He is correct that one 80 minute exposure photo contains less noise than eight 10 minute exposure frames that are stacked. You can minimize the noise in the stacked image using bias, flat, and dark frames but it still will have on average a bit more noise for a given exposure. But also the type of noise from photon to thermal to read noise also are exposure biased too like for example more thermal noise exists because the sensor is active continuously instead of a short intermittent rest periods between frames so in general the thermal noise trade off is negligible, photon noise depends on the ISO/ Exposure compensations you're making to get a properly exposed frame. So why in the astrophotography circles is stacking shorter exposure frames preferable than one long frame ... it's because of non-static artefacts or in other words an insurance policy against the random daily events. You have planes whizzing by in your FOV as well as satellites or the ISS. Having many moderately exposed frames means that when you are stacking the frames if two of the eight are bad you still have six of the 10 minute exposure frames to fallback on giving you a 60 minute integrated stacked image. That peace of mind is more worthwhile because you ultimately don't want to waste time capturing an image that the next day you realize was spoiled. Also 80 minute exposures will capture the faintest of lights that are from very distant objects ... in weird way I would say in astrophotography shorter exposures acts as a temporal filter. There's an amount of faint light that isn't worth capturing to get the image composition you are looking for. So through experience you'll figure out an exposure minimum like 1-2 minutes and exposure maximum like 5 minutes that is well suited for your camera and fov of the environment (ie. frequent air traffic). Actual filter usage (R, G, B, CLS, UHC, L-Pro, HIII, OII, Broadband, etc...) also plays a role in figuring out the proper exposure times. It's really a gut instinct trial/ error thing and I haven't found a formula based approach to this. Funny enough this comparison is also recursive in that a single 80 minute exposure might be able to have the SNR improved by stacking it with other 80 minute exposure frames. A better way to think of it is to establish your plan for the night ie. 4hr's on a target. Then live by a heuristic like for me it's no less than 15 images or no more than 30 images to stack per filter. With those constraints work backwards as to how to establish your session on schedule. Give yourself some cushion for test shots and setup like 30-120 minutes before the 4hr session begins. Also remember that calibration images require plenty amount of time too although you may have built yourself a library of calibration images so you might be able to not require some of these images hence saving you time, keep a journal, bring binoculars, tea, and you wait for the acquisition.
g'day Mark. just watched Jesse's short presentation and enjoyed his words and images. I did find it surprising that no mention (that I can recall) of light leakage from the camera view finder to the sensor during longer night exposures and how to prevent that from occurring. being a Canon user myself, I place a piece of black electric tape over the view finder which is handy if there is a light source from the rear of the camera. cheers
I wish Jesse had told us how he arrived at the exposure times, aperture sizes and speeds for each snapshot- rather than how clever he was to get the photos or where they had been taken!
Krishnan Raman Menon he did say he used 30 second test shots at 6400 ISO and fully opened aperture then adjusted to the longer exposure times from there
(bit late but..) he did. He said that he cranked up the ISO to maximum, so he can not only see the framing, but also the exposure. Now you only have to use a formular to calculate the aperture and ISO for longer exposure times, which you can do in most light meter apps.
He is correct that one 80 minute exposure photo contains less noise than eight 10 minute exposure frames that are stacked. You can minimize the noise in the stacked image using bias, flat, and dark frames but it still will have on average a bit more noise for a given exposure. But also the type of noise from photon to thermal to read noise also are exposure biased too like for example more thermal noise exists because the sensor is active continuously instead of a short intermittent rest periods between frames so in general the thermal noise trade off is negligible, photon noise depends on the ISO/ Exposure compensations you're making to get a properly exposed frame. So why in the astrophotography circles is stacking shorter exposure frames preferable than one long frame ... it's because of non-static artefacts or in other words an insurance policy against the random daily events. You have planes whizzing by in your FOV as well as satellites or the ISS. Having many moderately exposed frames means that when you are stacking the frames if two of the eight are bad you still have six of the 10 minute exposure frames to fallback on giving you a 60 minute integrated stacked image. That peace of mind is more worthwhile because you ultimately don't want to waste time capturing an image that the next day you realize was spoiled.
Also 80 minute exposures will capture the faintest of lights that are from very distant objects ... in weird way I would say in astrophotography shorter exposures acts as a temporal filter. There's an amount of faint light that isn't worth capturing to get the image composition you are looking for. So through experience you'll figure out an exposure minimum like 1-2 minutes and exposure maximum like 5 minutes that is well suited for your camera and fov of the environment (ie. frequent air traffic). Actual filter usage (R, G, B, CLS, UHC, L-Pro, HIII, OII, Broadband, etc...) also plays a role in figuring out the proper exposure times. It's really a gut instinct trial/ error thing and I haven't found a formula based approach to this. Funny enough this comparison is also recursive in that a single 80 minute exposure might be able to have the SNR improved by stacking it with other 80 minute exposure frames. A better way to think of it is to establish your plan for the night ie. 4hr's on a target. Then live by a heuristic like for me it's no less than 15 images or no more than 30 images to stack per filter. With those constraints work backwards as to how to establish your session on schedule. Give yourself some cushion for test shots and setup like 30-120 minutes before the 4hr session begins. Also remember that calibration images require plenty amount of time too although you may have built yourself a library of calibration images so you might be able to not require some of these images hence saving you time, keep a journal, bring binoculars, tea, and you wait for the acquisition.
g'day Mark. just watched Jesse's short presentation and enjoyed his words and images. I did find it surprising that no mention (that I can recall) of light leakage from the camera view finder to the sensor during longer night exposures and how to prevent that from occurring. being a Canon user myself, I place a piece of black electric tape over the view finder which is handy if there is a light source from the rear of the camera. cheers
sportz fotos Even later to the party, but, he actually says that when he gives the example of shooting long exposure inside a hotel room
Thank you very much
I wish Jesse had told us how he arrived at the exposure times, aperture sizes and speeds for each snapshot- rather than how clever he was to get the photos or where they had been taken!
Krishnan Raman Menon he did say he used 30 second test shots at 6400 ISO and fully opened aperture then adjusted to the longer exposure times from there
(bit late but..) he did. He said that he cranked up the ISO to maximum, so he can not only see the framing, but also the exposure. Now you only have to use a formular to calculate the aperture and ISO for longer exposure times, which you can do in most light meter apps.
Thanks unto L10