stephenbrooks.orgForumMuon1Generalv3.2/v4.2
Username: Password:
Search site:
Subscribe to thread via RSS
Stephen Brooks
2002-06-25 15:20:30
I was just thinking we have a lot of computing power here and was trying to imagine how it would all work if I allowed you to vary ALL of the solenoid radii and fields independantly (within engineering tolerances of course).

Things I plan to do by version 4.2 are to decrease the randomness of the results by making the pion decay be simulated more accurately and also include a intermediate-save feature so that simulations-in-progress are not lost.  I'm wondering whether it would be worth also completely generalising the genome at this point.

Alternatively I could test out the more-general optimisation on the decay-channel-only _first_, while doing the less-risky enhancements on the version4/bending chicane, so that I can compare it directly with the results of the last decay-channel-only run (v3.11).  This would mean the release of a version 3.2 and potentially two optimisation projects running in parallel, which fortunately is managable because the programs already upload to different directories for each version.

This is about a month or so away.  Sooner than that I plan to release 4.12 with a bugfix to the genetic algorithm (no more identical parents) and another optimisation elsewhere.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Stephen Brooks
2002-07-08 12:04:48
OK well version 4.12 is released now.  Does it work OK for everyone?


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Pascal
2002-07-08 12:18:17
Everthings ok, all systems working well.  When do you think about adding a new coloum to the stats tab?

___________________________
1: Athlon TB-C, 1.2 GC/s, 256 MB DDR-RAM, Erazor x², ADSL-Flatrate, NIC Intel, Win 98 SE Mainboard MSI-6380 Rev.  1
2: Pentium III, 600 MC/s, 256 MB RAM, NIC Intel, Win 98 SE

[This message was edited by Pascal [www.rechenkraft.net] on 2002-Jul-08 at 19:30.]

[This message was edited by Pascal [www.rechenkraft.net] on 2002-Jul-08 at 19:30.]
Stephen Brooks
2002-07-09 05:43:40
New columns appear on the stats table when the results from the different simulations have to be kept separate for some reason or other.  v4.00 v4.0x and v4.1x calculate scientifically-different simulations (it so happens that what the v4.0 branches were simulating was not what I wanted to simulate!), but as with the version-3 project I feel that I should still credit these people with the number of units they've crunched on incorrect versions.  The rightmost-column only displays the best muon transfer from the _correct_ version of the simulation.
v4.2 will supercede v4.1x, being more accurate, so those figures will be reset to 0.000% again and only v4.2x results used for them.  v4.2 will also go in another column by itself.  If the table gets too wide for people, I could always merge the first two columns.  Or just make the font REALLY WEENY so that nobody can read it.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
kruemi [Swissteam.NET]
2002-07-09 06:38:36
4.12 died silently after about 12hours last night.
I' restarted it and will have an eye on it.
It's the background-version on WinNT.

bye

kruemi (who is still wonderin, where his results have gone)

--
Don't read this.  And neither this!
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.orgTwitter: stephenjbrooksMastodon: strudel charactersjbstrudel charactermstdn.io RSS feed

Site has had 25162429 accesses.