Jump to content
  • Advertisement
  • entries
  • comments
  • views

Direct3D 10.1 is alive

Sign in to follow this  


Well a good evening to everyone (anyone?) still reading this journal [smile]

Things are still pretty hectic for me at the moment so I'm getting approximately zero time to just play with technology or get involved in any of the online forums. Fingers crossed it should sort itself out by the end of April and I can get back to things again soon.

That said and as much as I'm enjoying a break from coding I'm getting a little twitchy and wanting (needing?) to get working again [grin]

So I've managed to find a few hours to uninstall Vista SP1-Beta, install SP1-final, uninstall Nov'07 DXSDK and install March'08 DXSDK. The end result being that I can now compile and execute Direct3D 10.1 applications, unlike my previous attempt.

Now the problem I've realised is what do I actually want to do with D3D10.1. No, really, I can't think of much.

The main thing I wanted to mess with was the new CopyResource() semantics with bit-fiddling between ARGB128 and BCn[/n] texture formats. I figured this could be a simple and neat little way to get data into a compressed format. Say runtime deformed/generated normal maps being generated from a height map rendered on the fly.

I commented in my section of the D3D10 book that 10.0 was a bit lacking in getting data into the block compressed formats so I was quite keen to try out the improvements for 10.1.

However, the documentation I have doesn't make it easy unfortunately. The diagram of getting a 16 byte ARGB128 pixel twiddled into a single BC5 block is easy enough, but I'm not entirely sure about the source and destination format sizes. If a single ARGB128 pixel describes the usual BC 4x4 grid do I effectively down-sample to a 1/4 sized texture which gets expanded back up again:

1024x1024 (full colour) -> (RENDER) -> 256x256 (ARGB128 semi-compressed) -> (CopyResource) -> 1024x1024 (BC5)...


Then, even if I can figure that one out I'm not entirely convinced its worth all the effort. Given that techniques like parallax and ray-traced bump mapping always require a height map I'm left thinking that the few extra instructions to just run-time derive a normal map from a height map are plenty sufficient enough. Why store derived data, even if it is compressed??

Anyway... Think I'll just install Visual Studio 2008 Team Suite and watch a film instead.
Sign in to follow this  


Recommended Comments

If only MS would just completely agree that Vista is an epic fail and allow DX 10 support on XP all would be right with the world again.

Share this comment

Link to comment
I hear you... both, but especially the movie part.


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
  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!