15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes
7th December 2023: added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.
Hi,
I read the post that Ha-Oiii Extract Oiii was fixed/improved in Beta 6, so I downloaded it and loaded up a dual band OSC capture that I had stacked previously with the "Ha-Oiii color" mode. I tried Ha-Oiii extact Oiii and all frames resulted "Star Analysis Failed". I tried four options and two options out of four failed:
Algorithm:
Adaptive Airy Disk: Success
Ha-Oiii color: Success
Ha-Oiii extract Ha: Star Analysis Failed
Ha-Oiii extract Oiii: Star Analysis Failed
I had the same files loaded in both cases (A subset of my lights and bias frames to speed the test, about 20 lights and 30 bias frames). AFYI: These lights only have about 120 stars, so not a ton.
Also, I had not played much with the Ha-Oiii Extract Oiii or Ha options on this data set before, so it's possible the earlier one would have failed too, but given that Oiii is supposed to work now, I thought I would give it a try. But I'm just surprised that star analysis would succeed with “Ha-Oiii color” and fail with “Ha-Oiii extract Oiii”… seems star analysis should be the same in either case, I would have thought you would be looking for stars for registering and alignment purposes, not optimizing color extraction.
- Steven
Hi,
Can you try this : do star analysis with Ha-Oiii color, then registration.
After that, choose Ha-OIII extract Ha or OIII, and start at the normalization step.
David
That worked! I should have thought of that workaround, clearly it has an issue with analyzing stars in these other modes. Don’t know why the particular color mode would break star analysis.
I have got this error in beta6 when doing Extract Ha
That worked! I should have thought of that workaround, clearly it has an issue with analyzing stars in these other modes. Don’t know why the particular color mode would break star analysis.
We are looking at the star detection algoritm as someone else saw a big difference pre-beta, we're investigating.
I haven't given any specifics about this error because I have seen it several times and I can't replicate it at will. I'm hoping that you can figure it out by looking at the message. I have seen it about 6 times now and I just quit APP and restart and it seems to work the next time.
I was doing Ha-OIII extract Ha. The problem may be occurring when I do an ordinary integration then go back and do Ha-OIII extract Ha without restarting APP. I don't know what you mean by "what kind of data." It's UVIR and L-Extreme filter data.
If you can't spot the problem as it is then I'll pay attention next time so see if it repeatable.
@ddnum The error is a generic one unfortunately, might be a bug which pops up once in a while. If you have the exact data this happened on and are able to reproduce it a bit, we could use that data to hopefully find it.
I’ll see if a small manageable subset of my data triggers it.
Sorry for the delay.
OK, I think I found the general cause: I stored one data set as TIFF from NINA instead of FITS. As noted in another thread, TIFF headers are not as robustly used or defined. I'm wondering if this is it. I've captured several targets using FITS and have had not issues.
If you are still interested, here are examples of the TIFFs (two lights and one bias).
Yes there may be something going on with the TIFF's in that sense. You should be able to process different formats, maybe when you first do the tiff and then the fits, then combine those?
Yes there may be something going on with the TIFF's in that sense. You should be able to process different formats, maybe when you first do the tiff and then the fits, then combine those?
Well, I shoot all FITS now, which is a bit too bad as lossless compressed TIFFS are half the size. Nice on the drive space, I wish FITS supported lossless compression. For the stuff I have as TIFF, I worked around the issue per the suggestion on the person that responded. Thanks!
