Closed samreid closed 5 years ago
I added a prototype, the decoding of the FFT isn't quite right yet. Also, this relies on sherpa/lib/js-kiss-fft-a.out-1.0.0.js and sherpa/lib/js-kiss-fft-1.0.0.js being checked in.
This approach can handle 512x512 whereas turbomaze fails after 256x256, it seems faster than turbomaze. However, I don't understand the decoding and have a <canvas>
on the HTML to help figure it out.
After the above commit, I'm seeing this result:
Still some problem with decoding the data, not sure what I'm doing wrong.
I'm going to stash the latest version of DiffractionModel from the branch in case we ever need to go back to it and delete the branch:
Branch deleted, closing.
We have been using turbomaze fft implementation, but it would be good to compare with others to evaluate: speed, results and API. I'll create a branch for testing js-kiss-fft2