RE: PCI Politics (was RE: why Target cannot change its mind)

["Austin Franklin" <austin@darkroom.com>]

> >I have every idea how to get the signaling of PCI at 33MHz.  It
> is not the
> >PCI signaling that makes it difficult to be done outside of an
> ASIC.  It is
> >some of the cycle to cycle timing.  Initially designing PCI
> without certain
> >cycle to cycle timing constraints (basically too much logic
> between cycles)
> >would have made a negligible impact on performance.
>
> Are you talking about the parity checking, or something else?  What would
> you change, and how would it strike a better balance between the utility
> of PCI and the ability of letting "the little guy" in the door?

TRDY/IRDY timing.  Registering so there is one registered cycle in, and one
registered cycle out.

> > > The days of just slapping
> > > some CMOS gates on a connector and whomping up a new
> > > peripheral board are long gone, my friend.  Where have you
> > > been?
> >
> >Where have I been?  Designing (over 50) PCI interfaces since its
> inception.
>
> It sounds like you've been around a while, then.  I suspect you
> miss the days when a little company like Hercules could make
> a fortune on the graphics board business or AST & QuadRAM
> do their thing with memory cards and the like.  All with a
> four-layer fab (often with rework on it) and a handful of SSI.
>
> I can see your point that the investment bar is higher for
> a PCI design than an ISA/EISA/VLbus/whatever design because
> of the ASIC requirement.  Where I would differ is that the
> bar was not above the heads of a small company.  It was above
> the heads of a couple of guys horsing around with circuits
> in the garage.

I take it you've never priced an ASIC out.  Do the arithmetic.  Synopsys
$100k/Seat, VCS $30k/Seat, Design Engineer $100k/Engineer, Verification
Engineer $100k/Engineer, Fab ~$250k....plus spins...plus minimum orders.

One company I have been working with has burned well over $10M making two
ASICS, one is for PCI...and I'd say they have 5 Engineers on each ASIC...run
the numbers.

That is FAR more than two guys horsing around with circuits in the garage,
and well beyond the means of most small companies.