Deep Sky Astrophotography
- Thread starter jrista
- Start date
Here are 80D RAW files for Jon and anyone else interested in doing statistical analysis as discussed above: https://www.dropbox.com/sh/iuei2ks8jjn7pj5/AACjB1Rl9nEiZQYm2HTNGFf3a?dl=0jrista said:
Upvote
0
StudentOfLight said:
Here are the results for your data. I'm having problems loading the raw data into PixInsight at the moment...it does not seem to be quite up to date to handle 80D files yet, so I may do some additional fiddling to verify these values. In particular, the dynamic range values seem...unreasonably high for the higher ISO settings. That may be due to the fact that the overscan area is not being cropped out properly by PI at the moment, which seems to be giving me improbably high FWC values at each ISO setting. I do believe the read noise, offset and dark current levels are correct, though:
ISO 100:
Read Noise: ~6.3e- RMS
Dark Current: ~1.2e-/s
Offset: 128 ADU
FWC: ~139127e- (HIGHLY IMPROBABLE!!)
Gain: 8.4e-/ADU 14-bit
DR: 14.5 stops (HIGHLY IMPROBABLE!!)
ISO 200:
Read Noise: ~4.45e- RMS
Dark Current: ~1e-/s
Offset: 128 ADU
FWC: ~72394e- (HIGHLY IMPROBABLE!!)
Gain: 4.4e-/ADU 14-bit
DR: 14 stops (HIGHLY IMPROBABLE!!)
ISO 400:
Read Noise: ~3.63e- RMS
Dark Current: ~0.25e-/s
Offset: 512 ADU
FWC: ~36715e- (Improbable)
Gain: 2.25e-/ADU 14-bit
DR: 13.35 stops (Rather Improbable at this ISO)
ISO 800:
Read Noise: ~3.3e- RMS
Dark Current: ~0.23e-/s
Offset: 512 ADU
FWC: ~18900e- (Seems improbable)
Gain: 1.155e-/ADU 14-bit
DR: 12.5 stops (Rather Improbable at this ISO)
ISO 1600:
Read Noise: ~2.95e- RMS
Dark Current: ~0.18e-/s
Offset: 512 ADU
FWC: ~9550e- (May be possible, but unlikely for this camera and pixel size)
Gain: 0.583e-/ADU 14-bit
DR: 11.7 stops (Rather improbable at this ISO)
I think the issue may be the flats. Your flat levels seem to be very low...only a couple thousand ADU. Is there any way you can give me more exposed flats? If you can get the histogram to peak around 1/2 histogram for all the flats at each ISO, that should give me better results.
Another thing that seems to be a bit problematic is, the dark frames don't seem to have a higher level than the bias frames. Even with low dark current (and, based on the numbers I'm getting, the dark current is NOT low), the level of the dark frames should be higher than the level of the bias frames...however, that is not the case. That is resulting in some of the mean differences between darks and biases being negative, which might be contributing to the unrealistically large full well capacities.
I'll need better data to get you accurate results. For now, disregard the FWC and DR values above...as they are very likely NOT real. (And, well, if they are real...Damn Canon! Way to go!!!)
One thing I did notice. There is DEFINITELY a change in the offset level at ISO 100 and 200. The offset is 128 14-bit ADU until ISO 400, where it changes back to Canon's standard 512 ADU offset, which is also used for all other ISO settings.
Upvote
0
I shot all the images under the same indoor ambient conditions. I made sure to turn off all the noise reduction options in the shoot settings menu. The long exposures darks are what they are. Bias frame as well.
The original flats were perhaps a bit too dark. What I'll do is re-shoot the flats with ETTL flash (+0.7EV) at ISO 200, 400, 800, 1600 and 3200. That gives a neat peak in all channels slightly to the right of mid-tone, with no highlight clipping.
They'll take a while to upload due to my internet connection. Apologies for not doing them better the first time round.
The original flats were perhaps a bit too dark. What I'll do is re-shoot the flats with ETTL flash (+0.7EV) at ISO 200, 400, 800, 1600 and 3200. That gives a neat peak in all channels slightly to the right of mid-tone, with no highlight clipping.
They'll take a while to upload due to my internet connection. Apologies for not doing them better the first time round.
Upvote
0
Alright. Hopefully the flats will do it.
I evaluated the dark frames more deeply. It seems that Canon does indeed employ some kind of dark current offset removal technology. I don't know that I would call it dark current suppression per-se...I would need to evaluate a range of darks across a wide range of temperatures to evaluate that properly. However, it does appear as though any dark current OFFSET is removed by the camera. That means that the only thing that changes in the frames is the dark current noise, and that is definitely evident when comparing a 30s to a 300s dark frame.
I think there may be interesting implications for dynamic range with such technology. Usually, longer exposures suffer a loss of dynamic range, as the offset increases from dark current. As the offset grows, your dynamic range shrinks. However, if the offset never grows, then you would never lose any dynamic range when doing long exposures. That is quite interesting. The 5D III might have had an early form of that technology. While it did not have a 0 offset difference, the offset with a 600s sub was only slightly more than a 60s sub, while the dark current noise was much greater. Seems Canon has refined the technology.
I evaluated the dark frames more deeply. It seems that Canon does indeed employ some kind of dark current offset removal technology. I don't know that I would call it dark current suppression per-se...I would need to evaluate a range of darks across a wide range of temperatures to evaluate that properly. However, it does appear as though any dark current OFFSET is removed by the camera. That means that the only thing that changes in the frames is the dark current noise, and that is definitely evident when comparing a 30s to a 300s dark frame.
I think there may be interesting implications for dynamic range with such technology. Usually, longer exposures suffer a loss of dynamic range, as the offset increases from dark current. As the offset grows, your dynamic range shrinks. However, if the offset never grows, then you would never lose any dynamic range when doing long exposures. That is quite interesting. The 5D III might have had an early form of that technology. While it did not have a 0 offset difference, the offset with a 600s sub was only slightly more than a 60s sub, while the dark current noise was much greater. Seems Canon has refined the technology.
Upvote
0
Pacman Nebula in HSO
I thought I'd shared this before, but apparently I had not. This is a tri-channel narrow band image, using the HSO mapping (Red=Hydrogen-alpha, Green=Sulfur-II, Blue=Oxygen-III), albeit with a few blending tweaks of my own to produce a more pleasing blend.
This is about six hours of total integrated exposure time:
I thought I'd shared this before, but apparently I had not. This is a tri-channel narrow band image, using the HSO mapping (Red=Hydrogen-alpha, Green=Sulfur-II, Blue=Oxygen-III), albeit with a few blending tweaks of my own to produce a more pleasing blend.
This is about six hours of total integrated exposure time:
Upvote
0
I must agree with Lion Rock, the best astro images are ones like that where you can really get a sense of depth to the images. Achieving the necessary precision of capturing faint details and maintaining good tonality in highlights and shadows really gives a 3D-look. It takes many hours of dedication and true craftsmanship to create such amazing images. Jon, you are such an inspiration. Thanks for all your input on CR.
Upvote
0
StudentofLight,
Thanks for the support. Though I beg to differ on one point:
It takes longer than HOURS of dedication. It would really take years to get to this exquisite state of results. Some of us may take a lifetime and still not get there. I would like to attempt some shots of the night sky, but there is still work that gets IN THE WAY! Sob.
Jon is really an inspiration.
-r
Thanks for the support. Though I beg to differ on one point:
It takes longer than HOURS of dedication. It would really take years to get to this exquisite state of results. Some of us may take a lifetime and still not get there. I would like to attempt some shots of the night sky, but there is still work that gets IN THE WAY! Sob.
Jon is really an inspiration.
-r
StudentOfLight said:
Upvote
0
Geez. You guys are going to make my head burst into flames here. 
I really do encourage you guys to explore astrophotography. It is a very challenging hobby, and definitely more work than regular daytime photography (each and every single image I create these days usually involves tens of hours of actual exposure, and tens of hours more post-processing work, sometimes with second and third rounds of processing). It takes definite dedication, good software for processing, and also takes some technical and mathematical prowess, but if you put your mind to it, you can create some amazing images. In the end it can be a very satisfying hobby. Just be prepared to invest some TIME!
I've seen a number of you pick the hobby up and start creating some great images here just in the last year or so, which is great!
I really do encourage you guys to explore astrophotography. It is a very challenging hobby, and definitely more work than regular daytime photography (each and every single image I create these days usually involves tens of hours of actual exposure, and tens of hours more post-processing work, sometimes with second and third rounds of processing). It takes definite dedication, good software for processing, and also takes some technical and mathematical prowess, but if you put your mind to it, you can create some amazing images. In the end it can be a very satisfying hobby. Just be prepared to invest some TIME!
I've seen a number of you pick the hobby up and start creating some great images here just in the last year or so, which is great!
Upvote
0
Yes it definitely requires a great deal of time and dedication to get results like Jon has.
Pixinsight is definitely the tool to master but it takes more than a month to learn the software and unless you have $$$ and time to throw at it, it may not be the best tool for everyone.
The rest of us have to live with mediocrity using free tools like deep sky stacker and post processing tools better suited for weddings and bar mitzvahs. I have tried my best over the years and have realized that I just don't have the time, money, and dedication to throw at it...after all I have teenagers getting ready for college...one of them mentioned law school. 8/.
While I wait for an automated tool that can get me 75% there, I live with 10% and a lot of unfinished projects. I am resolved to just frequent this forum and eagerly await the latest image that Jon has collected and processed. Hopefully along the way, he will share something that will inspire me to try again and again and again.
I'm disappointed that he has moved away from using a DSLR as that is what the rest of us minions can afford to use in both dollars and time. Seeing images created using monochrome sensors and a plethora of filters is nice but I can see those by visiting NASAs HST site. It's great to see what people are capturing in their backyard using such but it doesn't help me get to my 75% goal and probably not applicable to this CR forum.
I'm hoping to get some time in this weekend if weather permits...maybe I can get to 15%.
Pixinsight is definitely the tool to master but it takes more than a month to learn the software and unless you have $$$ and time to throw at it, it may not be the best tool for everyone.
The rest of us have to live with mediocrity using free tools like deep sky stacker and post processing tools better suited for weddings and bar mitzvahs. I have tried my best over the years and have realized that I just don't have the time, money, and dedication to throw at it...after all I have teenagers getting ready for college...one of them mentioned law school. 8/.
While I wait for an automated tool that can get me 75% there, I live with 10% and a lot of unfinished projects. I am resolved to just frequent this forum and eagerly await the latest image that Jon has collected and processed. Hopefully along the way, he will share something that will inspire me to try again and again and again.
I'm disappointed that he has moved away from using a DSLR as that is what the rest of us minions can afford to use in both dollars and time. Seeing images created using monochrome sensors and a plethora of filters is nice but I can see those by visiting NASAs HST site. It's great to see what people are capturing in their backyard using such but it doesn't help me get to my 75% goal and probably not applicable to this CR forum.
I'm hoping to get some time in this weekend if weather permits...maybe I can get to 15%.
Upvote
0
jrista said:
Jon, I'm amazed at the images you produce and may someday try my hand at astro (when I have more time). In the meantime I'll keep learning from your informative posts.
Now that you've done the 5D IV, would you like to analyze a 1DX II? I posted a set of biases, darks (30s and 300s), and midtone flats for ISO 100 to 25600 at https://www.dropbox.com/sh/rqzzaff4jeu364a/AABRst_qMczLSIS8MNWC-nBia if you are interested. Hopefully I produced them correctly. If not, let me know and I'll make any necessary adjustments.
Upvote
0