Slashdot Mirror
nVidia's GeForce 256 Breaks Out; changes 3D world
Hai Nguyen writes " nVidia officially unveiled the GeForce 256 (the chip formerly known as NV10).
Its architecture emphasizes both triangle rate and fill rate, so the chip can render 3D landscape with highly detailed 3D environments and models, and smooth framerates. Go get the full info." Holy moses. I want one. Now.
Actually this is pretty big news. Up till this point fill rate has been all that mattered for 3D chip makers, providing the capability of allowing higher resolutions, more rendering passes (for different visual effects) and higher frame rates. This is the first (consumer level?) chip to add transformation and lightning to the 3D chip, thus offloading these duties from the CPU. This effectively means more polygons can be used and this will have a truely remarkable effect on the realism.
Aren't you tired of watching perfectly flat walls with big posters stuck on them?
Damn. I was hoping this would get linked up on the front page. :) Oh well, I took a trip to NVIDIA last week, and I'd love it if you guys checked my article out: riva extreme - geforce 256 coverage
- 15M triangles/sec - sustained DMA, transform/clip/light, setup, rasterize and render rate.
- 4 Pixels per clock (4 pixel pipelines).
- 480M pixels/sec fill rate - 32 texture samples per clock, full speed 8-tap anisotropic filtering.
- 8 hardware lights.
- 350 MHz RAMDAC.
- Most feature complete for DX7 and OGL - Tranform & Lighting, Cube environment mapping, projective textures, and texture compression.
- Will utilize 4x AGP performance with Fast Writes , which enables the CPU to send data directly to the GPU (1 GB/sec transfer rate), increasing overall performance and freeing the system memory bus for other functions.
- 256 bit rendering engine.
- Highest quality HDTV (High Definition Television) video playback.
- High Precision HDTV video overlay.
- 5 horizontal, 3 vertical taps.
- 8:1 up/down scaling.
- Independent hue, saturation and brightness controls in hardware.
- High bandwidth HDTV class video I/O.
- 16 bit video port.
- Full host port.
- Dedicated DMA video.
- Powerful HDTV motion compensation.
- Full frame rate DVD to 1080i resolution.
- Full precision subpixel accuracy to 1/16 pixel.
Snipped from www.bluesnews.comI just went and looked at the tweak3d guide ( http://www.tweak3d.net/reviews/nvidia/geforce256/1 .shtml ) and good god this card kicks ass.
The addition of Transform and Lighting really _is_ revolutionary. Once you've used one of these babies, you won't want to go back.
There's a list of useful links at Blues News (www.bluesnews.com)
My Journal
"hands-on tests" means prototype hardware already available, so this should be out fairly soon. None too soon, as S3 pulled ATI's trick and is coming out early with a chip at 0.18.
The Playstation II has a modified R10000 processor with very hefty floating point extensions - it won't have much of a problem doing geometry transformations. IMO, it will probably be about on par with the graphics cards floating around at the time of its release. It won't leave them in the dust, but neither will it be left in the dust.
OTOH, a friend in the gaming industry says that the Playstation II has architectural problems that might degrade performance (low system bus bandwidth, among other things). We'll see what happens when it ships.
So basically nVidia chose to make a high fill-rate card with hardware lighting and transforms (geometry acceleration). These aren't innovative directions -- they were the obvious ones. None the less, the other major player, 3dfx, has pulled back from these choices. I'll explain why:
... everything is (we once demonstrated that it's physically impossible under DX6 to be faster than a Voodoo3 under Glide). There are some downsides: if you want to do crazy weird stuff with your lighting (eg. wrong faster stuff, funky effects) you may not be able to get it to work. Similarly with geometry -- special fast cases will become normal cases. So there may be a 50%-100% gain in triangle rate, but it's unlikely geometry acceleration will ever be able to provide much more than that.
... 3dfx talked about putting a geometry accelerator on V4 but I believe they backed off from it. Voodoo4 is however still an SST and therefore still a true descendent of the original Voodoo chipset conceived as a flexible, long-term solution for both PCs and arcade games.
... it may be finally time to kill some very old paradigms in 3d hardware...
nVidia has a card which can do supported operations fast. It obviously has a lot of fill. It'll be a good board. Of course it'll still be slow in D3D
nVidia seems to have chosen not to support the hardware bump mapping of the Matrox G400, an extremely high fill (runs beautifully bump mapped in a window in 1600x1200x32bpp) card without geom accel. 3DLabs' long awaited Permidia3 will also have some kind of hardware bump. IMHO this is a relatively flexible feature -- you could do a lot with it. It remains to be seen how flexible nVidia's lighting and geom turn out to be.
I'll be impressed if D3D ever delivers real hardware geometry benafits. We have yet to see a single benefit of DX6 over DX5 (not screwing with the fp control word especially) actually work. I'm highly suspect of anything MS sez.
So what about the remaining behemoth, 3dfx? Their Voodoo4 is supposed to be an extremely high fill card (fill has always been their hallmark). It may not support any more hardware features (eg. bump, lighting, geom accel), but it will fill like crazy. It's supposed to do full screen anti-aliasing
I'm eagerly awaiting the new generation. But I expect the real crazy stuff to start happening in the following generation
The Voodoo 4 will be coming out around Christmas, and it will have hardware geometry as well. Rumour has it that it too will be at 0.22 micron instead of 0.18. I don't remember the name of the chipset off-hand.
S3 has also rolled out a new chip, with four pipelines and hardware geometry, at 0.18 micron. Check Sharkey Extreme for details.
Also, I've heard some reports that the PlayStation II will beat the living daylights out of PIII's loaded with then recent and most modern 3d accels. Even with this kind of chip, and most likely other chips to follow from nVidia's competitors, does this still hold true? Will the PlayStation II live up to the hype?
No, but it won't sink either. See my previous response on this subject (check my user info to find the post).
Get a load of the supplied pictures. Gee, low poly models sure don't look that impressive when you DON'T TEXTURE THEM ;P or do you really think the treads on the second tire are, or ought to be, geometry?
So it pushes 15 million triangles a second and a PIII only does 3.5 million. Well, where do they come from? Exactly what is used to store these geometries? I'd say that if they went with a rather Voodoo Glide-esque approach of putting all the geometries on the card and then giving minimal commands to position, scale and rotate them, then it could be significant. This, however, would be pathetically incompatible with all existing games- and frankly the bus is the bottleneck, that PIII is probably pretty comparable for doing transforms, it just cannot get them across the _bus_ as fast as a cached copy of the geometry on the card.
I saw what appeared to be a statistic that implied that games might see a 10% improvement in framerate. That, I think, is closer to the truth.
Sorry guys- you've been Hyped.
Um.
You do know what the word "hyperbole" actually means, don't you?
--
Don't imitate. Enervate.
3dfx is rolling out another chip, as people have been talking about for a while. It is rumoured to be at 0.22 micron too, and will have hardware geometry processing.
S3 already rolled out a new chip - at 0.18 micron. It too has four texel engines and hardware geometry processing.
IMO, the S3 chip is actually the one to worry about. Architecture may or may not be great, but at 0.18 micron it may outperform nVidia and 3dfx's offerings just on linewidth. ATI did something similar when it rolled out the Rage 128, if you recall.
What I'm waiting for is the release of the GeForce or (insert name of 3dfx's offering here) at 0.18 micron. However, I'll probably be waiting a while.
"Consumer level" is correct. High-end graphics workstations have been doing this for several years; in fact, the entire OpenGL pipeline has been in hardware for quite a while. Check out 3dlab's high-end boards for examples, or take a look at their competitors. These tend to be 64-bit PCI boards in the $2,000-$3,000 range.
The consumer graphics manufacturers have been making noise about using geometry processing for a while now, but have only recently gotten around to it. In that market, yes it could be called revolutionary (in that it substantially changes game design).
I read another article (the one on rivaextreme ... pretty good article), I can add a few more comments.
... imho this is kind of limiting ... we'll see.
... I again expect GeForce to better that ...).
GeForce has environment mapping (iirc so does Permidia3) but not bump mapping.
It can do 8 free hardware accel'd lights
A Voodoo3 on a fast machine under Glide can handle about a 10-12kpoly scene lighted textured w/effects and phsyics running about 20-25 fps on a 450a. I'll be very impressed if GeForce can do twice that -- 25kpolys at 25fps, or about 500k real polys/sec (BTW a Voodoo3 under ideal conditions w/out features can do 500k "fake" polys/sec
But 15million polys/sec is the kind of bloated number that usually comes out of graphics shops. Don't believe it for a second.
As for 100kpoly models lighted w/fx running smoothly, i'll believe it when I see it.
if DaveS or DaveR wants to correct me on any of this stuff, go right ahead guys...
Not if it can only transform 36 million polys per second, it can't (sustained transformation figure from an older slashdot article).
Based on all of the other numbers in that article, I suspect that they dropped a decimal point in the "75 million polys rendered" figure. That, or they're talking about flat-shaded untextured untransformed polygons.
Yes, but how well can the PCG systems do at movie speeds: on the fly rendering at 24 frames/second? nVidia is saying (and it remains to be seen how accurate their marketting info is) that this is the image quality quality that can move, not just a pretty static image.
----
----
Open mind, insert foot.
Microsoft writes each version of D3D by asking the manufacturers "What features do you guys need?" and then writing them in.
:)
...
There were no 3D games bfore D3D. No one had cards.
Just 'cos a card supports D3D doesn't mean you can assume your program will work right. You still have to test and debug each individual card. This is the voice of experience
Matrox's bump will be in D3D i'm pretty sure...
People use Glide rather than D3D 'cos it's way faster. Speed really is all that matters
Do you see GeForce boards on the shelves?
Do you see Playstation 2s on the shelves?
They've been conversation topics for months, but all either has now is alpha test hardware. It becomes difficult to see what point you are trying to make, given that.
That's easy- I presume you're rotating the tree realtime? _All_ that requires is that the tree can be cached on the card, which is then issued commands. ;) ) we have a comparable problem- there are 4X the voodoo cards as anything else, because of availability, and we're getting 'em off you PCers who are buying nVidias ( ;) dirt cheap, too! ), but Apple only supports ATI- so many important development tools are _not_ supporting 3dfx or Glide, and we are once again suffering the recurrent apple disease of Thou Shalt Use Only One Solution- in this case, ATI 3d acceleration. And I personally like 3dfx rendering better than even _TNT_, but this helps me not. (reading User Friendly I have been!). ;P
Why only one tree? What program, exactly, did this? There are some very serious questions to ask about demos like this. I, too, write software and try to come up with impressive claims. I can legitimately say that I'm writing a game with a ten million star universe with approximately sixteen million planets, of which the terrestrial ones (hundreds of actually landable-on planets) have terrains the size of the earth at 3 dots per inch for height information.
This is misleading as I'm doing it _all_ algorithmically- it's fair to ask 'well, what does it work like?' but nonsensical to imagine that somehow I'm messing with kajilliobytes of data. It's faked. (I have stellar distribution whipped, am working currently on deriving star types, slightly modified according to actual galaxy distributions- main current task is to come up with RGB values for the actual colors of star types, as this is more like white point color temperatures than anything else- very close to updating my reference pictures.
At any rate, will you believe me when I say that this reeks of demo? It wouldn't be that surprising if they used _all_ the capacity of the card to do that one tree. _I_ would. Might that be why there is only one tree and _no_ other detail at all (one ground poly, one horizon)?
More relevantly, what was used in doing that? If it was vanilla OpenGL, then okay, I concede this is very big. If they had to write their own software to do that, then you have a problem. Here in Mac land (also LinuxPPC land
You guys are looking at exactly the same situation here. Be damned careful. If you go with a proprietary technology you will fragment, and your developers will be faced with tough choices and could end up writing nVidia-only much as some developers in Mac land are writing ATI-only. This is bad. Do I have to explain why this is bad?
Let's get some more information about exactly how you operate this geometry stuff before getting all giddy and flushed about it, shall we? I don't see how software will use it without rewriting the software. And when you do that- it's an open invitation for nVidia to make the thing completely proprietary and lock out other vendors.
Or maybe they'd give the information out to people at no cost and not enforce their (presumed) patents for a while, only to turn around a year from now when they've locked in the market, and start bleeding people with basically total freedom to manipulate things any way they choose? But of course nobody (GIF) would think (GIF) of ever doing (GIF!) a thing like (GIFFF!) _that_...
Whoa! you have a 3D card on your sub-$400. WHY?
What? has S3 lost their minds? don't they know they can't make good video chips. I mean sure S3 makes a decent 2D video card, but common why are they trying to throw themselves into the market instead of just packaging someone elses 3d chip on their card like every other company does.
It's perfectly possible to do 3D in software. It's even somewhat more interesting, since you're not bound to and 3D card's paradigm. Software is where you get cool stuff like true voxels etc...
Glide probably wasn't out before DX3, but DX3 was pretty much useless (only a very minimum 3D API) so Glide may have beaten anything useful, tho not by much...
------------------------
Replying to another comment:
No, an API cannot be faster. But an implementation can. And an API's design can affect an implementation. In any case, Glide (the implementation) is a hell of a lot faster than D3D (the implementation), as per my original comment.
First off, I don't take this it as a given that just because you can't figure out a way to represent the meshes with variable levels of detail, that no one can. In fact, its my understanding that Quake 3 implements curves in a way that allows them to be retesselated to higher polygon counts depending on the graphics card and speed of the system. Second, even if a company didn't want to implement something like that in their engine, its not inconcievable that multiple environment resolutions could be placed on the game media. Many games already come with low and high quality sound samples to account for the wildly varying quality of sound cards out there.
Jherico
What can the average user can do to ensure his security? "Nothing, you're screwed"