Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!

1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


Member Since 25 Feb 2008
Offline Last Active Today, 06:08 AM

Posts I've Made

In Topic: Copy texture between contexts

18 December 2014 - 03:24 AM

Assuming you want to use the passed-in resource with a different device/context (than it was created with), take a look at resource sharing.

You might need to create a new texture resource on the first device (so you can correct sharing flags) and share that texture with the second device. Now you can copy (using the first context) the rendered frame to the new resource (and use it with the second context/device).

In Topic: Determine Shader Model for Compiled Shader Object

17 December 2014 - 10:21 AM

You should use the provided macros (defined in d3d11shader.h):


To get the shader type (returns a D3D11_SHADER_VERSION_TYPE):


where version is the value of D3D11_SHADER_DESC::Version

In Topic: Copy texture between contexts

17 December 2014 - 10:08 AM

Can't you simple make a copy of the resource handed over to you ?

In Topic: Sampling the Render target while rendering to it

28 August 2014 - 09:38 AM

If you want a overdraw 'debug' shader, you should simple render the scene using additive blending with a static color (for example float4(0.1, 0.1 ,0.1 ,1) ).

places with little overdraw will be quite dark while places with a lot of overdraw will be white.


In case everthing is black or white (the minimum/maximum), simple adjust the fixed color. for example change it to float4(0.25, 0.25 ,0.25 ,25) if it is to dark, float4(0.02, 0.02 ,0.02 ,1) if it is to bright. Even better is to have a user-controlled slider ;)


You could do it dynamically (determine the blend color based on the minimum/maximum overdraw count) but that will be quite a lot of work for a debug feature... If you chose to calculate it dynamically, I recommend something like this

In Topic: What you think of my Depth Pass AA?

25 August 2014 - 11:53 PM

The second image is the 8x MSAA and my Depth Pass AA implematation.

What is the point of using 8xMSAA + custom AA ?

Algorithms like FXAA exists because MSAA has quite an impact on performance/memory.

how does it look with 1xMSAA + your custom pass ?


It looks good but we need a whole lot more info in order to make a decent judgement.

- performance / memory usage

- compaire against other existing techniques (FXAA etc)

- how does it compaires to 'true' AA (supersampling)


I can't find zip about TXAA anywhere.

txaa looks to be a combination of 2xMSAA and temporal reprojection