stephenbrooks.orgForumMuon1GeneralHow do we get higher yields?
Username: Password:
Search site:
Subscribe to thread via RSS
Ripper
2004-03-09 11:49:36
I have noticed that some people are constantly getting higher and higher yields.  I am running my system using the top 500 of all my computers and yet the highest yield hasn't changed in over a month.

I have also noticed that there are a LOT of #gen entries that result in less than 15Mpts.  I thknk I asked this before, but either never got a response OR something else strange happened, as I can't find it using search.  How do I turn this off and what is the trick to getting higher yields?  Please don't say there isn't one as the stats page shows that there is obviously something that can be done to allow people and groups of people to constantly provide higher yields. 

When groups of people can pass those of working this project for some time with less than half the number of units being reported but higher yields (Mpts), then there must be a manual way to tweak the system OR the results are fictional.  If the software is designed to constantly provide random inputs that will gradually increase the yield based on results from other tests, then I would expect my system to be generating some higher numbers, not staying at the same level or worse, providing lower yields.

Any input???

Thanks

Ripper
=============================
Solution Submission:

SELECT * FROM people WHERE clue > 0

=============================
Nexus
2004-03-10 02:44:21
Yes, there are ways of manually tweaking the results.  Basically every result contains a great many variables.  If you looked in your results file and saw that an increase in muon yield corresponds to an increase in one of the variables you might try cranking that variable up to max and feed that citerion to the client to see how it goes.  I'm not quite sure how you would go about doing that though.  Never tried; never really felt the need.

As for the lower yields, there is a good reason for that.  Firstly, Stephen Brooks is interested in how the combinations of variables determines muon yields.  That requires some results in the low end of the spectrum as well as the high.  Secondly, by only focussing on getting the last dregs of performance out of one setup we would be ignoring the possibility that there is another totally different design that works better.  By periodically throwing in random configs the project reduces the risk of overlooking something good.
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.orgTwitter: stephenjbrooksMastodon: strudel charactersjbstrudel charactermstdn.io RSS feed

Site has had 25160947 accesses.