[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [f-cpu] Status quo



Le 2015-03-31 16:13, Cedric BAIL a ÃcritÂ:
Anyway, this was just to point out
that light compression do help a lot, if we can either have an
instruction or some trick with the MMU, they will definitively pay of.

"compression" is a pretty wide field (my algo "3R" is vaguely described
in Salomon's book that has almost 1100 pages, and it's far from exhaustive)
and it's hard to find something they all have in common.

Including in the architecture something that is too specific to an algo
is not a good idea. LZ77 works differently than LZW (GIF),
which has a totally different principle than BWT,
and oh, JPEG uses a different encoding, and MPEG has
how many variants ?

All I can think of is "bit field extraction/insertion"
(though it's not used by arithmetic coding)
and I think it's more or less possible to have it in the
instruction set. I don't remember if we have that instruction
in FC0. It uses 3 reads and one or two writes, so it
might not be implemented on the smaller implementations.

yg

*************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxx with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/