You may want to () around the pitch shift.
Another odd thing, though it shouldnt be a problem, is that blue is defined in the loop while they others arent. Strictly syntax if they are both UCHARs still.
I dont see anything inherently wrong with the logic though.
-Geoff
for (int index_y = 0; index_y < bitmap.bitmapinfoheader.biHeight; index_y++)
{
for (int index_x = 0; index_x < bitmap.bitmapinfoheader.biWidth; index_x++)
{
UCHAR blue = (bitmap.buffer[index_y * bitmap.bitmapinfoheader.biWidth * 3 + index_x * 3 + 0]),
green = (bitmap.buffer[index_y * bitmap.bitmapinfoheader.biWidth * 3 + index_x * 3 + 1]),
red = (bitmap.buffer[index_y * bitmap.bitmapinfoheader.biWidth * 3 + index_x * 3 + 2]);
UINT pixel = _RGB32BIT(0, red, green, blue);
back_buffer[(index_y * ddsd.lPitch >> 2) + index_x] = pixel;
}
}
[This message has been edited by Theses (edited November 26, 1999).]
[This message has been edited by Theses (edited November 26, 1999).]
int width_padding = ((width+3) & ~3) - width;
(What this line does is temporarily round the width up to the next multiple of 4, then subtracts it from the actual width. So, 599 temporarily becomes 600, and 600 - 599 = 1, so you have 1 byte of padding per scanline)
Every scanline has this many pixels of padding.
As a reminder, For 24bit bitmaps, every pixel padded adds 3 bytes, so don't forget this 
BTW Microsoft should recall the BMP format, too many people are getting this problem.. I must have answered this same question a dozen times just on this board alone
