>From: Roger Ivie <IVIE@cc.usu.edu>

>>> Might have to change it, make it
>>>SuperQ or something.
>
>Well, you could start by demultiplexing the address and data busses. Of
>course, this will take more pins, so you might have to trim the address
>range down a bit (512KB with a scatter/gather map might be enough, for
>instance; you'll only be using it for I/O) to save some pins. Then you
>could do a bit of optimization by (say) allowing the next DMA master to
>arbitrate for the bus while the current transfers are going on...

Yeah, and how about we call this new bus "Unibus"  :-)

Seriously, I think that it makes sense to retain a totally stock Q-bus but
NOT as the primary way to attach disks.  Have some kind of hot shot LVD
SCSI controller or something, which gets to do its DMA at memory bus speed,
and use the Q-bus for everything *else* like serial muxes and DIO boards and
weird custom stuff.  Just like what PCs *should* be using ISA for but M$ has
gotten all spiteful and wants to kill it because their PnP crap doesn't work.

However, I know people hate to hear this but, do you really think a tiny
company can really build its own VAX CPUs and have them consistantly outperform
software emulators?  It doesn't seem to be happening in the PDP-11 clone
market, and the VAX is a much bigger ball of wax.  I know purists think
a hardware clone is the only way to go, but such people are notoriously
reluctant to pay more for their computer than they did for their house.
I don't know *anyone* who used their own money to buy a brand new PDP-11
hardware clone just to use for fun, and the TOAD-1 didn't seem to do so
well even with the commercial customers (who could definitely afford it,
$100K or whatever to upgrade a dying $2 mil machine is a pretty sweet deal
for a large corporation).  But in any case I agree that VMS compatibility
is a must, people *definitely* won't pay through the nose for a CPU that
can only run an OS that's available in almost identical form for lots of
other machines that are cheaper/faster/bigger.

John Wilson
D Bit