Josh Borrow
Josh Borrow
There are many things going on here! First, I agree that if one wishes to prove that a numerical scheme is giving the incorrect answer, a reproducible test is always...
Thanks for that snippet; note that AREPO is not a smoothed particle hydrodynamics code. I am not sure how it is relevant to the discussion. If we wish to move...
> > Thanks for that snippet; note that AREPO is not a smoothed particle hydrodynamics code. I am not sure how it is relevant to the discussion. If we wish...
Great, thanks for checking that @nastasha-w. It seems the issue here may only then relate to frontends for codes that do not use this, including PHANTOM (SPLASH's primary parent code)?
Yes, we use pre-rendered kernels for mid-size (~1-10) pixel overlaps for accuracy. In all other cases we use an extremely simple numerical integration. The reasoning for not simply using the...
I've taken a brief look at the code and I can't immediately see if there is something wrong. However, if you are using the subsampling strategy that we demonstrated in...
Fix PydanticDeprecatedSince211: Accessing the 'model_fields' attribute on the instance is deprecated
Fix looks good and will reduce noise in our application and CI.
The only thing that broke things for us was the deprecation of `input_units` being turned into an error (from where we subclass unyt_array). Other than that we seem to be...
I want this feature and the code looks good to me. But the documentation is currently lacking (both on the developer side and on the user's guide side). I would...
The thing here that worries me is that 1e13 msun in those units is ~677, far, far, far from the precision limit in the array!