GDC05 presentations from ATI
- Thread starter tEd
- Start date
squarewithin
Newcomer
I like how everyone flips out and comments on the fact that someone may have something that they can't talk about, but when someone posts some informative with actual facts it gets no love.
I still admire Alex Evans (to quote Mark Healey (Lionhead graphics artist): "he has a brain the size of a planet, and is incredibly creative, too"). His presentation has some very nifty ideas (not all of which are immediately applicable to games). Especially the handmade illustrations are pretty. 
I hearby decree that all persons who have signed an ATi NDA pertaining to upcoming product releases are now free of them and may speak at will, preferrably in a thread right here on B3d.
(Interested ATi legal personnel who might wish to see my credentials may contact me at the following address:
Dr. No
Destroyed Bauxite Mine
Crab Key
Jamaica 747-F111*
*When writing, don't forget to ask for our free color brochure illustrating our latest Missile Toppling Systems!
Thank you.)
Ok, guys, you heard the man--spill it!--DaveB.
(Interested ATi legal personnel who might wish to see my credentials may contact me at the following address:
Dr. No
Destroyed Bauxite Mine
Crab Key
Jamaica 747-F111*
*When writing, don't forget to ask for our free color brochure illustrating our latest Missile Toppling Systems!
Thank you.)
Ok, guys, you heard the man--spill it!--DaveB.
pocketmoon66
Newcomer
[maven] said:
Wow that far cry paper is interesting:
should be:
The latter one is easier to type, the shader is shorter and it runs faster.
It's a good way to show the world how to write shader code...
Code:
const float4x4 cMatViewProj;
const float4 cPackedInstanceData[ numInstances ];
float4x4 matWorld;
float4x4 matMVP;
int i = IN.InstanceIndex;
matWorld[ 0 ] = float4( cPackedInstanceData[ i ].w, 0, 0, cPackedInstanceData[ i ].x );
matWorld[ 1 ] = float4( 0, cPackedInstanceData[ i ].w, 0, cPackedInstanceData[ i ].y );
matWorld[ 2 ] = float4( 0, 0, cPackedInstanceData[ i ].w, cPackedInstanceData[ i ].z );
matWorld[ 3 ] = float4( 0, 0, 0, 1 );
matMVP = mul( cMatViewProj, matWorld );
OUT.HPosition = mul( matMVP, IN.Position );should be:
Code:
const float4x4 cMatViewProj;
const float4 cPackedInstanceData[ numInstances ];
int i = IN.InstanceIndex;
IN.Position.xyz = IN.Position.xyz * cPackedInstanceData[ i ].w + cPackedInstanceData[ i ].xyz;
OUT.HPosition = mul( cMatViewProj, IN.Position );The latter one is easier to type, the shader is shorter and it runs faster.
It's a good way to show the world how to write shader code...
StellaArtois
Newcomer
Geeforcer said:
I'm sorry to bump this thread but there is an interesting new presentation from ati which wasn't there before http://www.ati.com/developer/gdc/Tatarchuk-ParallaxOcclusionMapping-FINAL_Print.pdf
and maybe somebody could make a demo out of it this because i really would like to see this in action.
I know that there are some demos with parrallax and horizon mapping but it seems that this one is a improved combination of these techniques
and maybe somebody could make a demo out of it this because i really would like to see this in action.
I know that there are some demos with parrallax and horizon mapping but it seems that this one is a improved combination of these techniques
