Alex Kaszynski

Please try: ```py pv_plotter = pv.Plotter(off_screen=True) pv_plotter.add_mesh(mesh, texture=texture, lighting=False) # lighting = True modifies the texture colors. image = pv_plotter.screenshot() ```

Ideally image would work, or if it doesn’t, it returns an intelligible error. We might just consider using screenshot and discouraging image, or at the very least providing better documentation.

I think you're correct @adam-grant-hendry, and we could use [pytest-benchmark](https://pypi.org/project/pytest-benchmark/) for this. Would be nice to have the tests in this repository, though I'm loath to adding any more workflows...

> If I can teach myself how to add workflows to our CI/CD, would you be willing to let me add it? We should consider creating a new repository that...

I would like to benchmark performance for the last few minor releases of critical workflows like: - Adding many actors to a `Plotter`. - Adding a single large dataset to...

Here's an initial stab at benchmarks: https://github.com/pyvista/pyvista-benchmarks

> Just as an idea: I personally like to use https://github.com/airspeed-velocity/asv for benchmarking - it is the benchmark-suite used, e.g., by NumPy and Scipy (an example for SciPy from Pauli...

This is to a large degree due to using a non-hardware accelerated picker, `vtkPicker`. There's a hardware accelerated one that's way better, but we'd have to do a bit of...

> And is there a workaround for this? Not a quick one. The picker just needs to be reworked.

I'm actually of the opinion of keeping each module as independent as possible. While `pymeshfix` is quite useful, it's not applicable for all users and any additional dependencies tend to...