🚀 go-pugleaf

In article <STORKEL.89Oct4144438@rudist.shell.com> storkel@rudist.shell.com (S. Storkel) writes:
>Why couldn't someone
>have designed things so that there were at most three types of
>interfaces (best bet: SE/30 and Mac Portable sharing the same type
>of interface). Sure it would have been a little bit more complicated,
>but I think the payoffs would have been well worth the effort. Where
>am I missing the boat?

At a guess, I'd say you're assuming a level of processor compatibility
between the 68000 and 68020/30 that doesn't exist.  There are a lot of
extra pins on the PDS for the SE/30 that don't exist on the SE.
They're there because the SE has a 16 bit data path and the SE/30 has
a 32 bit path.  The Portable, like the SE, has a 16 bit path.  So to
be compatible with the SE/30 it would need a bunch of extra logic to
take the 32 bit wide data coming in from the bus and ship it to the
processor in two clock cycles, reversing the process for data going
the other way.  

As to why the SE and the Portable can't use the same bus, there may be
a number of reasons relating to the difference in clock rate, bus
communication with the other slots in the Portable, power or cooling
requirements or some other aspect of the Portable's architecture which
make the simple approach taken on the SE unacceptable here.

Finally, there may have been packaging problems involved in making the
same size and shape space available inside the Portable as in the SE
or SE/30.  Any combination of these might have required the Portable
to use yet another incompatible slot design.



-- 
Hank Shiffman                                     (415) 336-4658
Marketing Technical Specialist
Software Engineering Technologies               ...!sun!shiffman
Sun Microsystems, Inc.                          shiffman@Sun.com