> I think the problem is with mapping of video memory at boot time, but
> Michael Zucca might be able to explain it better since he's the one who
> managed to get internal video working at one time.  I'm not sure how hard
> it would be to fix, tho.  I think that a hack to get it to boot might not
> be too hard, but the clean and "correct" solution might be a bit more
> difficult.  Good luck if you want to work on it, we're always looking for
> more volunteers!

Well, when I did it "back in the day" the trouble was two routines in
machdep.c that setup video and some other related memory things. The
code basically attempts to find and size video and NuBus stuff. The
trick to getting most PowerBooks going was to skip the NuBus probe
and set the base address of video properly. Other PowerBooks like
the 160/180's had some other bizzare problem where the mapping
used during a serial boot worked but the mapping used in intvid
booting broke. So I had a bit of code that fooled the machine to
map memory like a serial boot and things just sort of magically
worked. Weird stuff.

Scott moved the base address stuff into a table so that's about
half the problem. The other half, the weird memory mapping business
and the NuBus probe skip bear a more in-depth analysis so that
we're doing things "right" rather than just crossing our fingers.

I've gotten away from the "plug and pray" approach. I'm more
interested in doing things, once, correctly so that they will
stand the test of time. When I release Intvidtest2 I think
you'll see that ideology at work :)