[RC5] OGR search ignoring "as-good-as" candidates?

Cole Kitchen kitchenc at mindspring.com
Fri Nov 15 04:49:13 EST 2002


In the following Faq-O-Matic query, somebody raised a good point about the 
possible importance of "as-good-as" Golomb rulers. Have the "powers that 
be" at d.net taken this into consideration?

 From the OGR Faq-O-Matic:

What is the most optimal ruler found by d.net so far?

The client does not attempt to compute or return the fully "optimality" of 
the rulers it is evaluating. It instead only evaluates rulers to see if 
they are better than the best ruler speculated by mathematicians. As soon 
as a ruler (node) is determined to definitely be no better than best-known 
one, full computation of that ruler's optimality is discontinued and the 
next node is attempted. As such, the client will only transmit a "success" 
when it believes it has found one that is better.

There is certain scientific, and even potentially financial interest into 
finding alternate OGR's, which are just as short (or long) as the known 
ones, but are different. In addition they'd serve as a valuable control 
that the process is working. Has this been thought out, or would returning 
all such rulers create too much traffic?
Additionally, when would the contributors learn if a new OGR was found? 
Would Distributed.net strive to make such a discovery known as soon as 
possible, or would they wait for the end of the run, or something..? More 
precisly, have any been found yet?


--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list