home .. forth .. misc mail list archive ..

Re: P64 Math


Regarding the adder width and carry propagate time:

At the risk of putting my foot squarely in it, has anyone considered
staggering the read/write signals across the word to keep abreast with the
carry?  (i.e., the read/write operation is skewed slightly with the least
significant byte/word being accessed one minor clock cycle before the next
most significant byte/word.)  It seems at first blush that this could
offer the simplicity of ripple carry hardware with little real performance
penalty because of the minor degree of pipelining it could allow.

Doug
-------------------------------v-----------------------------------------------
Doug Girling                   | Voice: (604)291-4428  
School of Engineering Science  | FAX:   (604)291-4951
Simon Fraser University	       | email: girling@cs.sfu.ca
Burnaby, B.C. V5A 1S6 CANADA   | CompuServe: 73777,50 (73777.50@compuserve.com)