softviewer update

jReality backends
Post Reply
timh
Posts: 28
Joined: Wed 28. Jun 2006, 20:30
Contact:

softviewer update

Post by timh » Tue 21. Nov 2006, 13:00

The softviewer does Reflectionmaps now (this is quite slow of course) and evaluates the transparencyEnabled flag from the renderinghints for compatibility reasons (I am still not happy with this flag since it can cause a lot of confusion...).
BTW is it true, that the ImagaData now allways gives 4-byte per pixel arrays? the softviewer used the length to determine wether a texture has alpha and therefore the textured triangles need to be treated like transparent ones. Allways getting 4-byte textures therefore gave those textured triangles quite a performance penalty. Now of course the transparencyEnabled flag can force the renderer to ignore the (in fact absent) alpha.
Is there any other way to get information on wether a texture has transparency !=0 or not?

Tim

User avatar
steffen
Posts: 186
Joined: Fri 16. Jun 2006, 13:30
Location: TU Berlin
Contact:

Re: softviewer update

Post by steffen » Thu 7. Dec 2006, 16:06

timh wrote:The softviewer does Reflectionmaps now (this is quite slow of course) and evaluates the transparencyEnabled flag from the renderinghints for compatibility reasons (I am still not happy with this flag since it can cause a lot of confusion...).
Looks great!
timh wrote:BTW is it true, that the ImagaData now allways gives 4-byte per pixel arrays? the softviewer used the length to determine wether a texture has alpha and therefore the textured triangles need to be treated like transparent ones. Allways getting 4-byte textures therefore gave those textured triangles quite a performance penalty. Now of course the transparencyEnabled flag can force the renderer to ignore the (in fact absent) alpha.
Is there any other way to get information on wether a texture has transparency !=0 or not?
ImageData always had 4 bytes for each color as far as I could see from cvs history.

Steffen.

Post Reply