7 Years of 3D Graphics

xtra writes "At Accelenation they are running a nice timeline about 7 years of pc 3d graphics contains much info and even talks about some of the not so well known players anyone still remember rendition? or BitBoys?" How many cards on their timeline chart have you used?

Comments

Infinity and beyond...

[ackthpt]

At the going rate, the board with CPU and chipset will be a daughterboard of the graphics motherboard. :]

Re:Infinity and beyond...

[Boone^]

the GF3 uses vectors for calculations. An Athlon using the x86 CISC requires more overhead per instruction.

Athlon:
add ax,bx; % where ax and bx are 32-bit scalars

GF3:
add reg0, reg1; % where reg0 and reg1 are vectors of scalars

If you wanted to do the same work of the GF3 on an Athlon, you'd need 32 (or however deep the vector registers are) successive instructions.

Once you ditch the instruction overhead for doing an operation on X number of successive scalars, the processor spends more time doing the math and the FLOPS goes up. Take a look at the AltiVec unit in the G4, or CRAY vector supercomputers.

Why 7 years?

[Jacek Poplawski]

Why only 7 years of 3D graphics on PC?
What about Stunts, Elite, and other 3D games?

Re:Why 7 years?

[Rogerborg]

• What about Stunts, Elite, and other 3D games?

Given that ZX81 (aka TS1000) emulators for the PC were around from day dot (or a ZX81 emulator running on an Atari ST emulator running on a PC...) how about 3D Monster Maze? It's not that much more primitive than Doom. Run! Run from the scary Tyrannosaur!

(Yes, I know, the article is probably about hardware that draws triangles real fast, but it's Slashdotted hard, so we may as well have some fun reminiscing. If nothing else, it'll confuse the young 'uns ;-) )

How fast do we really need to go?

[TrollMan 5000]

Can graphics technology possibly get any faster? Well the GeForce2 GTS chip ran Quake3 at 80fps in May of 2000. Just twenty-two months later a GeForce4 Ti4600 can run Quake3 over three times faster. On that reckoning the GeForce6 in two years time should be running Quake3 at over 700fps. Is that fast enough for you!

Is there really much visual difference between 700 fps and 135 fps? I'm not really sure if the human eye can make the distinction. They're sure pretty-looking numbers, but do the results show for it?

And how long before video cards can render essentially photo-realistic graphics? Soon games will be more like interactive movies.

Re:How fast do we really need to go?

[Hieronymus Howard]

Ray tracing isn't the only way to achieve high-quality graphics. Renderman (used for Toy Story, Monsters Inc. and plenty others) isn't a ray tracer, but a sub-pixel renderer (if I remember my computer graphics M.Sc. correctly).

We've still got a very long way to go until we get Monsters Inc. quality real-time games. As you say, current cards render triangles. Curved surface rendering (e.g. NURBS) may come next. Anti-aliasing takes a lot of power. I think that current cards are still using Gourard shading, which is the most primitive shading model there is (correct me if I'm wrong here). The next step is Phong shading for highlight effects (there are hardware-optimised Phong shading algorithms, but they're still slower than Gourard). Then there's deformation mapping (Renderman again), etc. etc.

I believe that Quake 3 etc. does use radiosity algorithms, but that doesn't need to be done in real time, just when the level is compiled.

HH

Re:How fast do we really need to go?

[bribecka]

Renderman (used for Toy Story, Monsters Inc. and plenty others) isn't a ray tracer, but a sub-pixel renderer (if I remember my computer graphics M.Sc. correctly).

You're referring to PhotoRealistic Renderman (PRMan), the actual product developed by Pixar. It uses the REYES algorithm.

RenderMan is a specification for defining 3D scenes, much as PostScript is a specification for defining 2D documents/images. There are many renderers that are RenderMan compatible, including raytracers such as BMRT.

Only 7 Years?

[green pizza]

Heck, I still remember the "which is better, Silicon Graphics Reality Engine or Ferrari Testarossa?" threads in the USENET from the summer of 1992. Even the dual pipe / dual head SGI VGXT "Skywriter" from 1989 was pretty damned impressive. Even many, many years later.

Re:Number Nine

[Jeffrey Baker]

Uh? I don't remember any #9 cards with 128MB of memory. In fact, #9 was out of business two years ago. You may remember the #9 i128 series of cards, but those are very old and do not have 128MB memory.

You only need 16MB to handle th highest resolution computer graphics displays ever made.

Re:Demo scene.

[2Flower]

Future Crew reassembled, broke up, faded out, tried to get into the games biz, etc. Their site has been offline for awhile.

Fortunately the demo scene lives on; pouet hosts links to nearly every demo in existence across multiple platforms. And to keep us on topic, most demos nowadays are 3-D accelerated. It's become less a game of "What techie tricks can you do?" and more a game of "How artistically can you use the technology?". There's some visually striking demos being made nowadays, and not just because they have shadebobs or glenz cubes.

Two words :

[Viking Coder]

"Motion blur".

Or how about "rendering passes"?

Or how about "anti-aliasing"? (Kind of cheating on that one.)

Or how about "soft shadows"?

In short, more is better. If you give me higher framerate, I'll figure out what to do with those extra cycles.

Re:Two words :

[Viking Coder]

I forgot to mention "reduced latency."

If you can render the entire image in the 1/700th of a second before the screen refresh, you can much more accurately track events than if it took you 1/60th of a second to render the image. USB mice allow a much higher sampling rate that you could take advantage of, for instance. When you incorporate things like motion-tracking polhemus devices, and a head-mounted display, things get really interesting.

Re:Just incremental improvements

[green pizza]

My voodoo2 SLI still kicks all kinds of ass.

Amen, brotha... my only real PC at home is a PII/400 with a pair of Creative Labs Voodoo2 cards running SLI. Win98SE is stable (enough) for the few games and utilities I run on it. And 56 FPS in Quake2 and 41 FPS in Unreal is good enough for me.

calling Dr. Jim Clark...

[green pizza]

... it's time for the next wave of 3D.

I love playing with the SGIs at work and I enjoy playing with the wizbang PCs that my roommates and I have, but to be honest, I'm really not that impressed with modern gfx accelerators. The original geForce was pretty neat, and SGI's last big leap (InfiniteReality in '95) was cool... but golly, things really haven't changed much since Clark and his gang from Stanford opened our eyes to 3D in '82.

We've gone from cabinets to cards to chips to a single chip. We've added some gfx extensions and now do multiple rendering passes to make things look prettier... but really, nothing has changed much in the recent years. It's smaller, faster, cheaper. Steady evolution... but so is the scum growing in my bathroom sink.
Please excuse me while I yawn.

Re:Number Nine

[Milalwi]

You only need 16MB to handle the highest resolution computer graphics displays ever made.

This is true for 2D displays, but when you start having double and triple buffering plus z-buffers it starts to add up. Then add the texture requirement and you can see why most newer cards have 64-128MB of memory on the cards.

Milalwi

Rendition? Did you say Rendition?

[Rogerborg]

I still have nightmares about developing for the Rendition Verite 1000, which was a lovely graphics decelerator on anything faster than a P100. When we got our first batch of Voodoo 1's delivered, there was a brief but very ugly struggle to get our clammy hands on them. You ain't seen pathetic until you've seen geeks wrestling and squealing like stuck pigs over 4Mb graphics cards, let me tell you.

Question to anyone else who has developed 3D graphics: who did you find driving the demand? In our games house, there was a running battle between the programmers and the artists. Us code monkeys were forever on at the artists to cut down the polygon counts, but they kept trying to slip in models that were barely stripped down from the FMV sequences. In the end, we came to an equitable solution: they won, the game ran at 10fps, and all the programmers left.

I wonder how many other games were ahead of their time in that regard, and how many of them would be rescuable given cards that scoff at polygons and eat dozens of 256x256 textures before breakfast?

Re:Number Nine

[ram.loss]

I think you mean the first consumer level graphics card to use a 128 *bit* data path. I remember seeing them bundled with Dells in the back cover of Byte a few years ago.

More memory

[Traa]

> You only need 16MB to handle the highest
> resolution computer graphics displays ever made

you will allways need more memory (in 3D graphics accelerators), even if the display resolutions don't increase. Lets say we settle for a nice 2000x2000 ish display. Thats 4M pixels, at 32-bit is 16MB for the display.
At least double (32MB) but preferably triple (48MB) buffer this so you can create a new frame while the old one is being displayed. Then we need a Z-buffer (or W-buffer) to hold the depth values (24bit values) for each pixel, so we know what is in front of what, typically you might want to do some stencil effects to (8-bits, can be packed with the Z-buffer) that would be another 16MB. Now we have the basics for a 3D graphics display and are at 48-64MB.

But we are not done yet, now for some more interesting effects:
- Texture memory. Typically use the leftover graphics memory and swap the rest from host memory (but we don't like swapping, so preferably all textures should be in onboard mem) 2-64MB
- 2x Antialiasing (1 Backbuffer + 1 Z-buffer 2*2*size of display buffer) = 64MB (4x antialiasing = 256MB)
- Shadowbuffer (rendering into a kind of Z-buffer from the lightsource to create realistic shadows) 16MB
- Accumulation buffer effects like motion blur (very expensive, a good blur could take 4 to 32 frames) or depth of view could make us want another 4-32*16=64-512MB

I for one could easily use more then 1GB of onboard graphics memory.

Interesting trend.

[Frag-A-Muffin]

The first thing that jumped out at me was this interesting trend I see on the chart.

The companies that have long red lines (meaning the time it took for them to ship since their announcements, ie HYPE) are all gone! :)

The ones that kept a relatively consistent schedule are still around. Once again, a smart business plan wins, not super-hyped, non proven stuff. ;) (Yeah .. I know. I'm biased. Nintendo Forever! :)

(On a side note, I wonder how long the line would've been for the xbox! :)

Where are the real physics engines?

[linzeal]

Physics specalized processors? Can anyone show some nice linkage for them? That sounds like the next step for games today complete and utterly lifelike physics engines instead of scripted crap. Would make mapping much easier as well, imho. I know of geomod tech from the people that did red faction and freespace but what else is out there up to and including programs or languages for astrophysics and geomorphology simulations?

Re:Where are the real physics engines?

[Telastyn]

The only reason I mentioned the engine is because I'd love to design/write (I'd rather design) games one day, and that seems like something that would be a great thing for games, rather than re-implimenting Newtonian physics over and over again, there would be something akin to openGL that would have preset functions for things.

The processor itself would likely be a little specialized to handle (x,y,z,vx,vy,vz) style location/speed vectors and the such.

The closest thing I've seen to it was something one of my Materials Science TA's wrote to show and simulate forces on beams.

The real beginning of 3D hardware

[Animats]

The first real 3D hardware was the Evans and Sutherland Line Drawing System 1, in 1969. This was the first graphics device with a hardware 4x4 matrix multiplier, the basic geometry engine component.

I saw one once, at Case Tech, in 1969. About six racks worth of hardware. Nobody really knew what to do with it.

Re:Demo scene.

[Cruciform]

Skaven did the music for Bejewled (Windows Version).

Palm version here, but I'm guessing it lacks the tunes :)