#: 19628 S12/OS9/68000 (OSK)
    20-Jan-94  04:00:57
Sb: #19623-CDI Computer News
Fm: Eric Crichlow 71051,3516
To: Bob van der Poel 76510,2203 (X)

Bob,

    Some good looking suggestions there. I'll give them a try. However, as I'm
starting to get a little anxious about the ability to pull out some faster
performance, how about this...

    I see that memcpy () is quite a bit more efficient than I thought. However,
it has to be written to check and deal with a number of situations, possibly
some error checking, definitely a check of the length of the data to be moved
so that it knows whether there will be a remainder if longword moves are
performed that would require a final move.b and/or move.w. Since I know exactly
what the size of each move will always be, and since I know it is divisible by
4, wouldn't I save some time in having a shortened, specialized version of
memcpy() designed specifically for this task? And yes, I would definitely be
interested in the disassembly of memcpy() if you can find it.

  > I don't believe that Kevin used the pixacc in the 68070 for that (or
  > anything else).

    Actually, its my understanding that he does use the PIXAC for the mouse
pointer. I think it was Kevin himself who told me that. I have tried to get
info on using the PIXAC from 'C'. Only one person had even played with it, and
he'd done it in assembly, which I'm trying to avoid dealing with if at all
possible.

..Thanks...
..Eric...