Author: jeff (oh-lyndhurst2a-114.clvhoh.adelphia.net)
Date: 11-14-2001 05:43
Why do we have to port TOS? I would suggest this:
1. Port MiNT to whatever system. This would involve some rewrites of portions involving 68k assmebly in porting to other CPU's, like Intel. Now I know eveyone is about to jump on me about suggesting MiNT, but hear me out...
2. Port JET to whatever platform. JET stands for "Just Enough TOS" and provides the necessary GEMDOS routines that MiNT does not provide yet. JET is used to boot MiNT on a standard Mac. Yes, MiNT can run natively on a standard 68k Mac. And it works fine and has for many years.
3. With a GEMDOS ported, the GUI needs to be ported. Port fVDI to whatever platform. This provides a complete VDI replacement for the system. It can most likely be ported with ease to other hardware, including PC's.
4. Port XaAES for the GUI. XaAES is free, reasonably stable, and fast. No reason to try and use Atari's AES's, which were quite slow.
This replaces all the aspects of TOS. For TOS of any sort to run, it needs the GEMDOS, the VDI, and the AES. Other features can also be ported, such as the GDOS, the XBIOS (for Falcon-compatible sound), and hard disk drivers.
Notice my ideas use free software with available source code. No need to search for Atari's original code. And it provides a multitasking system. The OS would be native and an emulation core could be integrated on certain platforms to run 68k binaries.
PC-GEM is crap. I don't know if anyone's tried to use it, but it doesn't compare to any modern Atari AES, like MultiTOS (aka AES 4.x), XaAES, or N.AES.
And if we want support for the project, why not go through TOS International, which already exists?
Just some thoughts! My idea won't work, of course, for Magic, but ASH would have to port that themselves if they wanted it to run natively.
jeff
|
