[Solved] APP hanging after reaching Dynamic Distortion Correction Stage when constructing a mosaic using Seestar data

It may be APP is not hanging. It could be just taking an exponentially long long time to do a mosaic from that many individual frames, especially with DCC enabled.

Hi Narayan @seeker372011,

Thank you very much for reporting this issue and your clear explanation of it.

I think I understand what is happening. Let me explain. The current way for APP to calculate a mosaic is to try to register all those mosaic frames to each other to learn how they are connected. This process will take some time, but I see that you already got to the 80% point where indeed all the earlier information is used to build up the actual mosaic. Maybe the calculation becomes very big and it will ask a lot from your computer. How many cpu threads and RAM do you have and use?

Perhaps you want to upload all the 350 frames so I can have a try and find a solution to make it happen?

Can you upload it here:

https://upload.astropixelprocessor.com/

username: uploadData

password: uploadTestData

Can you make a folder with your name and issue like: Narayan-Seesstar-mosaic-issue

and upload the data there? Thanks !

Mabula

Hi Mabula

just uploading 352 subs from a Seestar mosaic

my laptop has 32 G of RAN and APP is using 31 threads

Thank you!

Narayan 

Hi, Former Seestar Owner, when stacking the Seestar Fits (especially when using the mosaic mode in the Seestar), under #4 Registration, don't use "Mosaic" in the Registration Mode, use "Normal."  Then under #6 Integrate under Composition Mode, Use "Full" not "Reference."  Alternatively, you can use "Reference" but you will need to manually add the Seestar Stacked Mosaic Fit File and manually select that file as a reference frame after Analyse Stars runs.  I posted response a couple months ago that has a more detailed set of instructions ( https://www.astropixelprocessor.com/community/main-forum/seestar-s50-new-mosaic-framing-mode-stacking-registration-takes-very-long-time/#post-31583i

Posted by: @seeker372011

↑

Hi Mabula

 

just uploading 352 subs from a Seestar mosaic

 

my laptop has 32 G of RAN and APP is using 31 threads

 

Thank you!

Narayan 

Hi Narayan @seeker372011,

Thank you very much for the upload. We have received it in good order 😉 I will test it in the coming days.

Posted by: @brannonqgmail-com

↑

Hi, Former Seestar Owner, when stacking the Seestar Fits (especially when using the mosaic mode in the Seestar), under #4 Registration, don't use "Mosaic" in the Registration Mode, use "Normal."  Then under #6 Integrate under Composition Mode, Use "Full" not "Reference."  Alternatively, you can use "Reference" but you will need to manually add the Seestar Stacked Mosaic Fit File and manually select that file as a reference frame after Analyse Stars runs.  I posted response a couple months ago that has a more detailed set of instructions ( https://www.astropixelprocessor.com/community/main-forum/seestar-s50-new-mosaic-framing-mode-stacking-registration-takes-very-long-time/#post-31583i

 

Thank you very much Brannon @brannonqgmail-com for your help and good suggestion. I think the registration mode normal should always suffice unless 1 or more frames don't have any overlap with the reference frame. I guess with the SeeStar mosaic mode, you would stlll have everything overlapping with the reference frame, right?

@mabula-admin The issue with Mosaic mode is the Seestar will take photos in a spiral pattern from the center of the frame (see below).  When you throw all the subs into stacking software, any reference frame will limit what gets stacked (unless you can choose 'full' instead of 'reference').  APP is by far the best at doing this but if you choose "Mosaic" in Registration it will take forever, I am guessing because there is so much overlap in the subs.  Also, since the Seestar is mainly 10 sec subs, there are thousands being stacked at a time.  At the end of the Seestar session, the unit will auto stack the mosaic image and it will be a separate fits file.  If you add that file to the lights and manually select that as the reference frame, APP will overlap and align all those subs to that image and produces great results.   

Hi Brannon @brannonqgmail-com and Narayan @seeker372011,

Thank you very much Brannon for your explanation. I fully understand it now. Indeed loading the mosaic image from the seestar and setting it as a reference should solve all issues and it should complete without issue using the normal registration mode. When integrating, I would suggest to even disable the reference from stacking, by deselecting that mosaic frame in the list after step 5) normalize, because the mosaic made by the seestar is already including all data and might be of less quality than the APP produced result. I will try to test it on Narayan's data.

Narayan, did you try to integrate your data with this tip from Brannon?

Mabula

Thanks Brannon and Mabula. Unfortunately due to some home renovations I have had to pack away my laptop for a couple of weeks and haven’t had a chance to try out your suggestion 

I’ll do that as soon as I am able to- maybe end of next week and for the sake of completeness report back here .

Thank you for taking the time to respond

clesr skies( haven’t seen any myself since December!)

Hi Narayan, @seeker372011

Have you tried yet?

Using the new APP 2.0.0-beta32 which you can download from the Downloads page, I was able to process your mosaic with issues using the mosaic registration mode.

In 3) Analyse stars, I set the star count target to 100 and the Cosmic Ray/Noise reducer to 16 and all went fine. This is very helpful because the seestar data with short exposures is a bit noisy. The registration of the whole mosaic for the 350 frames did not take that long even because of the low star count.

Mabula

@mabula-admin that’s fantastic news! Thank you so very much 

unfortunately my laptop is still packed away as my home renovations are still underway so I haven’t had a chance to try but sounds like the solution has been found

many thanks

Narayan

Posted by: @mabula-admin

↑

Hi Narayan, @seeker372011

Have you tried yet?

Using the new APP 2.0.0-beta32 which you can download from the Downloads page, I was able to process your mosaic with issues using the mosaic registration mode.

In 3) Analyse stars, I set the star count target to 100 and the Cosmic Ray/Noise reducer to 16 and all went fine. This is very helpful because the seestar data with short exposures is a bit noisy. The registration of the whole mosaic for the 350 frames did not take that long even because of the low star count.

Mabula

 

 

Hi Mabula

finally got a chance to try ..downloaded 2.0.0-beta 33 and the mosaic mode worked like a charm. Took a little while but not terribly long.

I much prefer the mosaic mode to using the Seestar stack as a reference as I can adjust the cropping 

thanks for your support !

Narayan