[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: What specs would your Apple ][ clone have today?

Tristan Mumford wrote:

David Schmenk wrote:


sicklittlemonkey wrote:

-snip-

Actually, PalmApple was a port of YAE.  It sort of worked, but took
about 30 minutes to boot ProDOS.  Yours truly wrote Appalm (a much more
clever name that PalmApple, I thought) in 68000 assembly and a little C.
The only thing used from YAE was the disk nibblezation and arm stepping
logic.


Oh I see. I misunderstood the lineage. From what I read I thought that
Appalm was a later iteration of PalmApple.
I sort of hijacked the project from Hilary. He did a good job of getting something to actually run on the Palm. It was much easier to get my code working by replacing PalmApple in sections. I didn't mean to suggest that Hilary hadn't done lots of work on PalmApple. Many of the hard PalmOS interfacing problems were solved in PalmApple. Without it, Appalm would never have happened. 


You wrote that? Very nice. Mind you I do tend to make a backup of my palm
before I use it. It has a tendency to explode on my T3, but the T3 is great
for upsetting some programs anyway.
Mostly due to allocating a linear 64K chunk of memory and then turning off the minimal memory protection mechanism. And the fact that PalmOS is a cobbled together piece of crap. 


Now this is where I think an Apple II clone would get interesting.  My
Treo 700 runs Appalm close to native speed of a real Apple II.  Appalm
has no speed regulation and just runs flat out.  So here is an XScale at
312 MHz, emulating a 68000, emulating a 6502.  At about the right speed.


Well, I guess you could always rewrite the core in ARM assembly. Or just do
the "easy" thing and compile a C based 6502 core as an ARMlet. That would
probably give a massive performance boost, removing the 68k middleman.



But then I'd have to implement some sort of speed control :-)

Dave...