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

Re: Anyone gotten A-10 or F-18 2.0 demos to work??



In article <4e0ooo$f5l@news.voicenet.com>, chessman@voicenet.com (ChessMan) 
declares...

>  The program is 
>essentially emulating System 6.0.7 which Apple stopped shipping with their 
>computers over 3 years ago
[snip]
>The software runs without any control panels or inits - a 'real' mac user 
has 
>many of these little programs at the tip of their finger tips - and of 
course 
>it's not multi-processing aware - to run two applications; you must run it 
>under windows and start it twice.  So it really doesn't have the  feel of a 
>MAC - but hey - it plays that little Risk game  like a champ.

I don't actually disagree with any of this, but thought it might be a good 
opportunity to remind people/clue in newcomers about the long-term ARDI 
strategy, at least as I, a registered Executor customer with no other ties to 
ARDI, understand it.

Right now, they're doing everything they can to get a commercially viable 
version 2.0 out the door.  The revenue from that will fund development for 
networking support, serial port support, better sound, more colors, etc.  
This may raise yet more money.  But the longer-term goal, for version 3.0 or 
so, I guess, is to make Executor a base onto which one will install a real 
copy of System 7.x and run all MacOS functionality.

There are three or so obstacles:

1) Dirty/Clean engineering.  ARDI has figured out most of the Mac ROM 
functionality through "clean-room" techniques; they haven't broken any 
copyright or patent agreements to figure out what the ROMs do.  Finding out 
the remaining undocumented features of the ROMs and other low-level features 
involves hiring a new, "dirty" team of engineers to reverse-engineer and 
otherwise mess around in Mac innards to find out such details. Those details 
will be used to draft pure specifications, which will go to the clean team, 
allowing them to cleanly implement that functionality.  A couple of 
iterations of this, and the above goal could be reachable, at least in 
theory.  So the 6.0.7 functionality is a way of *funding* System 7.x 
droppability.

2) PowerPC.  Mat Hostetler, the syn68k (ARDI's Motorolla emulator) has 
indicated that a PPC emulator won't be too much work, so Executor should 
eventually run PPC binaries.  He's even said that developers could include 
native x86 code, and that Executor could declare itself and allow Mac apps to 
run natively on the x86 (or whatever other processor Executor is ported to) 
if they were developed with such a "superfat" binary.  Tantalizing.  This, 
too, awaits more moolah.

3) Copeland.  System 7.x won't be a poor OS laggard for too many more years 
before Apple, provided it's solvent, releases Copeland, the next major 
upgrade.  (I know it *sounds* like I'm starting a religious war, but I'd 
rather not, as such wars on this group get intensely weird.  Best that we can 
all just emulate each other, IMHO :-)).  Presumably, once (2) is 
accomplished, (3) won't be a big problem, but there may be technical details 
I don't know about.  There are also Common Hardware Reference Platform specs 
I'm fuzzy on.

In other words, the 6.0.7 limitation isn't final.  BUT, and this is the 
tricky bit, it's against ARDI's long-term interests to spend much time 
hacking 7.x functionality now when it will eventually turn all that over to 
Apple.  They need just enough System 7 support to get a customer base; after 
that, lower-level issues have greater marginal utility.

Again, none of this is official, though the dirty/clean room stuff is 
summarized from one of Cliff Mathews' posts.

Scott Shuchart
shuchart@fas.harvard.edu



Follow-Ups: References: