Re: GS Emulator

[David A Lyons <dlyons@apple.com>]

In article <Cso8CK.Iwo@mail.on-line.co.uk> Richard King,
tombo@mail.on-line.co.uk writes:
> [...] Yeah - we all want to run an emulator, behind multiple
> layers of undocumented support-code, rather than the real thing. Not
me,
> I'm proud to say. I still can see the difference between talking to
the
> real silicon, and somebody's overly-complex 'emulation'. I just wish
you
> lot could, too, and weren't being blinded by the hype.

Interesting.  But _how_ can you tell when you're talking to the "real
silicon"?  I certainly can't tell most of the time...I have a Power
Macintosh 7100 and a Quadra 700 sharing a monitor + keyboard, with an
A/B box to switch between them.  I look at the switchbox when I care
which one I'm on--even in MacsBug.

In what way is the emulation overly complex and under-documented? 
Have you seen Inside Mac: PowerPC System Software?  It's good
documentation, and it's one of the thinner volumes.

On the topic of GS emulation, everyone should keep in mind an
important distinction:  On a Power Macintosh, the 680x0 is being
emulated, but the hardware is real (the PowerPC chip and the emulated
680x0 processor share the same address space).  On the other hand, a
GS emulator running on non-GS hardware would necessarily emulate both
the 65816 processor *and* the GS hardware, including the memory map. 
Just an observation.

Dave Lyons
Mr Tangent