Re: Second Sight

[mjmahon@aol.com (Michael J. Mahon)]

vector wrote:

>"Sheldon Simms" <sheldonsimms@yahoo.com> wrote in message
>pan.2003.06.02.04.11.56.167701@yahoo.com">news:pan.2003.06.02.04.11.56.167701@yahoo.com...
>> One problem with all this is that these really are things
>> that a IIgs can't do. With the hypothetical card, it would
>> be the card doing these things. In fact, it would seem kind
>> of silly to write 65816 code to do this stuff when it's just
>> being emulated (although quickly). Seems like it would be
>> better to just let the Xscale handle this stuff natively.
>>
>> I do think the project would be interesting, but I have to
>> agree with Michael's observation that the Apple II is more of
>> a peripherhal for the card than the other way around.
>
>This conversation reminds me of another old thread, where it was asked
>"wherein does a machine's soul lie?"  Is it the CPU?  The operating system?
>The particular configuration of parts on the motherboard?
>
>I've finally come up with my answer for that question.  For me, if a machine
>can run GS/OS or ProDOS without another OS underneath providing an emulation
>layer, then that machine is a IIgs, no matter how weird its specs are.  To
>put it another way: I understand that another CPU emulating a 65816 is still
>a type of emulator, but if the only way to get that processor to do anything
>is to program against it as if it were a 65816, then it IS a 65816 from the
>standpoint of the software developer, no matter how strange the system
>architecture may be.
>
>And therein lies the key difference in how you and Michael see this project
>versus how I see it.  You both seem to me to be Hardware Guys, whereas I'm a
>professional programmer.  I don't think much in terms of hardware.  To me,
>if it looks like a duck and hacks like a duck, then it's a duck.  Or an
>'816.

Interesting observation...  Actually, I'm a hardware-software guy--a computer
architect.  I certainly appreciate the notion of virtualizing an architecture.

I also really appreciate these older systems because they are "frozen
in time", and that makes doing things with them an exercise in finesse,
or "resource-constrained" programming (which I, personally, love).

Removing the resource limitations significantly seems to me like
playing poker with two decks, or playing chess on a 12x12 board--
the result still resembles the original in some way, but the strategies
and tactics are quite different.  The game is a different game.

The soul is a different soul.

And, of course, I have no illusions that great new software works
will be written for the "souped up" machine.  (Look at the power
we have available on PCs today, and how little new thinking?  ;-)
Come to think of it, look how multi-megabyte and multi-gigahertz
machines have essentially de-skilled or "de-crafted" programming.
Horribly inefficient bloatware has become the order of the day.
But I digress...  ;-)

The primary thing that a "souped up" machine does (beyond some
modest "souping"  ;-) is to give the owner of the now-unique
machine "bragging rights".  (And, to some minor degree, it tends
to trivialize all the great work done to make the original platform
usable without "souping".)

I should probably say that my own feeling is that the //e was the
culmination of the classic Apple II line, and the IIgs is its Rococo
decline. The complexity of the IIgs, and particularly the complexity
of GSOS, contradicts the simple, clear, elegant design that is
the defining characteristic of Woz's work.

Gad, look at the esthetic abomination of the 65816--with 24-bitness
grafted onto its 16-bit core like some digital Frankenstein.  This is
the kind of hack that gives "upward compatibility" a bad name.

Elegance to me means that a system can be completely understood
and programmed using documentation weighing less than the computer.
(Of course, featherweight computers are an exception.  ;-)

-michael

Check out amazing quality 8-bit Apple sound on my
Home page:  http://members.aol.com/MJMahon/