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.
Hello once again!
Got a quick question if it is possible to create a HDR stack, or generally, how APP would treat it for example, if I stacked a HDR target like M42 oder M31 with several lets say 30-60 second exposures alongside 5 minute exposures. Would the bright parts get burned out still?
CS Tobi
Did you get a reply. I'm trying to combine 30s and 180s exposures of M31. Looks like PixInsight has a HDRcombine tool but don't see any solution with APP.
@schakravarthi No, sadly no reply yet. 🙁 Hopefully Mabula or Vincent sees this post soon.
Before I open another topic, another quick question:
When I combine multiple sessions, and I want to add one Masterdark for all (for example). Would there be a difference between selecting "Only 1 session" or "All sessions" when loading it in?
@schakravarthi No, sadly no reply yet. 🙁 Hopefully Mabula or Vincent sees this post soon.
I do think, there is nothing new to report:
https://www.astropixelprocessor.com/community/postid/28343/
When I combine multiple sessions, and I want to add one Masterdark for all (for example). Would there be a difference between selecting "Only 1 session" or "All sessions" when loading it in?
Well, if you combine all single sessions in one session, then there is no difference (1 session = all sessions).
But if you do load multiple sessions, you surely need a masterdark for every session, these can be different ones (per session) or one masterdark for all sessions.
Or different speaking: every session needs its own calibration frames. but you surely can use a calibration frame for one, some or all sessions. depending if these calibration frames are matching the sessions lights.
Hello once again!
Got a quick question if it is possible to create a HDR stack, or generally, how APP would treat it for example, if I stacked a HDR target like M42 oder M31 with several lets say 30-60 second exposures alongside 5 minute exposures. Would the bright parts get burned out still?
CS Tobi
Hi @tobigh3 & @schakravarthi
Simply load all those different exposures into APP and it will automatically create a HDR stack composed of the different exposure times. If you want to favor the shorter exposures, to get sharper stars and less burnt highlights, choose a weighting method like star shap to favor the images with the smallest stars which are normally in the shortest exposures 😉
A special HDR combine is on our ToDo list to make part of the RGB Combine Tool going forward, but my suggested method above works well in my experience.
Mabula
@schakravarthi No, sadly no reply yet. 🙁 Hopefully Mabula or Vincent sees this post soon.
Before I open another topic, another quick question:
When I combine multiple sessions, and I want to add one Masterdark for all (for example). Would there be a difference between selecting "Only 1 session" or "All sessions" when loading it in?
LIke Sebastian indicates, if you do load lights with different session tags and you want the MasterDark to apply on all those session, either tag the masterDark with those sessions are use all sessions if it applies to all sessions. And you can always check what is applied to which light in the frame column. After 2) Calibrate, all masters should be attached in the frame column of the frame list panel to all lights, and if something is not right, you can see it there 😉 And that should point out what you need to fix then.
Mabula
Awesome, that is great to hear!
Just one last question regarding this:
Can I also use very few, or in theory even 1 subframe with a short exposure just so that the highlights aren't burnt out? Or is still a certain sample size recommended?
Simply load all those different exposures into APP and it will automatically create a HDR stack composed of the different exposure times. If you want to favor the shorter exposures, to get sharper stars and less burnt highlights, choose a weighting method like star shap to favor the images with the smallest stars which are normally in the shortest exposures 😉
A special HDR combine is on our ToDo list to make part of the RGB Combine Tool going forward, but my suggested method above works well in my experience.
It‘s surely exciting news that this method does work well.
But now I got some more questions about this (sorry to highjack your post @tobigh3)...
Will the described method suppress nebulosity, especially fainter regions in some way? (Or dont you „loose“ information at all by stacking this way (except outlier rejection)).
If you stack different sessions (and the long and short ones surely are two different sessions, because i do need at least different dark frames), how will APP handle this?
As far as I can imagine APP will stack the different sessions and then stack the results. And I do expect they will be weighted by their stacked exposure times? (So stacking everything in one go or stacking the results would give more or less the same result (as I do assume these are commutativ/associativ in some way).
Last question is based on these assumtions:
If I do stack the „star session“ by star shape and the other session by quality, and then stack the results again, can I expect do get a „better“ result than stacking everything by star shape?
Sorry, that these are mere „basic“ questions about stacking, but they are directly related to this topic.
Can I also use very few, or in theory even 1 subframe with a short exposure just so that the highlights aren't burnt out? Or is still a certain sample size recommended?
If you stack only a few short exposures to a big stack of long exposures, I do think, they will have only a little impact on the whole stack. Imagine if only 1 subframe of short exposure would be enough to have a big impact, you would insert all the noise of this 1 subframe. So, to supress noise in the short exposure stack, you should get some more short exposures, just to even that out. But this is only my "opinion"...
I do think this is why it will be nice to have the combine tool upgraded to hdr some day. So we can be selective on what to blend into the final image.
@xyfus I understand what you mean. My point was, that maybe APP would detect the burnt out parts of the long exposures, and then just take the bright parts of the short exposure so that they are not burnt out anymore, if you get what I mean. 🙂
Awesome, that is great to hear!
Just one last question regarding this:Can I also use very few, or in theory even 1 subframe with a short exposure just so that the highlights aren't burnt out? Or is still a certain sample size recommended?
@tobigh3, the sample size will affect how much weight in total it will have influence in the final result. So yes, the more samples the more effect.
Will the described method suppress nebulosity, especially fainter regions in some way? (Or dont you „loose“ information at all by stacking this way (except outlier rejection)).
No, there are no algorithms at all that would do such a thing. It is not that magical, what you enter the machine will be used and will be part of the integration, the weights determine how it will look.
If you stack different sessions (and the long and short ones surely are two different sessions, because i do need at least different dark frames), how will APP handle this?
At the top of 6) Integrate, you can choose this. You can stack per session or have them all combined into 1 stack. APP will apply darks to lights by checking the exposure times. If you load the darks for all sessions, the correct masterdark will still be applied to the correct lights.
As far as I can imagine APP will stack the different sessions and then stack the results. And I do expect they will be weighted by their stacked exposure times? (So stacking everything in one go or stacking the results would give more or less the same result (as I do assume these are commutativ/associativ in some way).
The session stacks or the stack of all sessions will use the weights as chosen by you. If you integrate in automatic mode, the quality weights are used. So you have full control over which weights are used per stack if you do not use the automatic integrate mode.
If I do stack the „star session“ by star shape and the other session by quality, and then stack the results again, can I expect do get a „better“ result than stacking everything by star shape?
No, this all depends on the whole dataset what results you would get with using different weights. But, in general, the quality weights is the best compromise leading to the best stacks normally but it does not mean it will always be the case. And which stack is "better" is not that objective if you ask me, it can be rather subjective and it depends on what you aim to get. Some users only focus on least noise, others only on the most sharp details... and you can not have it both ways. I prefer the middle ground between these 2.
If you stack only a few short exposures to a big stack of long exposures, I do think, they will have only a little impact on the whole stack. Imagine if only 1 subframe of short exposure would be enough to have a big impact, you would insert all the noise of this 1 subframe. So, to supress noise in the short exposure stack, you should get some more short exposures, just to even that out. But this is only my "opinion"...
I do think this is why it will be nice to have the combine tool upgraded to hdr some day. So we can be selective on what to blend into the final image.
Indeed, the sample size controls this 😉 With a HDR dedicated option you will get better control over this.
@xyfus I understand what you mean. My point was, that maybe APP would detect the burnt out parts of the long exposures, and then just take the bright parts of the short exposure so that they are not burnt out anymore, if you get what I mean. 🙂
@tobigh3, no that is not what is happening. we do not detect burnt out areas and skip those in the pixel stacks, it would lead even to ugly artefacts I would think at the borders of these areas in the stacked result.
I understand, thanks for the explanation.
Still, I can't really wrap my head around on how to make use of this exactly.
For example, if I want to image the orion nebula, and I do the regular 2-3 minute exposures, the core will obviously be massively burnt out.
Let's say I then just want to add a few 4-5 seconds exposures, just for the core. Lets say I do 50 subs each 4 seconds, that is only 3 minutes, but would be enough.
But if I then have let's say 8 hours of 2-3 minute exposures, and just the 3 minutes of 4 second exposures, will this even be noticable?
Or should I rather take the "conventional" approach, and separately work with the short exposures and use masks for example in PS?
If you woudl use exposures as integrating weights, than the short exposures will not cotribute. But with the other weighting methods like quality weights or star shape weights, then the amount of images is the factor. So combining 50 exposures of only 5 seconds with 50 exposures of 2-3 minutes will work nicely for such a HDR object. The proof is in the pudding, simply try it 😉
Mabula
