• entries
    422
  • comments
    1540
  • views
    488796

Part 7: Learning the Programmable Pipeline

Sign in to follow this  
jollyjeffers

80 views

I've tidied up the source code for the cubic patch shader I mentioned in my previous journal entry.

I also managed to create an optimized version of the shader that knocked 6 instructions off and gave me around 500,000 more vertices per second [wow]

64 Instruction Listing and Optimized 58 Instruction Listing. Complete HLSL Code.

You can download all of the files here (751kb)

I think I'm gonna implement some lighting for the surface next. I've messed with a 3-point-per-vertex sample which works but it's slow. Given it's a mathematical surface there should be a mathematical normal, if I can bend my head around that I can probably write some clean(er) code [grin]

Comments appreciated - especially performance related if you have a GeForce 6800/7800 [wink]
Sign in to follow this  


2 Comments


Recommended Comments

Quote:
Very nice Jack.

Thanks!

Quote:
I've only got a 5200 but I'm getting 400 fps.

That aint bad at all [smile] If you do the following you can get quite a good guage of the vertex throughput of your hardware:

1. Switch to fullscreen mode, any resolution
2. Hit PAGE-UP until LOD is 8 and it's saying there are 65536 vertices
3. Scroll up (or down?) with the mouse wheel until the geometry disappears off the screen (reduces/eliminates fill-rate).

If you then watch the 3rd (might be 4th) line of text - "Performance : ___ MVerts/Sec" you can get a reasonable benchmark.

For reference mine is usually between 10.25 and 10.40 @ 158fps [grin]

Jack

Share this comment


Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now