70" was clearly chosen to coincide with the GeForce7 series. If nVidia refreshes at least some of its lineup in the Spring on 90nm, what's to prevent them from calling these parts the GeForce8 series? That would likely coincide with some G80 name (and G8x derivatives). There may, after all, be a performance boost to warrant such a name change.

[Maintank]

If G80 has unified shaders I'd expect it to be after R600. They seemed to be fighting the notion while ATI was pushing for it.

Doesnt DX10 requires a unified design?

[Maintank]

well i said this before but i guess i can do it again.

Nvidia is against Unified architecture citing that they wont use it until it shows benefit, and will problably continue to operate on highly programmable pipelines instead. This can benefit both in cost and dificulty to produce as well as overall speed (perhaps). I do not think they will launch their DX10 compatable products without a Unified part however. I expect them to very much launch either/or meduim/low end parts based on unified architecture. Both to get a feel for production and to aid in driver maturation for a flagship Unified part which may or may not come sometime in 2007 so that they arent releasing products that prove to be immature due to drivers. Lets face it, a retail product release helps both companies mature drivers far more then in house driver production. People can say company X has had so much time to do drivers that when it launches it will already be top notch, but i have never seen that as the case. The best performance drivers seem to come between 3-6 months after a launch and driver performance imrovments, both insignificant, and significant, continue through the products cycle.

The NV50 core has been in production for Vista and DX10 for quite a long while. Almost 2 years by my judgment. Both companies have had access to the ever changing DX10 API for well over a year. I see the "G" code named cores, as the stop gap between the NV40 and NV50. Think of it as hey look we've had this core on the burner for awhile but Microsoft keeps changing things as well as pushing release dates, we need to do something about our product inbetween then and now or we'll be infringing on codenames. Obviously Nvidia wont change their time table or core succession, so enter the G70 and departure for the time being of the NV codename. If there is infact a G80, i very much suspect it to be launched early or mid next year, and most certainly prior to their first DX10 part. And as soon as that DX10 part is introduced, i think we'll see Nvidia go back to NV codenames. This is literally, the best and most logical reason i can come up with for any reason of their code name departure from what they have been using for the last 5+ years.

They will keep riding this tech, NV40 derivative, modifying it through-out, keeping essentially the same SM3.0 technologies, until Vista launches (now late 2006/early 2007). Once that happens, we should see a very matured and substantially impressive/complex core, technology wise from them, as i think they have been working on it(NV50) for quite a long time.

R600 im sure will be its own wonder. Although even in its launch time table i dont think it will use imbedded DRam. Costs too much and will cause problems in games. I believe it takes specific coding to use it.

Thats my theory in a nutshell.

That is a very interesting thoery indeed. The G70 isnt much different than the NV40 is it? It is just 8 more pixel pipes and 2 more vertex shaders, high clocks, and faster memory?

We have heard about dual core solutions from Nvidia that sound like they are based on the G70 and of course SLI.

Nvidia could just be positioning themselves for the interim until Vista is near release by adding more pipelines, high clocks, more memory, or even dual core solutions.

Once DX10 shows up they will have their unified GPu ready to go.

[nAo]

Doesnt DX10 requires a unified design? No, it doesn't require that. AFAIK it requires to have the same features set on fragment and vertex shaders, but you can have that even with a non unified design.

[DegustatoR]

G70 is because there were no more NV4x names left -- NV40,41,42,43,44,45,48. So they have only 46,47 and 49. 47 is G70. We know that there are G71,72,73 -- that's already one codenumber more than available in NV4x. Then there may be die shrinks of NV4x/G70 to 90nm and new NV4x-cores before NV50 which i believe is G80 now.

[Monty]

im sure i heard a while back that vista wasnt going to be launched with wgf2 but with wgf1.

if that were true then wgf2 would be released in 2007,

so g80 and r600 may be further away than expected and all were going to get in 2006 in upgrades of current architecture,

this would give r520 more time in the market before r580 is released.

[Fodder]

G70 is because there were no more NV4x names left -- NV40,41,42,43,44,45,48. So they have only 46,47 and 49.Did NV48 ever show up? I was under the impression it was morphed into G72..

[Graham]

Ok, here is my competly far-fetched theory on R600.

Please feel free to rip it to shreds and proove it's utterly impossible. I would appreciate this actually :-) - as I have little knowlege of the techincal production side of graphics chips afterall.


Anywho.

Looking at what ati have done with both r500 and r520, then applying these together (but far moreso) to r600, this is what I come up with.

what interested me the most about r500 was the separation of cache/low level fixed pixel functions from the mem controller and shaders. This could be taken further.
what interested me the most about r520, of course, was the new memory controller.

Combine the two. But go a bit further.

Currently, the memory controller on r520 circles the internals of the chip (from what I've read at least - although the die shots suggest otherwise). Now I was thinking, if you add pipelines, that should increase the die size. Therefore the memory bus gets bigger, therefore slower and requring more die... So. Why not actually seperate the memory controller from the shader logic. The memory controller + scheduler + ring bus can sit on it's own (far smaller) chip... Ideally this will speed up the ring.
At each node on the ring, use a similar interconnect found on the r500 (between chips), instead, connect this to a smaller daughter die.. This small chip could, say, contain 8 alus, 8 general purpous memory read/write units, 8 texture lookup units, 8 z compare units, small bit of cache, etc etc etc.
If the ring bus then has 8 nodes, there can be 8 of these smaller chips placed around the outside of the controller (forming a square 3x3 chips). physically they could be closer than the r520 daughter die, and they have the advantage of not sharing the same peice of silicon therefore hopfully significantly reducing yeild problems (and cost). Plus, of course, creating scaled back versions is far easier, and doesn't require using the full die from a more expensive product. etc etc. I know it probably wouldn't be as efficient, but surely producing 9 working 50 million transistor chips is *far* easier than producing one working 450 million transistor chip?

yeah. Sure it won't happen, but thats what I see as being a good idea.

feel free to shoot me down.

etc.

[Chalnoth]

The problem with that idea is that the memory bus needs to be able to get data to/from the pixel processing units.

[Ailuros]

Threw it away since it's 5 year old tech now? Everything from 3dfx IP is pretty outdated nowadays.

There are quite a few tidbits and past experience that has been used since the NV25 in their products that came from the former 3dfx and GP patent portofolio.

[Ailuros]

Not necessarily. As David Kirk noted, it is an implementation detail. It's performance in real games that is going to be important.

Part of me really hopes that nVidia will go for a unified architecture, just because the idea of a unified architecture is so simple and beautiful. Part of me doesn't just so that we can see a good showdown between a more traditional architecture and a unified one.

The benefits IMO stretch way beyond normal 3D needs but could expand even more for GPGPU functions and many other markets NV is addressing.

ATI seems to also want to address the PDA mobile market (2nd generation) with a USC as just one example.

[CarstenS]

I rather like the idea or the way, 3dlabs have gone: Simple scalar units, then you don't have to worry about anyhting else, than how to improve your dispatcher and how to cram the largest possible amount of them anywhere on the chip.
(Well, not that easy, but a start)

AFAIK the US of the Xbox360-GPU are Vec4+scalar, right? Does anyone know about their "splittability"?

[_xxx_]

There are quite a few tidbits and past experience that has been used since the NV25 in their products that came from the former 3dfx and GP patent portofolio.

Sure, some of it is there in the GFFX but today, I doubt there is anything useful left. But then who knows? ;)

[_xxx_]

Currently, the memory controller on r520 circles the internals of the chip (from what I've read at least - although the die shots suggest otherwise). Now I was thinking, if you add pipelines, that should increase the die size. Therefore the memory bus gets bigger, therefore slower and requring more die... So. Why not actually seperate the memory controller from the shader logic. The memory controller + scheduler + ring bus can sit on it's own (far smaller) chip... Ideally this will speed up the ring.
At each node on the ring, use a similar interconnect found on the r500 (between chips), instead, connect this to a smaller daughter die.. This small chip could, say, contain 8 alus, 8 general purpous memory read/write units, 8 texture lookup units, 8 z compare units, small bit of cache, etc etc etc.

The point of the memory "ring-bus" is to speed up the communication between different parts of the chip. It would rather make it slower to implement that in another external chip, that really makes no sense. No matter how you do it, the internal version will always end up MUCH faster than the same thing implemented on external die.

[Chalnoth]

The benefits IMO stretch way beyond normal 3D needs but could expand even more for GPGPU functions and many other markets NV is addressing.

ATI seems to also want to address the PDA mobile market (2nd generation) with a USC as just one example.
Perhaps. But if the instruction sets are identical, one could just use nothing but the pixel shaders, of which there are many more, and still get most of the performance. Unless you have an algorithm that requires a lot of linear interpolation (so that you could make use of the interpolated registers for efficiency purposes), I don't see why this wouldn't work.

[Ailuros]

Perhaps. But if the instruction sets are identical, one could just use nothing but the pixel shaders, of which there are many more, and still get most of the performance. Unless you have an algorithm that requires a lot of linear interpolation (so that you could make use of the interpolated registers for efficiency purposes), I don't see why this wouldn't work.

My mind was directed more in the power consumption/die size direction and that's why I also mentioned the PDA/mobile market. In that market those two aspects are on the top of the list of any priorities. Having a base platform then that you can scale downwards does make more sense to me.

[Ailuros]

Sure, some of it is there in the GFFX but today, I doubt there is anything useful left. But then who knows? ;)

This is way off topic, but I often have the feeling that NVIDIA wanted 3dfx more to gain a contract with Quantum3D than anything else. Quantum contracted NV immediately after the buyout and they were building multi-GPU configs for their Independence Systems ever since (2001?).

As for today:

Indeed, such a texturing mechanism sounded similar to 3dfx's fabled "Rampage" chip, and it didn't come as much of a surprise to find that Rampage's chief architect was in charge of NV40's texture and shader engine. We asked Emmett if this was the case with NV40 and he replied that "some in the NV4x range would feature this as it has benefits and drawbacks".

http://www.beyond3d.com/previews/nvidia/nv40/index.php?p=9

Kilgariff used to be one of the key engineers of the Rampage project and we're still in the NV4x/G7x era aren't we?

[Dave Baumann]

Currently, the memory controller on r520 circles the internals of the chip (from what I've read at least - although the die shots suggest otherwise). Now I was thinking, if you add pipelines, that should increase the die size. Therefore the memory bus gets bigger, therefore slower and requring more die...
Although it wasn't spoken about at the launch, I'd heard references to the Ring Bus last year and the one of the reasons gave for its implementation then was because of die size issues.

R520's memory controller itself is not a ring, the data return path is a ring round the chip; on other chips the memory controller is the same but both the client and request paths and data paths all point to that controller. If you increase the number of transistors then the conections between the clients and the memory system will increase in both cases - I think the point was that because the data return ring spans all the way raound the trip, the data return paths to the clients are likely to be shorter, so although it uses more space initially, the larger the chip grows the more savings are made in relation to their previous memory architecture.

(The size of the chip was one of the reasons I was given for the Ring Bus not existing on Xenos and for these reasons I was rather surprised to learn that it was included in RV530)

[DegustatoR]

Did NV48 ever show up? I was under the impression it was morphed into G72..
Click (http://forums.guru3d.com//showthread.php?s=&threadid=141559). AFAIK NV48 is the TSMC version of NV40, nothing else.

[Ailuros]

Considering the age of the actual G70 chip i tend to believe that their next flagship product with significant improvments will be the G80. Anything relating to the G7x series should be minorly modified or on a new process as well as mid-low range series. However, combinging my theory with popular theory creates a strange launch schedule unlike what we've seen before if Vista launches on time. 3 high end products between january 06 and january 07.

We "know" that the G70 is what was formerly known as "NV47" and was supposed to be cancelled. Taking under account that logically anything NV4x= DX9.0 and probably anything "NV5x"= DX10, the G80 might as well be what was formerly expected to carry the NV50 codename.

Codenames though are less relevant; what I consider the most certain speculation is another SM3.0 high end part at 90nm. That's two high end parts for 2006 from NVIDIA including a DX10 part.

G7x die shrink with speed increases, which i dont think will ever exist. They'll shrink the die if they feel they have enough time to benefit cost wise from switching production to 90nm. Although i must use harsh fact and point to how fast they shut down NV40 production do to cost and the G70 supersession. So i dont expect to see this core on 90nm unless it launches before christmas time. But architecture differences as well as 30-40% higher clocks to that? I think thats pushing the realistic spectrum. Dont forget both companies are more then willing to delay launching to stock product for hard launching in quantity.

Possibly early 2006 = G7x@90nm and late 2006 DX10 GPU. Where's the third one?

That means both the R580 and the 90nm G7x (assuming its 90nm, had its architecture changed significantly, and has much higher clocks, have to be in full swing now or very soon if they're going to launch as early as people keep saying (early next year?). Something which i just dont see. I think we'll know alot more about plans in Febuary.

Timeframe for either/or depends often on uncontrollable factors, but yes early 2006 sounds reasonable for either/or. I don't expect personally and signficant architectural changes for either/or parts. One sounds like triple the ALU amount, the other with more quads.

G80 with significant improvments to the architecture, 90nm, what everyone expects from the G7x. Still remaining SM3.0. I'd expect this to be the real R580 competitor for spring/early summer. I'd also expect significant advanced shader enhancments from Nvidia here as well although perhaps not quite up to par with the R580.

G80 would build an oxymoron since it would signify a new generation in my mind. Granted "generation" has become a tad fluid these days, but since it will most likely be another SM3.0 part, it would make more sense to call it say G75 or something similar. Both are refreshes in a relative sense to today's GPUs and not a real new generation.

As for being up to par or not in terms of ALU throughput, doing some speculative math I could still figure G-whatever to be slightly ahead.

NV50 speculate what you will, perhaps 80nm, real monster fully DX10 compatable. Winter launch time frame just before holidays. IF MS can stick to a schedule. I'd expect many improvments to come like free AA/AF from both ATI and Nvidia by the time the NV50 and R600 are launched.

We already have in relative terms "free" 2xMSAA for quite some time now. Since I don't expect IHVs to move to single cycle 4xMSAA with the next generation either, I don't expect any significant changes.

As for the supposed "free AF", the only thing that still is for free on today's GPUs is garden variety bilinear; you don't even get yet trilinear for free, let alone anything AF.

We have high chances to see more sophisticated and higher quality algorithms for both AA/AF, but nothing that comes for free.

[Geo]

As for the supposed "free AF", the only thing that still is for free on today's GPUs is garden variety bilinear; you don't even get yet trilinear for free, let alone anything AF.

We have high chances to see more sophisticated and higher quality algorithms for both AA/AF, but nothing that comes for free.

I want an interview with "Tony the texture God". Somebody go make that happen. Thanks ever so much.

:wink:

[nonamer]

*BUMP*

Now let's see our predictions from a year ago.

I rather like the idea or the way, 3dlabs have gone: Simple scalar units, then you don't have to worry about anyhting else, than how to improve your dispatcher and how to cram the largest possible amount of them anywhere on the chip.
(Well, not that easy, but a start)

AFAIK the US of the Xbox360-GPU are Vec4+scalar, right? Does anyone know about their "splittability"?

:yep2:

[Neeyik]

I'd say _xxx_ was pretty much on the money:

I say both will have unified shaders (correct), decoupled tex units (correct), advanced memory controller and scheduler along the lines of R520 (well, not quite but certainly a 'different' controller) and at least 24 "pipes". (at least bit was right) HDR+AA and correct AF will be there as well. (both correct) G80 will of course feature multiple clock domains (yup), R600 maybe as well.

I also think we'll see a replay of this year (business-wise). G80 in late summer, R600 a few months later (later than anticipated but in the right order). I do expect ATI to have parts which are actually available this time around, but nV doing the same trick they did this year.

Tell me something _xxx_ - what numbers do I need to win this week's Euro lottery?

[nAo]

Wow..I like this game! Can I quote myself (http://www.beyond3d.com/forum/showpost.php?p=631920&postcount=36)?
I'm also wondering how many ALUs slots go wasted when you're doing 1,2,3 components ops and at the same time another op can't be co-issued.
Obviously it's not going to happen but it would be nice to push 'diversification' even further and to have only scalar ALUs..:wink: LOL..I'm glad it happened :razz:

[Geo]

Gee, how embarrassing for David Kirk that 3 years into a 4 year project they still hadn't told him it was unified. Doh!




:cool:

[kyetech]

It would be interesting to scrape Titanios one liners he made over the course of the last three months, that he was saying as 'fact' and see just how much FUD or fact he was spreading as gospal?

I gather Uttar was getting pretty pissed off with it at one point !?

[Jawed]

Presumably you meant Trumphsiao.

Jawed

[kyetech]

yeah, I couldnt be bothered to look up his actual name, lazy, but hey you know who I mean..... :-)

[_xxx_]

Who brought this thread alive again?

BTW, nice to see that my predictions (in the first post) from over a year ago were absolutely correct :cool:

[chavvdarrr]

Who brought this thread alive again?

BTW, nice to see that my predictions (in the first post) from over a year ago were absolutely correct :cool:predictions?!
why don't you just admit that your account is managed by both Jen-Hsun Huang and Dave Orton ?
:D

[LunchBox]

Who brought this thread alive again?

BTW, nice to see that my predictions (in the first post) from over a year ago were absolutely correct :cool:

Whoa!
That's way kewl!

[_xxx_]

predictions?!
why don't you just admit that your account is managed by both Jen-Hsun Huang and Dave Orton ?
:D

Hehe! Nope, I manage theirs! :D

[_xxx_]

I'd say _xxx_ was pretty much on the money:

Tell me something _xxx_ - what numbers do I need to win this week's Euro lottery?

Now hey, that was just logic, not magic! I like 7 and 17, if that's of any help ;)

And you forgot the physics reference, that one was also correct :mrgreen: