[HARDWARE] Just Courious......

taniwha taniwha at taniwha.com
Sat Sep 18 23:05:16 EDT 1999


Ha! - I finally solved my mail problems (forgot which email address I
originally subscribed to the list with). I did a paper design  of
a cracking chip last year - you can find it at:

	http://www.taniwha.com/~paul/rc5.html

along with verilog source. In silicon it will do ~2.5MKeys/sec
200MHz clock in a proto 0.15u library - it's ~50K gates 
from a naive synthesis run - in practice it's mostly memory
and datapath (adders and shifters) and would benefit greatly from
some real datapath design - it could easily end up running at 
500MHz and be 1/2 the size - so you might get 6MKeys/sec
out of it - however I designed it for FPGA implementation (it
has a larger number of pipe stages because flops are cheap in
FPGA clbs) I'd guess you could get it to run at maybe 50MHz
in an FPGA (it takes 3x26 clocks to do a key so that would
give you about 640Kkeys/sec/chip).

I have some experience running open-source hardware projects,
where you build it yourself, and I had had this in mind
when I was investigating this - the big problem with these designs 
is that the FPGAs have way too many pins making them unsolderable
by mear mortals (and hence any kit unbuildable).

My design hooked to a PC parallel port and multiple chips
could be assembled into a tree structure to build arbitrary 
sized cracking engines - check out the article on how to win 
the rc5 contest in 7 days :-)

	Paul Campbell
--
To unsubscribe, send 'unsubscribe hardware' to majordomo at lists.distributed.net



More information about the Hardware mailing list