Re: [info-performer] a question about transparency

New Message Reply Date view Thread view Subject view Author view

From: MLM Veraart (veraart++at++fel.tno.nl)
Date: 09/26/2002 03:36:15


You have a Z-buffer problem. Your polygons are co-planar and some of them get
lost in the Z-test.
If you draw your special effect in the transparent buffer you just disable the Z-test
with a draw callback.
Otherwise use the pfLayer technique (stencil planes) to cope with the coplanar geometry.
Use a base layer that does no update to the color-buffer and is as large to contain all the
animated geometry.

Mario Veraart

P.S. Please adjust your email tool so you send only plain text to the mailing list.
     A reply was not possible without copying from the original message.

----------------------------------------------
Archer Chen Wrote:

Hi all,
 
I find that I can't read the attach file, so I post it again , sorry.
 
I had a problem about transparency , and have no idea to solve it .
Please someone can give me some suggestion.
 
My problem is that I can't show the Second transparent texture where after the other transparent texture.
Because , the first transparent texture will make the Second texture to transparent on the overlape region.
How can I make the first transparent texture near camera unable to affect the Second texture's transparent effects.
  Please give me a solution to handle it .
  Thanks.
 
 
  pfGStateMode(shared->rain.gstate, PFSTATE_TRANSPARENCY, PFTR_BLEND_ALPHA);
  pfGStateMode(shared->rain.gstate, PFSTATE_ALPHAFUNC, PFAF_ALWAYS);
  pfGStateMode(shared->rain.gstate, PFSTATE_ENFOG, 0);
  pfGStateMode(shared->rain.gstate, PFSTATE_ENLIGHTING, PF_OFF);
  pfGStateMode(shared->rain.gstate, PFSTATE_ENLPOINTSTATE, 1);
 
  pfGSetAttr(shared->rain.gset, PFGS_COLOR4, PFGS_PER_VERTEX, shared->rain.colors, NULL);
  pfSetVec4(shared->rain.colors[i], 0.6, 0.6, 0.5, 0.3);
 
 
 
Regards
archer chen


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Sep 26 2002 - 03:36:54 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.