JB,
Thanks, after modifying it so that I could compare apples to apples, I came up with the following function. It tests to between 7-8 times as fast as the original, which is substantial. I'm pretty sure the overhead is in the binaryEncode function, which you only do 4 times, while I call it once for each pixel. Thanks for your help!
For those interested, in the messagebox type: go URL "http://www.altuit.com/webs/altuit2/RunRev/altMakeTrans.rev";
best,
Chipp
function altMakeTransAlpha pTransColor,pImgData,pAlphaData --> pTransColor IS THE COLOR TO MAKE TRANSPARENT --> EX. "255,255,255" IS WHITE --> pImgData IS THE IMAGEDATA FROM WHICH THE MASK IS CREATED --> pAlphaData IS THE ALPHADATA OF THE ORIGINAL IMAGE --> RETURNS THE ALPHADATA TO USE FOR TRANS MASK
put binaryEncode("C",0) into t0RGB
put t0RGB into tAlphaTrans put binaryEncode("C",item 1 of pTransColor) after t0RGB
put binaryEncode("C",item 2 of pTransColor) after t0RGB
put binaryEncode("C",item 3 of pTransColor) after t0RGB put 0 into tPos
repeat forever
put offset(t0RGB,pImgData,tPos) into t
if t=0 then
exit repeat
else
add t+3 to tPos
put tAlphaTrans into char tPos/4 of pAlphaData
end if
end repeatreturn pAlphaData
end altMakeTransAlpha
_______________________________________________ use-revolution mailing list [email protected] http://lists.runrev.com/mailman/listinfo/use-revolution
