Had the chance to use the R5 for astrophotography

Nov 12, 2016
910
615
I used my new R5 for some astrophotography this past week. I'm pretty happy with the results. Even pushing the shadows pretty hard at ISO 1600, the foreground detail held up fairly well. Here is a sample image for anyone considering the R5 for astro work. Please note that I don't consider myself to be a pro at editing astro photos. This was just done with some quick basic edits and noise reduction in Adobe Camera Raw.

Taken with a Rokinon 14mm f2.8, 35 seconds, ISO 1600
1R5_0651.jpg

And the original, no edits, for comparison.
1R5_0651-copy.jpg

And I know people want to pixel peep, so here's a 1:1 crop of the foreground detail in the first photo. This has some noise reduction applied. It's not perfect by any means, but I'm fairly happy with it, especially being that it's in the corner of the photo. Also keep in mind that this is a 45mp photo, so looking at 1:1 crops is looking a lot closer than what 1:1 would be from a 5D4 or EOS R.
1R5_0651crop.jpg
 
  • Like
Reactions: 3 users
Nov 12, 2016
910
615
That 1:1 doesn't look good to me. How much did you raise the shadows and or exposure by?
Pretty much as much as I could without the image totally falling apart. Again, not claiming to be an astrophotography pro, just putting this out there for people to look at what the camera can do and make their own decisions.

adjustments.jpg
 
Upvote 0

pmjm

R5, 1DX Mk II, 5D Mk IV, four 90D's
Sep 8, 2016
78
46
I grabbed the moon at 1200mm (Tamron 150-600 gen 1 with a 2X teleconverter @ f/11) on the R5, it came out decent, but not as sharp as if I had taken it without the teleconverter. I was just going for high pixel count on this one. I no longer have the original but here it is with contrast, detail tweaks.



Moon_080320_R5_Teleconverter_1200mm_#2.png

For comparison, here it is the night before at 400mm with no teleconverter. Much sharper.

Moon_080220_R5.png
 
  • Like
Reactions: 1 users
Upvote 0
I used my new R5 for some astrophotography this past week. I'm pretty happy with the results. Even pushing the shadows pretty hard at ISO 1600, the foreground detail held up fairly well. Here is a sample image for anyone considering the R5 for astro work. Please note that I don't consider myself to be a pro at editing astro photos. This was just done with some quick basic edits and noise reduction in Adobe Camera Raw.

Taken with a Rokinon 14mm f2.8, 35 seconds, ISO 1600
View attachment 191999

And the original, no edits, for comparison.
View attachment 192000

And I know people want to pixel peep, so here's a 1:1 crop of the foreground detail in the first photo. This has some noise reduction applied. It's not perfect by any means, but I'm fairly happy with it, especially being that it's in the corner of the photo. Also keep in mind that this is a 45mp photo, so looking at 1:1 crops is looking a lot closer than what 1:1 would be from a 5D4 or EOS R.
View attachment 192001

Thanks for posting this. It looks like you could have gotten away with a brighter exposure as a starting point. Raising the shadows by +100 at 1600 iso is less than ideal ;)
 
  • Like
Reactions: 1 user
Upvote 0

SteveC

R5
CR Pro
Sep 3, 2019
2,678
2,592
Well I got much better results with the Milky Way. The biggest problem is I can't lock the lens focal length and if I accidentally touch it fumbling in the dark, I have to focus again. But by putting the camera on spot focus and putting that spot against a light on the horizon, I can then back-button focus it. Then I just have to swing the camera back to where it should be. (I realize if you live in a wooded or urban area finding a single light several miles away on the horizon might be impossible.)
 
  • Like
Reactions: 1 user
Upvote 0
Jan 29, 2011
10,675
6,121
You should be able to get a much cleaner image than that. Here is a badly underexposed test shot of a wall from a 1DX II (so last generation sensor) I input the same RAW settings as yours and then opened it in PS and resized it to the same size as the R5, 45mp. The first shot is before and after in LR and the second screenshot is a crop of 1:1 copy of that resized PS file. Only LR NR done nothing fancy, I only moved it to PS to resize it for a more relevant comparison.

1596949011748.png

1596948874847.png
 
Upvote 0

SteveC

R5
CR Pro
Sep 3, 2019
2,678
2,592
Nothing earthshattering here, but... In both cases I raised the contrast a bit, in one case I cropped out really bright ground.

R5_L2720_small.JPG

The really bright "star" is Jupiter, up a bit and well to the left, that second brightest "star" is Saturn. They're going to get a LOT closer to each other in December...close enough to possibly appear in a telescope simultaneously. 24.0 mm 10sec, f/2.8, iso 10000.

R5_L2726_small.JPG

The continuation of the Milky way upwards from the horizon takes you into the Summer Triangle (if you're in the Northern Hemisphere). Same exposure info. The bright star top and center is vega, the brightest star in the lower right quarter is altair, and Deneb is at about mid height, a quarter of the way into the image from the left. to the right of Deneb are three stars (fairly widely spaced) that form the "crossbar" of the northern cross, the other two stars in the vertical (which is horizontal in this picture) are a lot harder to pick out.
 
  • Like
Reactions: 5 users
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
You should be able to get a much cleaner image than that. Here is a badly underexposed test shot of a wall from a 1DX II (so last generation sensor) I input the same RAW settings as yours and then opened it in PS and resized it to the same size as the R5, 45mp. The first shot is before and after in LR and the second screenshot is a crop of 1:1 copy of that resized PS file. Only LR NR done nothing fancy, I only moved it to PS to resize it for a more relevant comparison.

View attachment 192007
Your wall looks a lot brighter than the OP's image after the push to me. So it appears you have more signal here.

Also, as was mentioned by the OP, magnification matters a lot. The 1:1 crop from the OP can't be compared with a 1DX II shot. One that is downsampled to 20 MP could, but the OP would either have to upload the original file or give that to you.

And lastly, the OP's exposure time is 35 s. How long was yours? I don't expect Dark Current noise to be much of a factor here, but it tends to show up in strong shadow pushs in my experience, so is a thing worth considering.
 
  • Like
Reactions: 1 user
Upvote 0
Nov 12, 2016
910
615
Your wall looks a lot brighter than the OP's image after the push to me. So it appears you have more signal here.

Also, as was mentioned by the OP, magnification matters a lot. The 1:1 crop from the OP can't be compared with a 1DX II shot. One that is downsampled to 20 MP could, but the OP would either have to upload the original file or give that to you.

And lastly, the OP's exposure time is 35 s. How long was yours? I don't expect Dark Current noise to be much of a factor here, but it tends to show up in strong shadow pushs in my experience, so is a thing worth considering.
They did say they enlarged the brick wall photo to be equivalent to 45mp. But otherwise I agree with you. How long was the exposure of the brick wall? Unless it was also 35 seconds, I don't think it's an apples to apples comparison.
 
Upvote 0
Nov 12, 2016
910
615
Thanks for posting this. It looks like you could have gotten away with a brighter exposure as a starting point. Raising the shadows by +100 at 1600 iso is less than ideal ;)
Yeah maybe. Like I said, not claiming to be a pro at astrophotography. Still figuring out the tradeoffs between running a higher ISO, or boosting it in post.
 
  • Like
Reactions: 1 user
Upvote 0

cornieleous

5D4 + R5
Jul 13, 2020
208
737
Yeah maybe. Like I said, not claiming to be a pro at astrophotography. Still figuring out the tradeoffs between running a higher ISO, or boosting it in post.
Until the sensor is completely ISO invariant (not really an accurate term but a popular one) using the correct exposure is best or you end up adding back that perceived savings in noise plus more in post. While the latest Canon sensors are closer to the so called ISO invariance, still not quite there. Also I feel its better to see what the image will look like as a RAW than underexposed for noise in camera.

I leave dark areas dark in my astro landscapes, pixel peeling beyond print size for astro doesn't make sense to me, and use of denoise and stacking can help tremendously. From my first test of this sensor it is better than any previous Canon and very likely better than most cameras out there since for low light high ISO Canon has always been near the top.

Here is a quick test at F2.8, 15s, 6400, Samyang 24mm on R5. Image is extremely compressed for web and downsized to 20% original. Light post processing.
117528259_10100499118655218_9181429015684955488_o.jpg
 
  • Like
  • Wow
Reactions: 3 users
Upvote 0
Jan 29, 2011
10,675
6,121
They did say they enlarged the brick wall photo to be equivalent to 45mp. But otherwise I agree with you. How long was the exposure of the brick wall? Unless it was also 35 seconds, I don't think it's an apples to apples comparison.
I used the only 1600iso exposure I had that was also pretty underexposed (I normally delete them).

Anyway I wasn't trying to either put you down or pick a fight, merely point out that you should be able to process the image you got even better than you did even just with LR.

I took a picture of the inside of my wife's closet to get a 35 second 1600iso image that had a decent bit of underexposure.

Here is the before/after in LR with your same settings.

1596986377079.png

And here is the 1:1 of an upsized file to 45mp.

1596986565002.png

And then your 1:1 laid onto of my 1:1. Yours should be a touch better than mine if you are managing to get the cleanest results optimal processing can give you even just out of LR.

1596986926603.png
 
Upvote 0
Aug 27, 2019
667
1,414
There is no form of photography that excites me more than astrophotography.

There is no form of photography that pisses me off more than astrophotography.

I have spent many cold nights taking what can only be the most amazing photos while on site only to be revealed as the complete shit they are once downloaded on to my computer.

I have read/watched everything I can on the subject and I am convinced the sky hates me and I will never never waste my time again with it....

Hmm there is a meteor shower this week.

I guess I can give it another go.
 
  • Haha
  • Like
Reactions: 5 users
Upvote 0
Nov 12, 2016
910
615
I used the only 1600iso exposure I had that was also pretty underexposed (I normally delete them).

Anyway I wasn't trying to either put you down or pick a fight, merely point out that you should be able to process the image you got even better than you did even just with LR.

I took a picture of the inside of my wife's closet to get a 35 second 1600iso image that had a decent bit of underexposure.

Here is the before/after in LR with your same settings.



And here is the 1:1 of an upsized file to 45mp.



And then your 1:1 laid onto of my 1:1. Yours should be a touch better than mine if you are managing to get the cleanest results optimal processing can give you even just out of LR.

No offense taken. I'm always open to learn and understand more. I'm curious to know how you got so much better results getting rid of the noise. I just played around with the noise reduction slider until I found a decent compromise between detail and noise.
 
Upvote 0
Mar 25, 2011
16,848
1,835
I wonder if there is a significant difference with a R6 and its 20 MP sensor. I'm afraid I have no clue as to what is needed for astrophotography except for light sensitivity. Since individual stars are so tiny, does the extra resolution of a 45 MP sensor outweigh the extra light per pixel in a 20 mp sensor? I know that large sensors do much better, but they are still pretty specialized items.

Do the sensors used in big observatories use bayer filters or AA filters? I'm thinking the super large one made by Canon.


I know it gets extremely complex with custom designed cameras linking many sensors to get the larger area. CCD's are also used but I don't know if thats because appropriate CMOS sensors are not available or if CCD's just work out better for the application.

 
Upvote 0
Jan 29, 2011
10,675
6,121
No offense taken. I'm always open to learn and understand more. I'm curious to know how you got so much better results getting rid of the noise. I just played around with the noise reduction slider until I found a decent compromise between detail and noise.
Sharpening and NR are not easy to standardize, especially as I started out with a lower resolution image, although, as you said, I did normalize mine to yours.

But try this as a starting point.

1596992653724.png

If you like PM me a link to the RAW and I'll do what I would do and send you the exact settings for that image, that would give you a much better starting point for your files.
 
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
Do the sensors used in big observatories use bayer filters or AA filters? I'm thinking the super large one made by Canon.
I have no idea with regards to 'big observatories' but a lot of people in the astro photography enthusiast community use monochrome CCD cameras. So no bayer filters. You get more light, and pick the wavelength (a specific color, infrared, ultraviolet, specific molecule emissons like hydrogen-alpha, ...) that you want to capture by installing a filter.

As for resolution, if you capture stars, I don't think there's much of a benefit to be had from the additional resolution of the R5.
 
Upvote 0