It just means there's a 64 byte span between one bitplane line and the next, and so you can address individual lines very fast by using a single shift instruction to multiply the Y coordinate by 64.
For a typical lowres screen you would have 64*256 bytes per bitplane and set data fetch registers as usual, but set the bitplane modulos to 24 so it reads the usual 40 bytes and then skips over the last 24 to get to the next line.
Those extra bytes between the lines are not lost or anything, they are fully useable like any other chipmem, just in a poor arrangement.
Actually, I will adjust my introsystem to such aligned bitplanes. To only have to do an "lsr.w #6,dx" to determine the Y coordinate speeds up the pixel drawing considerably.
But ATM I'll stick to standard 40-byte aligned planes b/c it would be too much of a work to change my framework by now.
