WBP 2.0 issues

I haven't watched them again......will watch them and look for those suggestions that I can try. I wasn't having issues when you started releasing them ...and with the exception of the groupings, everything else seem pretty straight forward and similar to the first release of the new WBBP in Oct. A few options were removed as in Calibrate Only, and Flat Darks check box. But the Oct release worked flawlessly .....it is this last release that has failed with four different data sets taken over one night using two rigs. I have used masters and have processed with flats, flat dark, and dark files. Although I haven't used Bias frames with the ZWO cameras in some time...I even tried that in lieu of the Flat Darks. I was able to manually calibrate, register, and integrate the images using the Bias master successfully. I have tried default settings and tried using Warren Kellers suggestions. I have used cosmetic correction and tried without. I have zero problems with all data sets using Deep Sky Stacker and the results look good for me. But......I will re-watch your videos looking for the help I obviously need. I do not mind putting all of the images, lights, flats, flat darks, and darks in a zipped file for DROPBOX. It will take a long time to upload and download as the file will be enormous. But, I have the time is someone wants to take a crack at it. Hopefully I will have takers who will figure out what I am doing wrong.
 
I was able to image last night with both rigs....each rig imaged two different targets. I have processed two sets of images from the SCT kit and both sets failed ....just more of the same.....STAR ALIGNMENT issues. I assume the images from the widefield refractor (other kit) will also fail. I have a question. If I download and install the last version of Pixlnsight..the one that was released in October (that worked flawlessly) will I get the complete older version. I am trying to avoid using Deep Sky Stacker! I have attached both sets of logs.
 

Attachments

  • Logs.zip
    93.4 KB · Views: 45
EOVER

This gets a little complicated but something I consider very important is just because you get files to flow through a process (with WBPP in the past or DSS) it *does not mean* the results were optimal or correct. They may or may not be. One of the interesting things about the new version of WBPP is that it communicates exactly what is going on with respect to image calibration.

So, if the problem you are having is specific to matching data for purposes of proper calibration- that is easy to check.
The results literally cannot be different if you configured things in the same way in both the "old" version of WBPP
and the new version. What is best, I think, is to move forward with getting things configured correctly in the new
version...and then you can retroactively determine the difference to what you were doing in the past.

I did download and look at your data. Roberto is correct ... but I will say it more strongly your 300sec dark frame *does not match and will not calibrate* your H-alpha data. No program can make those two files work together. I posted a screen shot of subtracting the two files. All of the stuff that follows (like StarAlignment) will be tripped up with crazy data due to consequences of one sort or another. E.g. A reference file was not assigned since it failed to find any signal there... who knows...stuff like that will cause the error that started the process. It is better to just do Calibrate only without aligning and combining...and check the data.

I created a video that demonstrates the DARK level changes with your exact camera on short timescales. Taking new darks was the solution. There is no guarantee that new darks will match your old data unfortunately. But new darks with new data will. Even if you did "get it to work" I suspect you will have other problems (with the flats working in particular). You can see this all in action with this video:
 

Attachments

  • Capture.JPG
    Capture.JPG
    287.4 KB · Views: 39
I was able to image last night with both rigs....each rig imaged two different targets. I have processed two sets of images from the SCT kit and both sets failed ....just more of the same.....STAR ALIGNMENT issues. I assume the images from the widefield refractor (other kit) will also fail. I have a question. If I download and install the last version of Pixlnsight..the one that was released in October (that worked flawlessly) will I get the complete older version. I am trying to avoid using Deep Sky Stacker! I have attached both sets of logs.

I strongly suggest you need to stop using Weighting, StarAlignment and Image Integration. You just need to Calibrate the data only..and look at it. StarDetector is basically saying you have black images or black-clipped images. Do you? You can check that now by looking at your *_c images.

It is all here in the log:
* Begin generation of image descriptors
[2021-02-23 19:57:21] ************************************************************
[2021-02-23 19:57:21] Computing descriptors for image 1 of 20
[2021-02-23 19:57:21] /Volumes/Astro II/2021/2_February 2021/2-22-21_NGC2359_Duck Head Edge/1-Pixsight/calibrated/light/NGC2359_Duck Head Edge_300sec_1x1_Ha_0001_c.xisf
[2021-02-23 19:57:21] Loading image: w=4656 h=3520 n=1 Gray Float32
[2021-02-23 19:57:21] 74 FITS keyword(s) extracted.
[2021-02-23 19:57:21]
[2021-02-23 19:57:21] StarDetector: Processing view: Image01
[2021-02-23 19:57:23] 300 star(s) found
[2021-02-23 19:57:23] 0 PSF fittings
[2021-02-23 19:57:29]
[2021-02-23 19:57:29] ** Warning: No convergence in MRS noise evaluation routine: using k-sigma noise estimate
[2021-02-23 19:57:30] ------------------------
[2021-02-23 19:57:30] FWHM : -
[2021-02-23 19:57:30] Eccentricity : -
[2021-02-23 19:57:30] SNR : 0.000
[2021-02-23 19:57:30] ------------------------
[2021-02-23 19:57:30] Failed to measure /Volumes/Astro II/2021/2_February 2021/2-22-21_NGC2359_Duck Head Edge/1-Pixsight/calibrated/light/NGC2359_Duck Head Edge_300sec_1x1_Ha_0001_c.xisf
[2021-02-23 19:57:30] Image will be ignored.
[2021-02-23 19:57:30] Computing descriptors for image 2 of 20
[2021-02-23 19:57:30] /Volumes/Astro II/2021/2_February 2021/2-22-21_NGC2359_Duck Head Edge/1-Pixsight/calibrated/light/NGC2359_Duck Head Edge_300sec_1x1_Ha_0002_c.xisf
[2021-02-23 19:57:30] Loading image: w=4656 h=3520 n=1 Gray Float32
[2021-02-23 19:57:30] 74 FITS keyword(s) extracted.
[2021-02-23 19:57:30]
[2021-02-23 19:57:30] StarDetector: Processing view: Image01
[2021-02-23 19:57:32] 238 star(s) found
[2021-02-23 19:57:32] 0 PSF fittings
[2021-02-23 19:57:38]
[2021-02-23 19:57:38] ** Warning: No convergence in MRS noise evaluation routine: using k-sigma noise estimate
[2021-02-23 19:57:39] ------------------------
 
First and foremost - thanks to all for hanging in there with me. I ran Calibration only on the Ha images. They were black with stars. I ran a histogram transformation (stretched) and there was no change. Clearly the darks were clipped. If I "Blink" the ha images they look good. So I am lost - is it my supporting files that are causing this?? I plan reshoot flats, flat darsk, and darks (new library) and perhaps I will find out......but I am puzzled if those are the culprits how was I able to use them without issues prior to this most recent release. I am just lost in trying to understand what is going on.
 
Ok....where do I begin! First, if there are others who are passionate about this hobby and want to understand what is under the hood - I cannot compliment Adam Block's Fast Track and Fundamentals course enough. If you just want to know how to drive the car, there are youtube videos on PixInsight training that can do that for you....but if you get stuck and are not pleased with what is going on......and you need to look under the hood, These courses are designed for EVEN ME to understand. BTW, my problems are solved - I have re-shot my calibration frames and know now I am much better off using current calibration frames rather than 1 month old library saved ones. Again, I am not in the habit of recommending businesses that I personally do not know but these course at www.adamblockstudios.com are priceless. From my small part of the planet, the cost for knowing how to process my data is small in comparison on what I have spent on equipment to acquire the data. Thanks to everyone for your help - I apologize for taking up this thread for issues of my own making and thankful beyond words for the work down by Mr. Block. For the record, I have added the Horizons training as well. Who knows, one day I may be on this forum helping others -who would have thunk!
 
Back
Top