Re: Apple IIe TCP-IP w/ LANceGS Card

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

Marsha wrote:

>I haven't looked at the binary of the workstation ROM to figure out what 
>executes on card vs. what executes on the motherboard, but in terms of low 
>level coding it would have been simpler to tell the programmers to program 
>for a 65C02 for all the code rather than some 6502 only and some 65C02...

I don't think there would be much simplification, since one set of software
is "firmware" and the other is driver software.

Except for religious reasons, they might just as well used a Z-80 for the
card.

>Speed and code size being somewhat critical, taking advantage of the 65C02s 
>extra opcodes, like BRA, come in handy, over saving register flags, setting 
>carrys, branching, restoring register flags and continuing with the code.

In my experience, it is seldom necessary to do much extra, if you think
about the problem from a 6502 point of view.  For example, when I need
an unconditional branch, the "worst case" is a BCS+BCC if they need to
be relative, and I almost always know the state of a flag that I can use
for a simple "branch always".  Having to move data through the A register
to get into and out of the stack is usually just something to get used to,
though it can occasionally get sticky.  STZ is cute, but hardly necessary.

In short, I find the "universal" 65C02 extensions rather trivial, and, while
they can save a few bytes in a difficult situation, they don't measurably
improve my life as an assembly language programmer.

The advantage of having code that can run on any Apple II outweighs
the smaller advantage of saving a few bytes, in my opinion.

-michael

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