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 10 Aug 2007
Offline Last Active Feb 25 2015 09:52 PM

Topics I've Started

User Clip Plane Aliasing Issue

27 January 2009 - 11:41 AM

I've been having some weird aliasing-type issues the past couple of days with my stencil mirrors and their associated user clip planes. In the particular scene in question, there is a floor and mirrored wall that are flush against each other, forming a perfect right angle. Both the wall and floor are perfectly planar. The reflection renders fine, except for a weird 0-2 pixel border at the bottom of the mirror that flickers wildly with the camera angle. I can reproduce it with only this one floor and wall in a scene, so, its not something external to that. I've investigated in PIX and found that the stenciled off region is exactly the pixels that it should be, forming a line without gaps along the wall. The floor always renders completely ok, the deviant pixels in question are on the mirror itself. The pixel history also seems to indicate that these pixels are being discarded by the clip plane, and not as the result of a depth/stencil test failure. I have a user clip plane defined to match the mirror plane perfectly. If I disable this user clip plane, then the weird edge artifacts go away. Though, this hurts performance a lot in some of my mirror heavy scenes and introduces clipping artifacts in other scenes. I tried implemented Near-Clip Plane Shearing as per a paper I found on Nvidia's website, and it suffers exactly the same problem as the user clip planes do. Has anyone encountered anything similar to this and/or have any advice on what to try next? I could put a frame around the mirror which would hide the problem, but, that isn't completely to the spec of the scene.

[XACT] - Determining What Wavebanks A Soundbank Uses

17 October 2008 - 05:27 AM

I'm looking into the process of adding in support for sound, and Xact seems to do basically everything that I'm looking for, with the added bonus of having a content creation tool pre-built for me. However, I'm puzzled by something here. How do you determine what wavebanks a soundbank/cue needs? The names of the wavebanks seem to be stored in plain text near the start of the .xsb file, but, that file format isn't documented (at least, no where that I've seen) and possibly subject to change. The Xact project file does contain this information in a human readable file format, but, it seems silly to be shipping that .xap file just for this purpose. Though looking inside of Sounds.xsb in tutorial 3, there isn't any mention of Sounds.xwb. Plus, ideally, one would be able to query somehow what wave banks a specific cue needs, so that one isn't loading in any unnecessary wave banks into memory. Or does Xact take care of all of this behind the scenes for me? The tutorials are loading wave banks directly, but, what happens if I tell Xact to play a Cue without its underlying wave bank loaded in? I don't see any Xact notification callbacks for missing wave banks. Even if you have all of the sound banks and wave banks loaded into memory, there doesn't seem to be a way of querying for what wave banks are needed by what cues. Also, just noticed that there's no audio programming forum here... maybe there should be?

Enabling Transparency AA In DX9 On ATI Cards

11 August 2008 - 10:59 AM

Is there a way of enabling transparency anti-aliasing in DX9 on ATI cards? I tried this method from Nvidia: (http://developer.nvidia.com/object/devnews028.html see the tip of the month section) pd3dDevice-> SetRenderState(D3DRS_ADAPTIVETESS_Y, (D3DFORMAT)MAKEFOURCC('S', 'S', 'A', 'A')); And while that works on Nvidia cards, it has no effect on any Ati cards.