Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 13 Aug 2000
Offline Last Active Jun 26 2013 09:46 AM

Topics I've Started

Havok Physics

06 May 2013 - 07:05 AM

I'm doing some testing of the Havok Physics sdk to see if it meets the requirements of my engine. Does anyone know if a hkpRigidBody can be created with out any shapes added? I'm trying to create a dynamic object with out any collision but I'm getting a crash when I try to create a new  hkpRigidBody with bodyCinfo.m_shape set to NULL. Seems strange that such a popular physics sdk wouldn't support such a basic feature.



Direct3d slowing down other threads

16 April 2013 - 11:44 AM

I've got a MFC program that does all the rendering in the main tread and uses a seperate thread for physics. For some reason when I comment out the main render update function my physics thread runs faster even though no communication is happening between the two threads. If I replace the render update function with code that wastes time in a while loop the physics thread also runs faster. The Direct3D device is being created with D3D11_CREATE_DEVICE_SINGLETHREADED. Does anyone what might be going on here?

Direct3D custom z testing.

27 September 2012 - 10:34 AM

I want to write one pixel depth value to the z-buffer but test against another value with a offset added. This is to overcome objects that are z fighting against the terrain. I want to write the normal depth value to the z-buffer but have a offset added when testing agains the terrain z values. Is there any easy way to do this?

Creating adjacency information in a domain shader.

05 May 2011 - 01:02 PM

I'm trying to tessellate a quad for some displacement mapping. I also need to do some edge detection in a geometry shader once my quad has been tessellated but I can't seem to find a way to generate any adjacency information in the domain shader. Does anyone know how I might go about doing this? If I could figure out how the SV_DomainLocation coordinates are generated based on the tessellation factors are generated then I might be able to predict where the neighbor vertices would be.

Question about spherical spreading of light.

15 December 2010 - 04:44 AM

A light source at distance d1 from a object attenuates at 1/(d1*d1). If the point at which the light hits a object is at distance d2 from the camera shouldn't the light also attenuate by 1/(d2*d2) as it travels from the reflection point back to the camera? I know sonars works this way but I don't really see lighting calculations done like this. Anyone know why?