Reusing masters in DSS

[philhilo]

Using DSS is there a way of telling it to reuse use masters it has already made when reprocessing data? Such as the master flat, or master dark, rather than it churning through everything when you redo a set of lights?

[philhilo]

And my second question, why does DSS sometimes throw out perfectly good looking data? Thought I was having a great night and with over 2 hours of the Horsehead and it only wants to use 6 or 7 of the 44 subs, and they look fine, seeing was pretty grim but that's not been a problem before, baffled.

[spillage]

1st answer is that DSS saves the mater to the same folder location as the single calibration frames. So you can just load the master on its own if you are reusing the calibration frames.

2nd answer, Have you checked the number of stars found using the slider bar to make sure enough stars have been detected?

[philhilo]

3 minutes ago, spillage said:

1st answer is that DSS saves the mater to the same folder location as the single calibration frames. So you can just load the master on its own if you are reusing the calibration frames.

2nd answer, Have you checked the number of stars found using the slider bar to make sure enough stars have been detected?

Thanks for that. I wondered about that (been there ages ago and it was a focus issue, but this looks fine - just done M45 from same session and no problems but obviously much higher in the sky out of the murk), with the detection at 20% most were 50-70 stars detected, and some 250. Now I have pushed the star detection threshold down to 10 or 5% and it is picking up 250+ stars, although looking through the files 6 or 7 have 3 or 4 times that amount. It is still rejecting the bulk of the subs. Is it processing those and rejecting the others, do I need to split it into 2 batches? Can I see which subs it is rejecting and which it is processing? Cheers.

[Tommohawk]

9 minutes ago, spillage said:

1st answer is that DSS saves the mater to the same folder location as the single calibration frames. So you can just load the master on its own if you are reusing the calibration frames.

2nd answer, Have you checked the number of stars found using the slider bar to make sure enough stars have been detected?

.. and if you prefer you can put the master in a separate library folder. It will always open masters by default from the last location used.

[Tommohawk]

2 minutes ago, philhilo said:

It is still rejecting the bulk of the subs. Is it processing those and rejecting the others, do I need to split it into 2 batches? Can I see which subs it is rejecting and which it is processing? Cheers.

Just do the registration first without stacking. You can then see the scores each sub gets and decide how many to stack. Or maybe it isnt even registering them?

[spillage]

4 minutes ago, Tommohawk said:

Just do the registration first without stacking.

+1

I am not sure what camera you are using but with a mono I only bother doing calibration frames from the L filter.

Once the master is created I dump the other files and then redo as and when is needed. As long as your filters are kept clean this should be okay but it depends on your set up and situation.

Edited December 23, 2019 by spillage

[philhilo]

Hooray, and apologies for the panic. Looked back through the files and discovered there was a shot of M45 in the set. I shot two targets last night, but not sequentially as I was cloud dodging. When I looked at the files APT had labelled them in such a fashion that they went into the folder interleaved, thus I had to sort them visually off of small tiles (whilst doing breakfast, fielding small children etc). During that process one ended up in the wrong folder and that is what lead to DSS rejecting 90%, rather than rejecting one outlier as you might expect. 

So moral of the story is when shooting multiple targets check the files very closely and DSS does not always respond in an expected way to bad data in.

Thanks again everyone for their help, cheers.

[philhilo]

2 hours ago, spillage said:

+1

I am not sure what camera you are using but with a mono I only bother doing calibration frames from the L filter.

Once the master is created I dump the other files and then redo as and when is needed. As long as your filters are kept clean this should be okay but it depends on your set up and situation.

I should set up my kit list, however a lowly Canon 650D, recently modded - hooray, and added an IDAS LP D2 filter and wehay the nebula are actually there, not just 'fake news'. Yes I was beginning to consider storage - huge piles of calibration frames that are surely redundant once a master has been made (I am building a calibration library but researching furiously before I bin stuff).

[spillage]

Okay I think with the dslr you can get away with reusing your flats and bias but I would take new dark every session to match the temperature of your lights. I would also try dithering and then you might be able to do away with the bias frames.

[philhilo]

I do dither but was beginning to wonder via research if there was any point in the bias with dithering. I did read that one set was enough. Flats - I have already figured that they aren't going to change unless the camera orientation changes. My thoughts around darks makes me wonder if the time of exposure is in fact irrelevant - is it just temperature critical, the temperature being a factor of the time/ambient. I did read that sensor temp can lag a long way behind the temp read out which is in fact the chip temp. Also use of warm air blowers used to bring the whole lot UP to its likely final temp. I can see the logic in this but wonder if in a situation where time is of the essence if its benefits are more than negligible.

[spillage]

when I used to use a dslr I would always run my flats either at the beginning or end of a session and you do need to make sure they are the same exposure time as your lights.

[Tommohawk]

2 hours ago, spillage said:

when I used to use a dslr I would always run my flats either at the beginning or end of a session and you do need to make sure they are the same exposure time as your lights.

Do you mean darks? Darks definitely need to be same exposure time as lights, but pretty sure flats don't. Flats just need to be about 50% on the histo. it would be pretty difficult to get the exposure time the same as the lights without overexposing.

Flats just correct for the optical issues ie vignetting and dust bunnies.

[spillage]

sorry typo. Yes I meant darks not flats.

[philhilo]

6 hours ago, spillage said:

when I used to use a dslr I would always run my flats either at the beginning or end of a session and you do need to make sure they are the same exposure time as your lights.

Jeepers had me going there for a moment. Definitely a beginner but thought I had that covered off!

[Tommohawk]

16 hours ago, philhilo said:

Jeepers had me going there for a moment. Definitely a beginner but thought I had that covered off!

LOL. But Spillage makes a good point - the best way to temperature match the darks is just to run them off after your session, if possible. Either cap the scope (if its lightproof - but remember light can leak in especially round the mirror if its a newt) or remove the camera and put the cap on, then leave it say in a garage or somewhere at similar temperature. Kind of assumes you have an intervalometer or similar for taking your subs.

Either that, or just do the darks one night when its cloudy - plenty of those! -  at a similar temp. You're right the EXIF temp is the chip/CPU temp, rather than the sensor temp, but if the ambient is the same ish you should be OK.

TBH when I used DSLR I used just a few sets of library darks.

[philhilo]

On 24/12/2019 at 15:45, Tommohawk said:

LOL. But Spillage makes a good point - the best way to temperature match the darks is just to run them off after your session, if possible. Either cap the scope (if its lightproof - but remember light can leak in especially round the mirror if its a newt) or remove the camera and put the cap on, then leave it say in a garage or somewhere at similar temperature. Kind of assumes you have an intervalometer or similar for taking your subs.

Either that, or just do the darks one night when its cloudy - plenty of those! -  at a similar temp. You're right the EXIF temp is the chip/CPU temp, rather than the sensor temp, but if the ambient is the same ish you should be OK.

TBH when I used DSLR I used just a few sets of library darks.

I definitely don't waste clear skies on darks, or I get to a point where I can do no more and leave it with APT to finish up the darks - come back in the morning. Had wondered about light getting in around the mirror of the Newt, I tend to chuck the scope coat on so that's going to be pretty dark combined with the lens cap. I am trying to build a darks library but have just changed DSLRs so its start again time.