Opened 15 years ago

Closed 13 years ago

#22 closed feature (fixed)

experimental planning tools

Reported by: butler Owned by: kline
Priority: minor Milestone: SANS Reduction 5.2 Release
Component: SASCalc Keywords:
Cc: Blocking:
Task:

Description

One of the things that DANSE eventually plans to provide is experimental planing tools, hopefully to include full MC simulation of the instrument and samples. However, it would seem that it would not be too difficult to take the output of SASCAL and our analytical models to see what one would expect on an absolute scale with error bars which depend on the run time desired. This would clearly be a long term idea as I suspect there are many other priorities at this point, but Steve suggested I put this in as a ticket so it doesn't get lost.

Change History (6)

comment:1 Changed 15 years ago by srkline

  • Milestone set to SASCALC_2.0_Beta

comment:2 Changed 14 years ago by srkline

from (closed) ticket #53:

currently, only a simple Debye function is shown. It would be instructive to show different models (sphere, peak, vesicle, cylinder, etc) on the plot.

  • show the smeared data set that would be measured
  • show the unsmeared data set for the same data points
  • allow some simple input of the sample parameters

comment:3 Changed 14 years ago by srkline

from (closed) ticket #16:
For even more help in planning experiments, data could be simulated for one of a set of scattering functions to give a more realistic view of what the expected result for the experiment would be.

Ideally, one could set SASCALC to a given counfiguration, pick a "sample" (size shape contrast) and count for X seconds - and get Y detector counts. Then either 2d data could be shown, or averaged to 1d with proper statistical erorr bars.

How this is actually implemented could be in several steps.

This is an idea that the PAC has asked for in the past as a tool for all users to assess feasibility and improve experimental design. It would be a valuable teacing tool as well.

comment:4 Changed 14 years ago by srkline

First pass at the MonteCarlo? has been added in r429 and r430. This appears to work very well for single scattering simulation. Multiple scattering is yet to be verified. There are still some open questions that are stated at the top of the ipf file.

Any model (almost - see comments in the ipf) can be plotted as the SASCALC model. The analysis package must be simultaneously loaded, and an unsmeared model plotted. Then the unsmeared model coefficients and model are used for either "normal" SASCALC or for the MC simulation.

An XOP (Mac only) version is working, but not under SVN until the MonteCarlo? has been verified.

Interface/functionality/feedback/etc. are still crude.

comment:5 Changed 13 years ago by srkline

  • Milestone changed from SANS Reduction Wish List to SANS Reduction 5.2 Release

comment:6 Changed 13 years ago by srkline

  • Resolution set to fixed
  • Status changed from new to closed

done.

XOP to SVN and Windows port has been opened in a new ticket

Note: See TracTickets for help on using tickets.