On Sun, May 15, 2005 at 09:43:27AM -0400, Timothy Miller wrote:
>
> Adding chips always costs money. But we (the users) will come up with
> some LGPL processor cores for this thing, for fun.
That answers one of my questions. I was thinking about ways of
implementing a piggyback board as an optional accessory to convert the unit
to fixed-frequency video (later, when there's attention to spare for that).
If we build an SPI port into the HDL, it just needs to be brought out to a
6-pin header. The piggyback board could consist of nothing but jumper
headers and the shift registers to read them in. If the core interrogates
the SPI port and gets back anything other than all-1s, it reads the mode
from the SPI bit string, and disables mode commands from the host.
Mind you, actually writing the HDL for the SPI feature could be left
until very late. That's the beauty of field programmability. All it would
take to preserve this option is to assign the FPGA pins, and include the
site for the header and the holes for the piggyback mounting hardware when
the board is laid out.
What do you think, Tim? Should there be an SPI port just on general
principles?
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
