nickkarpowicz / lightwaveexplorer Goto Github PK

I was making a 2D parameter scan (delay1, delay2) on a sequence, and it crashed under windows, reproducibly

Currently only double precision is used; I would like to make this flexible so that single precision can be used (allowing it to run on Intel Xe-based iGPUs that don't support 64-bit FP), or even quads in the case that very small nonlinear effects are to be studied

It would be useful to be able to apply transmission and phase from a file in the course of a sequence. Should be a 3xN array, giving frequency, transmission, phase. People might prefer wavelengths but it will be more reliable if I don't extrapolate lower frequencies.

Sequence should look like
loadOptic("filenameOpticZero.txt")
loadOptic("filenameOpticOne.txt")
do some stuff
applyOptic(0)
applyOptic(1)

This means supporting quotes in the argument field of functions. Optics should just accumulate sequentially

Hello there,
I've been playing with the beautiful application you wrote and enjoying it lately.
In my research i'm interested in the following task -
Propage a 100 fs pulse through air with external focusing (lens of 500mm) conditions and create a plasma filament.
Is this doable under your simulation?
I haven't found a way to put a lens.

Thanks,
Ivan

should have minimal changes to the data structures, but possibly prevent undefined behavior if the slider is dragged to the active plot

Use FDTD modes to calculate nonlinear reflection:
Structure should be to place the observation plane in front of the crystal and do two propagations, one with crystal and one without, taking their difference.

I am currently working on simulating SHG of low pulse energy, tightly focused light. Due to the presence of group velocity mismatch and spatial walk off in the system, the generated SH should exhibit signficant pulse front tilt and a distorted SH beam profile. I am trying to use python to plot the spatial profile of the output of a simulation to characterise this (e.g. a heatmap of the electric field over the x-y axes). However I am having trouble deciphering the way the electric fields are stored in a python lightwaveExplorer object. Do you have any sample code that I could use to get started, or just some tips on the way the electric field is stored in the lightwaveExplorer object?

Any help would be greatly appreciated!

could you provide a version for CW, I do not know how to fix it to CW SHG.

I could not reproduce it on any of my systems, but it was the case on Max S's computer at MPQ - antivirus issue?

The sequence helper icons use emoji but the Mac version uses letters, because of a current limitation of Cairo on mac. This will hopefully get fixed in a future release, or if there's a workaround I would implement it

A simple nonlinear fiber mode would be useful.
Elements to add:

• Apply boundary conditions in real space
• For gradient pressure, recalculate propagation parameters when necessary
• Add some gases to database

Dear Nick,

I tried to compile the code on an Apple Silicon Mac and had quite a hard time. It turns out that vcpkg relies on the distutils library, which is now outdated and no longer part of Python since version 3.12. One has to use a virtual environment with Python 3.11 or earlier, and it works.

It might be worth mentioning this in your documentation.

Thanks for your great software; I love it.

Cheers,
Eliott

Currently, the simulation only handles propagation through a simple planar crystal; would be good to have the option of a spatial map of materials
Needs:

• File format for grid (ascii matrix of integers would be best)
• Memory efficient way of storing the material indicies and only allocating an oscillator set when required

Selecting EOS in the pulldown menu for an input file time will simply use the parameters from the interface since no function exists yet to load it

it does work in normal mode, counting parameter must be getting lost somewhere

In the case where reflections are a significant part of the dynamics (thin samples, nonlinear reflection from a surface) we will sometimes want to solve Maxwell's equations rather than the unidirectional propagation equation. This solution could be integrated within a sequence as a function for calling the Maxwell's equation solver and returning a new space vs. time grid from a given plane within the Maxwell grid.

First version will assume planar sample in vacuum for simplicity. Basic structure:
input field (taken from current value in solution) injected from lefthand side of maxwell grid (3D or 2D space only). Material pixels have running calculation of polarization/current which will only work for Lorentzian/Gaussian materials.

Inputs of function:
material
z extent of grid
start z of material on grid
end z of material on grid
z location of output plane
integer divisor of time step
integer divisor of z step