[RC5] Is my 386 any use?
Jim C. Nasby
jim at nasby.net
Sat Jan 16 13:18:13 EST 1999
While smaller block sizes would make 386s more useful in DES, they would also mean
a much higher load on our network as well as changes to every part of the network,
from the keymaster down to the clients. (I won't go into details, but every piece
of the network would need to be able to handle the smaller blocks). It's basically
a step in the wrong direction, unless someone comes to us with a huge farm of 386s.
We're not trying to be biased against 386 owners or anything, but we simply have to
draw the line somewhere. You'll notice that 8086s and 286s aren't supported at
all... that's a line that was drawn early on. Granted, there are more 386s than
286s, but it's just not worth the amount of effort involved just to allow slow 386s
(iirc, the faster 386s can do a 28bit block in a few hours, which is still
reasonable at this point) to participate in one contest. They are still more than
welcome to work on rc5, and any other contests where block latency isn't an issue.
d.net Human Interface
"Art Coates Jr." wrote:
> Hence the use of smaller, more manageable block sizes for slower machines. The
> benchmarks on version (.427+) will recommend a block size for the particular
> machine. A nice feature that comes in useful for 386 & 486 type machines.
> Carl Johnstone wrote:
> > >I think you forget the PURPOSE of the d.net experiment, to use
> > >spare cycles of ALL machines, even low end machines to increase
> > >power, rather than only high end machines.
> > The point is that on "quick" contests (like DES) a machine which connects
> > downloads half a dozen blocks spends a couple of days cracking them and then
> > uploads them is probably wasting it's time, because the contest will be
> > over.
> > If a machine gets a block and cracks it and then uploads that one block 24
> > hours later may find that the block has already been re-issued to another
> > machine thats cracked it in 5 mins and returned it already! As these
> > contests get quicker, the usefulness of low-end machines gets less and less.
> > When you get to long-term contests (RC5) then the odds of the contest
> > finishing or a block being re-issued is that much lower that low-end
> > machines are still an important part of the project.
> > Carl
> > --
> > To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
> > rc5-digest subscribers replace rc5 with rc5-digest
> To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
> rc5-digest subscribers replace rc5 with rc5-digest
Jim C. Nasby (aka Decibel!) /^\
jim at nasby.net /___\
Freelance lighting designer and database developer / | \
Member: Triangle Fraternity, Sports Car Club of America /___|___\
Give your computer some brain-candy! http://www.distributed.net Team #1828
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5