I'm running the system with an AMR dataset, and I'm getting some results that confuse me somewhat. I have a suspicion why they are doing what they are doing, but I cannot yet verify this.
To set the stage, I've loaded up a "fake" AMR dataset using yt.testing.fake_amr_ds(geometry = "spherical") and I'm rendering the field "ones".
With a random normal, [0.18617054, 0.15260178, 0.83324573], I get an image that looks like the attached. The min/max are 0 and 3.971; this is just under what I'd expect the max to be (4.0, for through-the-center).
The AMR grids are well-defined, and the total enclosed volume is correct. It looks to me, however, like it's not fully traversing some of the grids. I am not sure if this is because they need to be entered by rays twice (which is definitely possible), but I am working on figuring that out. So I will get back to you. I just wanted to give an update.
I'm starting to suspect that I'll need to use the walkSphericalVolume function to compute the order of traversal for the grids.
I'm running the system with an AMR dataset, and I'm getting some results that confuse me somewhat. I have a suspicion why they are doing what they are doing, but I cannot yet verify this.
To set the stage, I've loaded up a "fake" AMR dataset using
yt.testing.fake_amr_ds(geometry = "spherical")and I'm rendering the field"ones".With a random normal,
[0.18617054, 0.15260178, 0.83324573], I get an image that looks like the attached. The min/max are 0 and 3.971; this is just under what I'd expect the max to be (4.0, for through-the-center).The AMR grids are well-defined, and the total enclosed volume is correct. It looks to me, however, like it's not fully traversing some of the grids. I am not sure if this is because they need to be entered by rays twice (which is definitely possible), but I am working on figuring that out. So I will get back to you. I just wanted to give an update.
I'm starting to suspect that I'll need to use the
walkSphericalVolumefunction to compute the order of traversal for the grids.