SCEI & Toshiba unveil 65nm process with eDRAM

[Fafalada]

Apparently SH3 uses soft edged stencil volumes (I still would like to know how that is done without using several jittered volumes) so aliasing never becomes a problem.

Well it wouldn't matter, even hard stencil shadows have far less aliasing then shadow bufers.
But, considering what method SH3 appears to use, I wouldn't say it can "never" become a problem.

Chap, sure.
The guy programming the engine (me) has been having health problems lately. How's that for gossip

 

[Guest]

Fafracer2k3 it iS!
Hope you get well soon.

 

[archie4oz]

Throw a little .hack into the equation and you could possibly see release dates being pushed back...

 

[BenSkywalker]

Cybermac-

Classical meaning of shader. Vertex Shader is merely a buzz word invented by Nvidia PR.

Microsoft is now nVidia's PR firm? Didn't know that one.

When ppl talk about shaders I think of something that describes the properties of an object... not geometry transformation and whatnot.

Deforming an object doesn't relate to its properties?

I haven't played it but there are plenty of screens on the net proving that is not the case.

You haven't played it?

Well... you seem to like DOA3's lighting and that is done per-vertex save for a very few effects.

Vertex will be fine for a great deal of things, using a suitable substitute for those areas is a mish mash.

Like RL you won't notice it most of the time but there are times when the maps are stretched across the screen that it becomes very apparent.

After you play it, tell me where. In RL is screamed out in certain areas(the underside of Corvettes, AT-ATs etc).

Faf-

Eh... would those people be referring to PC users - same PC users perhaps of which 90% thinks Doom3 is high poly?

I'm not saying that most people get it right. Doom3 is obviously very low poly(hard to believe some people can't see that).

Anyway, from what I've seen of people's estimates on console games, I don't recall anyone coming within 50% of actual numbers(except maybe by pure luck), though granted usually they were mislead by developer PR releases.

Sega's 63Million polys per second for VF4 springs to mind

The guy programming the engine (me) has been having health problems lately.

I hope it isn't anything serious. Get well soon.

 

[zidane1strife]

Why would I argue vertex performance? I have never argued against the PS2 in that aspect, certainly no reason to start now. Even if the XBox was capable of tripple the geometric complexity of the PS2 at the level we are already at it isn't going to make a huge difference. We are at the point now where hard edges are increasingly rare. Telling the difference between a 100poly sphere and a 500poly is easier then comparing a 500 with a 1,000,000. There is certainly still room for improvement, but it will take a generational gap to make it much of a factor.

The argument here is Geforce 1 v.s. Ps2 given the gf1 has no pixel shaders, we should discuss the other non pixel shader areas... I brought the xbox because it seems that in many of those other areas the ps2 is not far behind what is basically a Gf4... thus if it's not too far behind a Gf4 in non-feature-pixel shader related areas... and thus compared to gf1... u get the idea.

Anyways the cpu side of this argument is kind of funny so since the gf1 will always be cpu limited we can use cpus from yrs, decades, or centuries later.... we can take benchmarks out of games that do part of their T&L on the cpu and will massively benefit from such upgrades, and attribute the perf. to the gf1....

PS: hope u get well soon faf.

 

[BenSkywalker]

Zidane-

The argument here is Geforce 1 v.s. Ps2 given the gf1 has no pixel shaders, we should discuss the other non pixel shader areas... I brought the xbox because it seems that in many of those other areas the ps2 is not far behind what is basically a Gf4... thus if it's not too far behind a Gf4 in non-feature-pixel shader related areas... and thus compared to gf1... u get the idea.[/qoute]

What are the major differences between the GF4 and the GF3? Another vertex shader and LMAII. What are the major differences between a GF2 and a GF1? Build process and another TMU per pixel pipe. What is the fill difference between the NV2A and the NV15(GF2)? Comparing to the particular NV15 that I run, 1.86GTexels v 1.76GTexels. You take away vertex shaders and pixel shaders and the GF3/GF4 aren't all that different from the GF1/GF2(there are some differences particularly pertaining to efficiency of the architecture, and supported filtering modes, but overall....). If you say outside the pixel shaders the PS2 isn't that far behind the XBox, you are saying that it is comparable to a GF2 with a strong vertex unit which in itself is marginal over the GF1.

Anyways the cpu side of this argument is kind of funny so since the gf1 will always be cpu limited we can use cpus from yrs, decades, or centuries later....

No, it won't. Once the framerate ceases to rise running 640x480 with faster CPUs then the limit is the GeForce chip. Load it up with enough T&L work, or stressful enough conditions in terms of fillrate and it will happen even with current CPUs. I haven't seen it yet though.

we can take benchmarks out of games that do part of their T&L on the cpu and will massively benefit from such upgrades, and attribute the perf. to the gf1....

I haven't seen a game that has an edge running soft T&L over hard T&L @640x480 yet, no matter which CPU it is.

 

[cybamerc]

BenSkywalker:

> Microsoft is now nVidia's PR firm? Didn't know that one.

Didn't Nvidia develop that part of DX8?

> Deforming an object doesn't relate to its properties?

Let me be more specific then. When I think of shader I think of materials. Displacement mapping certainly relates to that but physics and animation don't.

> You haven't played it?

That would be what I meant when I said "I haven't played it..."

> Vertex will be fine for a great deal of things

Just not when it's on PS2?

> After you play it, tell me where.

I don't need to.

> In RL is screamed out in certain areas(the underside of Corvettes, AT-ATs etc).

Yeah, that's one of the problem's you face when using shadow buffers and not confining dynamic shadows only to small rooms

 

[marconelly!]

Ben, Splinter Cell doesn't have full scene dynamic shadowing. It's rather obvious. Many of the shadows are static and pre-calculated.

I have numerous duplicate screenshots one with extreme aliasing and one with almost none for each. All you have to do is adjust the contrast on your capture device.

The shot I posted is a framebuffer grab, most likely provided by the developers. It's obviously not a video grab made by capturing device.

 

[zidane1strife]

with a strong vertex unit which in itself is marginal over the GF1.

Dude if u'r talking about ps2... it's two vertex units running at more than 2X the speed of the gf1... and yes they're more flexible too.

I haven't seen a game that has an edge running soft T&L over hard T&L @640x480 yet, no matter which CPU it is.

Well according to my quotes(dunno if they're right) quake III does part of it's T&L on cpu, it's not h/w versus s/w T&L, it's part software part h/w. Thus it truly does benefit from 3Ghz PIV cpus and the like.... clearly the s/w T&L perf. of what's being done on such a cpu is FAR FAR beyond what a measly gf1 could ever do.

EDIT: Not to mention that if u mean that at 640x480 the cpu is not limiting the gf1... it's perf. is pathetic.... and with it being easy to dev. for and well suported, i doubt optimization of the kind that takes place in fixed platforms will get u more than 30-40% improvement...