dx_ron wrote: ↑Mon May 01, 2023 12:02 am
It's been cloudy, of course - so I went back and re-processed with the current alpha. In addition to jacking up gamma post-tracking I thought I'd go back to this off-the-wall-sounding suggestion of yours. Set RGB:RBG, cap green 100%, Keep. Open color again to put back to RGB. Interesting.
M42_739x15s_norm_RBG-cap_gamma-boost_1.25+740k.jpg
I can't take credit! I think you invented something new.
So based on your description, it seems you swapped G and B, and thus running green cap turned it into a blue cap.
The purple or magenta cap I am aware of actually uses the Layer module for a full inversion (green and magenta being opposites), after which you'd green cap and then invert back.
Of course the M42 and Running Man region is replete with colors, including I believe blues, purples, and greens, so any kind of global capping could end up altering too much.
dx_ron wrote: ↑Mon May 01, 2023 12:02 am
Upon pixel pre-peeping (before Mike gets to it), I found a weird dark artifact to the lower-right of a lot of stars. Here's a crop of one where the artifact actually makes a donut. This is in the purple-star version. I need to go back and start over to see where it first appears. It's not there in the initial OptiDev.
star.jpg
I wonder if this is the same as what Dietmar pointed out a few days back. And I've seen it before also. Did you have intra iteration + centroid tracking turned on in SVD? That might help alleviate it. Or look for a star sample that wasn't very good.