DirectX Next

[DemoCoder]

(I must say, excellent game and graphics) X-Box TopSpin

Off topic warning but:

I used to play tennis games on the amiga back in the good old days. Haven't played one tennis game (though i haven't tried Virtua Tennis) that i have been interested in since then until i tried Top Spin. Awesome game.

Yes, top-spin has amazing gameplay and even more amazing graphics and animation. Some of the instant replays are amazing for the smoothness and naturalness of the player's motions. Far less "jerky" transistion than most sports titles.

More than that, I am impressed by the fact that the entire stadium is modeled, every audience member is a 3D model, and they all clap/animate at different rates. All geometry on the screen is also fully stencil volume/shadow buffered, even down to the individual people in the audience.

The game also features nice hair on the players, bloom lighting effects, and a host of other stuff. If it had anti-aliasing, I'm sure people would be accussing the cut-scenes of replays of being pre-rendered.

I was really wowed by this game, and X-Box live play is addictive. The other game I'm playing alot now is Return of the King on the PS/2. I got the PS/2 version for online play, even though the gfx are somewhat inferior. ROTK is one of the most "polished" game's I've played in awhile.

 

[OICAspork]

Out of Curiosity...

How does the Roadmap of OpenGL compare with the functionality of DX Next in the same time frame? OpenGL 2 seems slower in coming than I had expected, is that because it is getting beefed up, or something else?

 

[Ilfirin]

Actually there are other possibilities to create instances of geometry. Take for example the possibility of automatically looping an index buffer. So you store the indices (into a vertex buffer) for say a single sphere and if you loop over that list of indices multiple times you get multiple spheres. To get the spheres into different positions you can then use a second input stream which runs at a lower rate (instead of reading a value for each index you read one every x, say 100, indices) - this stream could indicate the transformation of the sphere or a lookup value in a table of constants. Lots of possibilities... and this stops the nasty driver call overhead etc... not to mention it would be much cooler if you could actually just procedurally generate your sphere from an input containing centre point and radius. Ah well the developer in us all will never be happy with the functionality given

K-

I'm curious how this would work.

According to the docs, you can't use an index buffer with the frequency divider, so would you have to make several instances of the mesh inside the vertex buffer (with 3 unique vertices for each triangle, or stripify everything) and use the stream frequency divider to change transformation matrices for each mesh?

That doesn't sound like a very good idea..

P.S. This isn't what was meant by the mesh instancing ability in DX Next was it (just vs.3.0)?

 

[JohnH]

I'm curious how this would work.

According to the docs, you can't use an index buffer with the frequency divider, so would you have to make several instances of the mesh inside the vertex buffer (with 3 unique vertices for each triangle, or stripify everything) and use the stream frequency divider to change transformation matrices for each mesh?

That doesn't sound like a very good idea..

P.S. This isn't what was meant by the mesh instancing ability in DX Next was it (just vs.3.0)?

Rate dividers don't work on the current ref-rast when using indexed prims, but they do get passed down to the driver so if your HW supports it... Ouside of this you just need to pick up a modulo value for the index stream from somewhere TBD.

John.