Next gen consoles: hints?

3 billion polygons ? I'm sorry I can't believe that for a second. Mabye a couple hundred thousand peak and a hundred or two hundred thousand max a second . But to 3 billion polygons is a bit nuts. I highly doubt even a 1tflop cell would put anything close to 1 billion let alone 3 billion


edit * the hundred thousands should be hundred millions sorry
 
well, look at current PC GPUs (R3x0 & NV3x) they do over 300 million peak right now. the R420 and NV40 should bring us to at least 500 million peak. its not hard to imagine next gen consoles being several times more powerful than the upcoming R420 and NV40.

we should be seeing 1-3 billion polygons peak performance for next gen consoles, and with that, several hundred million with all the good stuff applied like pixel shading and lighting, in games, at least by the end of their life cycles. we'll probably start off with games that do tens of millions of polygons instead of hundreds of millions.
 
Super Grafx said:
well, look at current PC GPUs (R3x0 & NV3x) they do over 300 million peak right now. the R420 and NV40 should bring us to at least 500 million peak. its not hard to imagine next gen consoles being several times more powerful than the upcoming R420 and NV40.

we should be seeing 1-3 billion polygons peak performance for next gen consoles, and with that, several hundred million with all the good stuff applied like pixel shading and lighting, in games, at least by the end of their life cycles. we'll probably start off with games that do tens of millions of polygons instead of hundreds of millions.

The only question is exactly what is the max they put out. If the r3x0 and nv3x do 300 million peak what does it actually do. If it only does 50 million a second thats not all that much is it. So even in 2 years if we do hit a billion peak what if it only can sustain 200 million ? I don't see them hitting much higher. And most def not with all the effects turned on
 
archie4oz said:
Shaping controller as a number 3 would be cool too indeed.

ARGH! You thief!

Too bad this isn't Xbox3 or something, then the 3 idea would REALLY go over well... Hehe, they could emblazon xXx all over it and get sued by Sony... :p

Anyways in keeping with the theme of 3's how about instead of 256 shades of green, how about dark grey (or for the geek folk, #333333)?

The Ramans do everything in threes. At least according to Arthur C. Clarke.
 
Well X-box had some success, I expect the sequel XX-box to bring in more of
the college and high school crowd, and finally the main stream will be achieved with the XXX-box... ;) ( This may also be called the R-box )
 
The only question is exactly what is the max they put out. If the r3x0 and nv3x do 300 million peak what does it actually do. If it only does 50 million a second thats not all that much is it. So even in 2 years if we do hit a billion peak what if it only can sustain 200 million ? I don't see them hitting much higher. And most def not with all the effects turned on

For example in 3dmark2001 i get ~85millions/poly/sec with one light and ~27million with eight lights in the FF test, thats on a NV35@400Mhz. Depending on how the shader units look like in R5xx(probably unified) i would think that a "peak" vertex number dont mean anything but the shaderops/ipc is more important to know.
 
overclocked said:
The only question is exactly what is the max they put out. If the r3x0 and nv3x do 300 million peak what does it actually do. If it only does 50 million a second thats not all that much is it. So even in 2 years if we do hit a billion peak what if it only can sustain 200 million ? I don't see them hitting much higher. And most def not with all the effects turned on

For example in 3dmark2001 i get ~85millions/poly/sec with one light and ~27million with eight lights in the FF test, thats on a NV35@400Mhz. Depending on how the shader units look like in R5xx(probably unified) i would think that a "peak" vertex number dont mean anything but the shaderops/ipc is more important to know.

Well see thats just the thing. I don't see 1 light as awe inspiring. 8 lights are much better but look at that drop off less than half of what 1 light gives. So what the r500 can mabye do a 100 million polygons / sec with 8 lights . Which would be great and will be a huge leap foward but it wont be enough to do what the ps3 would render like pics. Not even the ps3 will be able to do those. The rendering power isn't there yet. A big leap from last gen console hardware but not from this years pc hardware .
 
jvd said:
A big leap from last gen console hardware but not from this years pc hardware .

I am not correcting you, merely asking why you believe this to be the case. Why can't, say PS3, be a large leap up from the R350 and NV35? If any of the consoles will be a large leap forward, it'll be the PS3. The other architectures have too much legacy, too much linear thinking/progression from what we've heard (eg. ATI/DirectX next)
 
Well see thats just the thing. I don't see 1 light as awe inspiring. 8 lights are much better but look at that drop off less than half of what 1 light gives. So what the r500 can mabye do a 100 million polygons / sec with 8 lights . Which would be great and will be a huge leap foward but it wont be enough to do what the ps3 would render like pics. Not even the ps3 will be able to do those. The rendering power isn't there yet. A big leap from last gen console hardware but not from this years pc hardware .

Agreed, as you said it will naturally be a "huge" step from current consoles but not as big from current cards from ATi and nVidia that are real fast ones.
Something i would like to see is more detail in nextgen games, something like a new AceCombat with citys full of cars and real windows on the houses with real geometry and not just a texture. As a big Starwars fan i would like too se a full simulation of Coruscant and Naboo.
 
Vince said:
If any of the consoles will be a large leap forward, it'll be the PS3. The other architectures have too much legacy, too much linear thinking/progression from what we've heard (eg. ATI/DirectX next)

Tell us, in your informed opinion - where is this linear thinking/progression actually lacking? How is progression from whats known and accepted a bad thing? Surely the unknown also has the largest risks?
 
Biggest risks, biggest rewards... <shrugs> That's why it can be such a wild card. I figure next gen will be a big hop from this year's hardware simply because we have two years+ yet for next generation to appear, while PC's themselves tend to never look as good as they could anyway because of the broad multi-purposing of the PC itself, along with developers always having to support aging tech.
 
Tell us, in your informed opinion - where is this linear thinking/progression actually lacking? How is progression from whats known and accepted a bad thing? Surely the unknown also has the largest risks?

There is nothing wrong with it, it's the safest and most surefire way, but it also yields the most predictable results.

Your right, the unknown has the largest risks, but the unknown also has the largest returns if everything turns out right.
 
There is nothing wrong with it, it's the safest and most surefire way, but it also yields the most predictable results.

And when new technology is given to developers this is something thats quite desireable - the least amount of effort for them to learn something new can often pay great dividends, and then they can explore the limits from there.

However, I've not heard any any actual, real deficiencies yet.

but the unknown also has the largest returns if everything turns out right.

Ummmm, well if its unknown, then surely thats not known.
 
Vince said:
I am not correcting you, merely asking why you believe this to be the case. Why can't, say PS3, be a large leap up from the R350 and NV35? If any of the consoles will be a large leap forward, it'll be the PS3. The other architectures have too much legacy, (snip)

These are the significant legacy bottlenecks on a PC, and how the Xbox1 overcame them.
1) Poor memory bus - 256 bit DDR bus @ 100 Mhz, something nonallowable on PC do to the wider tolerances needed in an open platform but quite possible in a closed box.
2) Lack of CPU access to video memory - Unified Memory Architecture
3) Only high level access to coprocessors - Low level push buffers
Thats about it. What's a PC holdover that holds Xbox back? I can't think of any.

(unsnip) too much linear thinking/progression from what we've heard (eg. ATI/DirectX next)

And yet the PS2 suffered more from this philosophy than anything else I can think of. It's one big extrapolation from the MMX/Voodoo 2 era, just turbo-charged. Sure, there's many improvements in a macro sense, but Xbox has all of those as well.

Bottom line - the Xbox chipset has an extremely similiar silicon cost to PS2's chipset and yet which chipset is vastly superior? It's really obvious on anything better than a NTSC/PAL/SECAM set. I really hope PS3 is a "paradigm shift" from PS2, last thing we need is another MS monopoly or a return to US$89.99 games from "consumer-friendly" former monopolist Nintendo.
 
And when new technology is given to developers this is something thats quite desireable - the least amount of effort for them to learn something new can often pay great dividends, and then they can explore the limits from there.

However, I've not heard any any actual, real deficiencies yet.

Ok.

It's better for developers I admit that, you also get more out of it in a shorter ammount of time.

There are no real deficiencies though, this I never stated.

Ummmm, well if its unknown, then surely thats not known.

Your right. It's the wildcard, given nothing, you cannot predict it. It could suck, or it could turn heads. From what we see Sony wanting to do with it, it looks like if everything goes 100% their way and if the patent BE is what the final PS3 BE turns out to be, it will turn heads.

Bottom line - the Xbox chipset has an extremely similiar silicon cost to PS2's chipset and yet which chipset is vastly superior?

Can't be compared.

Xbox is based off of better lithography and came out 18 months after PS2. If both came out at the same time your argument would be great, but I feel that it just isn't fair.

Obviously if Sony were targetting Fall 2001 as a release date the PS2 spec would be different would it not? The lithography better, thus the spec higher.
 
Your right. It's the wildcard, given nothing, you cannot predict it. It could suck, or it could turn heads. From what we see Sony wanting to do with it, it looks like if everything goes 100% their way and if the patent BE is what the final PS3 BE turns out to be, it will turn heads.

Much like most of the other new generation hardware will turn heads I'll guess.
 
As far as what you see on screen? Yes for all, concidering the fact what we're seeing a Xbox do with a Geforce 4 and 64mb of main ram.

Games like Ninja Gaiden are better than what I'm seeing done on the PC, well excluding around 5 games. The joy of dedicated coding and hardware.

Can't wait until what developers can do with 512mb of faster DRAM and a R500 derivative.
 
Games like Ninja Gaiden are better than what I'm seeing done on the PC, well excluding around 5 games. The joy of dedicated coding and hardware.

OT from the point here, but....

Dedicated hardware is but one issue, and one that increasing less important with each revision of DX. One of the primary reasons PC games look as "clanky" as they do is merely due to the fact they have to code for a wider range of older hardware. DX8 and 9, with VS/PS and vertex buffers can already largely bypass much of the PC deficiencies - look at 3DMark03, by and large much of this is quite invariant to the host - and this is only increased with DX10. I'm playing NFS: Underground at the moment and I'm fairly positive the 9800 PRO I'm using would happily chuck a bunch more polies around - the consoles are holding me back here.
 
Back
Top