stephenbrooks.orgForumMuon1Q&AA few questions about the program
Username: Password:
Search site:
Subscribe to thread via RSS
Jeff Hammersmark
2003-04-10 22:50:59
1. How long does a typical chunk of work take?
2. Is there any way to reduce resolution but fit the whole thing on the screen?
3. Is the frame rate always 1 ?
4. Is there any log of processed data kept for users to look at ?
Stephen Brooks
2003-04-11 01:07:59
1. It varies.  If the results has to be rechecked, it will take 5x as long.  Also there will be a factor of 10 variation depending on efficiency and the number of particles simulated in the design.  I'm aiming for between 5 minutes and 5 hours for each simulation, meaning that an automatic send will occur about 100x less frequently (i.e. after 100 results), so from once/twice per day down to every week or two on slower PCs.

2. You press V to disable auto-view and then P to pause.  Then you can use the mouse and its buttons to navigate around.  I think if you unpause but don't reactivate auto-view, it will stay showing what it was.

3. The frame-rate will go to (1/2), (1/3) etc. when the number of particles reduces towards the end of a bad simulation, to prevent too much of the CPU time being spent on graphics when the calculation itself is fast.

4. results.dat is the cumulative log of all simulation results processed.  results.txt is the log of those produced since the last send.  There used to be a results viewer program for these files but I'll need to adapt it for the new v4.3 format.

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
Jeff Hammersmark
2003-04-13 12:54:08
One last question (sorry for being such a newbie)......

Are the results from this project actually scientifically applicable (are they going to make a difference in the future building of particle accelerators if good results are found) or is this more of a computer programming experiment?  Basically, do the high muon transfer results matter?
Stephen Brooks
2003-04-13 16:54:34
quote:
Originally posted by Jeff Hammersmark:
Are the results from this project actually scientifically applicable (are they going to make a difference in the future building of particle accelerators if good results are found) or is this more of a computer programming experiment?  Basically, do the high muon transfer results matter?


I can't give you a guarantee that this thing is going to get built (that is up to the various governments whether they decide to fund it), but the idea is that what we are simulating are options for the design of a section of a real life machine.  In fact, the higher the efficiencies get, the more likely the thing is going to get built.

It is also true that we are still in a sort-of warm-up phase, because although our results really are real-world applicable, what I hope to do in the end is to run several qualitively-different design options through this optimiser in parallel so I can compare the potential usefulness of each one.

So in short - yes, the high efficiencies _do_ count, but the actual thing might not get built for another 10+ years.  (It's going to send a beam of neutrinos through the centre of the Earth, you know Big Grin )

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
Jeff Hammersmark
2003-04-13 23:24:27
Excellent......my CPU time shall be yours from here on in.
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.orgTwitter: stephenjbrooksMastodon: strudel charactersjbstrudel charactermstdn.io RSS feed

Site has had 25160563 accesses.