Sega Multigame and Vector Generator Card...

Just misc. updates here.

DaveF looked through the Tac/Scan code and found a few locations that
needed some attention. I'm running the new code on real hardware and it
seems to have fixed the problems I saw with the "cracked" ROMs. (Atta-boy
to DaveF, again!) Travis (my "beta-tester" :-) hasn't found anything else
suspicious with the rest of the games so far. I've been building
daughtercards at the rate of about one a night (fully "testing" with a game
or two of Zektor and Tac/Scan... ;-) so I'll probably get some shipped out
to people that have pre-paid in the next week.

--- end sega multigame stuff ---

As for the PC-vector generator card... (Zonn! Input! Need Input!
Hint-hint don't let that job get in the way of the important stuff! ;-)
I've got a VHDL compiler all set up and targeting to the Cypress 371i.
(About $6 in singles.) If the adders get too big we can target to the 372i
(64 macrocells instead of 32) for about $10 a pop. The Lattice rep is
bringing me a copy of Synario with VHDL support, so I'll see how that looks
too...

Trivia... Ripple-adders are kind-of a pig when it comes to programmable
device resources. Each pass (bit) takes a macrocell, so a 12 bit adder
eats ~38% of a 32 macrocell PLD! Still though-- a comparitor, an adder, a
subtractor, and some registers should cover what we need for Bresenham's in
hardware.

( And for just $12 a pop the CY7C373I is in-circuit programmable, comes in
an 84 pin PLCC (with 64 I/O's!) and has 64 macrocells... Could probably
glue that directly to a pair of 12-bit parallel DACs and a databus without
any external parts. Kinda cool. ;-)

All this PLD stuff is still orders of magnitude faster than we need, even
with a BUNCH of passes through the array. (Back of envelope calculation of
12 bit ripple-adders resulted in an 8MHz draw rate. 3-bit look-ahead
adders were several orders of magnitude faster, but eat P-terms much
faster, yet save Macrocells... Gotta love CPLD lingo.)

Al-- did you by chance look up that old hardware-assist Bresenham's dual
68K project you did way back when?

-Clay

Clayton N. Cowgill Engineering Manager
_______________________________________________________________________
/\ Diamond Multimedia Systems, Inc. clay@supra.com
\/ Communications Division http://www.supra.com/
Received on Mon Dec 15 18:21:33 1997