Thank you,

I would *still* like to know why pixel_to_bmap() wont write pixels to a 24-bit bmap, though. I can work around it using 32-bit bmaps, but its still annoying to not know the why of the problem.

Because in your code you used 8888 as format. But 24bit needs to be 888.

How exactly did you test this,

changing bmap_createblack()to 24 bit for pnlFrom.bmap
changing bmap_createblack()to 24 bit for pnlTo.bmap
changing bmap_createblack()to 24 bit for pnlFrom.bmap and pnlTo.bmap

all work...?

I tested using 8888 and 24 bit bmaps, then 8888 and 32 bit bmaps. After that, i tried 888 and 24 & 32 bit bmaps. the only time anything worked was when i used 8888 and my bmaps were 32 bit.

I actually ended up dropping the bmap_createblack() call and using 2 bmaps instead. (BMAP* bmp1 and BMAP* bmp2). As stated, wdlmasters code worked perfectly once I saved the bmaps as 32 bit in Photoshop - pixel_to_bmap() also worked fine as well.

Once I saved the bitmaps in photoshop as 24 bit( as well as changing the code to use 888 format), pixel_to_bmap() simply wouldnt write pixels to the bitmap - as well as wdlmasters code.

Curiously, the format returned by bmap_lock was 8888 - period. Whether I was locking a 24 bit bmap or a 32 bit bmap - it returned 8888 like it thought it was dealing with 32-bit. At the time, I didnt pay any attention because the manual had stated that formats are automatically "upconverted" depending on the video card. Maybe bmap_lock() is the culprit?