[mcstas-users] Could Virtual_output/Virtual_input raise efficiency?

[wokaoyan1981]

Hi colleagues,




     Learning to use Virtual_output/Virtual_input, I found something interesting:

I put Virtual_output and PSD after monochromator to end the fisrt split simulation I1:

DEFINE INSTRUMENT I1(…)

<CUT. Source, Channel, guide, Collimator and Monochromator. See my previous mail>

/******************psd_after_mono********************/

COMPONENT psd_after_mono = PSD_monitor(

    nx = 90, ny =90, filename = "psd_after_mono ",

    xwidth =0.2, yheight =0.2)

  AT (0, 0, 0.4) RELATIVE mono_out

/********************Virtual_output*********************/

COMPONENT DataAfterMono = Virtual_output(

    file = "DataAfterMono")

  AT (0, 0, 0.4) RELATIVE mono_out

 

Then run it with ncount=1E8 and Scilab format. The obtained data file “psd_after_mono” suggests I I_err N=[4.02214E+8 2.18807E+6 172701].

The second split simulation I2 only includes a PSD behind:

DEFINE INSTRUMENT I2(…)

COMPONENT SourceAfterMono = Virtual_input(

    file = "DataAfterMono_list.p.x.y.z.vx.vy.vz.t.sx.sy.sz", repeat_count = 1)

  AT (0,0,0) ABSOLUTE

COMPONENT PsdAtEntrance = PSD_monitor(

    filename = "PsdAtEntrance", xwidth = 0.2, yheight = 0.2)

  AT (0, 0, 1.1) RELATIVE SourceAfterMono

 

Then ncount is set as 1E8 corresponding to repeat_count=1.To my surprise, running I2 needs 20 minutes as much time as I1.It means that, in this case, the use of  Virtual_output/Virtual_input would’t help to raise simulation efficiency.

 

I suspect the phenomenon is related to high I and N at SourceAfterMono. But I’m not sure. Would you please give some explanation?

 

Best wishes,

Tom

 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20080806/3ac2eaa9/attachment.html>
-------------- next part --------------
_______________________________________________
mcstas-users mailing list
mcstas-users at mcstas.org
http://mailman.mcstas.org/cgi-bin/mailman/listinfo/mcstas-users