This project has moved. For the latest updates, please go here.

Bug - NURBS Forward Solver

Aug 31, 2011 at 10:59 PM

There appears to be a bug in the NURBS forward solver when you try to make repeat calculations.  

To repeat:  solve the attached problem in the frequency domain, and keep plotting the reflectance for the same parameters:

Fwd Solver Engine: Scaled Monte Carlo - NURBS

Solution Domain: frequency-domain, R(rho,ft)

Independent Axis: rho at ft= 0.1GHz

Detector Positions: 0 to 5 mm, Number 51

Optical Properties: ua=0.16, us=1.7, n=1.4

Output: R

Keep plotting reflectance (set y axis to Log10 for easier viewing), and the solution is qualitatively the same 5 times (although if you look at the data, it's slightly different).  The sixth time the solution is completely different.  Every subsequent solution is wrong, until you reload VTS.

I will try to upload a screen shot and the exported data, but I'm not sure how to do it from here. 


Sep 1, 2011 at 5:23 PM
This discussion has been copied to a work item. Click here to go to the work item and continue the discussion.
Sep 1, 2011 at 5:24 PM

Thank you for your feedback, I have created an issue from your post.

Sep 1, 2011 at 6:08 PM

Thanks Tom! Looks like there may be some static state retained between transformations in the discrete fourier transform of ROfRhoAndT. For our records, here is the screencapture of the 6th (yellow) data set going awry.