I''m working with DirectDraw 7.0 and I have an assembly additive blending function but it is too slow. I need some help on how to speed it up.

Yeah man, you could just use BSP like quake does, and forget about all else... I use a combo of portals between bsp-encoded sectors so I can iterate whole objects, or whole rooms in 3d , or whole areas of space if not a room. Iteration rocks !!

If your asm code is *well commented*, you can sent it to me and I will look at it over the weekend. (email in my profile)

*travois*

* Utilize somekind of RLE blitting scheme (encoding all the completley transparent pixels for example and/or encode the ''runs'' as usual to reduce memory and cache usage)
* Since your obviously aiming for the x86 platform you could probably use MMX to speed things up, if you aren''t already.

Oh and keeping both the surface you are blitting from and the surface your blitting to in system memory usually speeds things up quite a bit. The same old "never read from video memory unless needed" applies here.

"This album was written, recorded and edited at Gröndal, Stockholm in the year of 2000. At this point in time money still ruled the world. Capitalistic thoughts were wide spread. From the sky filled with the fumes of a billionarie''s cigar to the deepest abyss drenched in nuclear waste. A rich kid was a happy kid, oh..dirty, filthy times. Let this be a reminder."
- Fireside, taken from back of the Elite album