Hi Greg,
Stepping through this:
- There’s only one instance of
radand no more than eight instances ofrpictrunning. - There is no other error message to indicate that an
rpictprocess died. If one is dying, it is either not showing an error message, or the message is not going to stderr. - The image is 512x512, so it shouldn’t be anywhere close to the file size limit. The unf file (what does get produced) is 4064 kb, and the hdr file produced successfully from the single thread version ends up being 625 kb.
- I can use the ambient file for subsequent regular
rpictruns, so it doesn’t seem to be corrupt.
Also looking back at Randolph’s thread, this model also contains lights, not mkillum necessarily, but ies2rad. I’m not sure if that’s a factor, but maybe I can try out some other models.
Nathaniel