Canon will release the EOS Ra astrophotography camera

KirkD

EOS T7i
Nov 23, 2017
68
44
Canada
kirkdurston.wixsite.com
From the above mentioned leaflet I can see that even though I dabble at astrophotography, this will not be useful to me, as I do a lot more nature and landscape shooting and they say, "shooting normal subjects with this camera is not recommended." The hydrogen-alpha wavelength is good for extended objects such as nebulae, but you really need to stick your camera on a telescope to get satisfactory magnification of the nebulae. So I think this is not for fellows who stick their camera on a tripod. Instead, it will be for those who mount it on the back of a telescope, which is fairly niche.
  • The EOS Ra is a version of the EOS R designed for astrophotography. This camera has approximately four times
    the transmittance of hydrogen-alpha light (656 nm) as the EOS R.Photographs of subjects that reflect a lot of infrared light will thereforeappear redder than they actually are. Also, as it may not be possible to obtain an appropriate color balance or uneven colors may result,shooting normal subjects with this camera is not recommended.
 
  • Like
Reactions: prodorshak

CanonFanBoy

EOS 5D SR
Jan 28, 2015
4,173
1,757
Irving, Texas
From the above mentioned leaflet I can see that even though I dabble at astrophotography, this will not be useful to me, as I do a lot more nature and landscape shooting and they say, "shooting normal subjects with this camera is not recommended." The hydrogen-alpha wavelength is good for extended objects such as nebulae, but you really need to stick your camera on a telescope to get satisfactory magnification of the nebulae. So I think this is not for fellows who stick their camera on a tripod. Instead, it will be for those who mount it on the back of a telescope, which is fairly niche.
  • The EOS Ra is a version of the EOS R designed for astrophotography. This camera has approximately four times
    the transmittance of hydrogen-alpha light (656 nm) as the EOS R.Photographs of subjects that reflect a lot of infrared light will thereforeappear redder than they actually are. Also, as it may not be possible to obtain an appropriate color balance or uneven colors may result,shooting normal subjects with this camera is not recommended.
Kirk, could it make for some strangely quirky landscape art shots? Just asking because I have no idea.
 
  • Like
Reactions: KirkD

amorse

EOS 7D MK II
Jan 26, 2017
487
488
www.flickr.com
For those who are curious, here's a photo I took a few weeks ago with my Canon EOS R and Canon EF 35mm f1.4L II, 15 seconds at f1.4, ISO 400. The image is not cropped, so you can see how well it does in the corners. Even at 15 seconds, the star trails are starting to show.

View attachment 186675
I'm not surprised about the trailing, if I'm honest. I use a 5D IV for astro occasionally, and I've found that when I have proper focus, the rule of 500 is not enough. I would commonly use a 14mm f/2.8 open for 30 seconds on a 6D and be satisfied with the results (albeit maybe with less than perfect focus), but I've found that with the same lens on the 5D IV I am hesitant to even go for 25 seconds - much better results at 20 seconds when focus is bang on. I'm pretty confident that I don't have any movement anywhere, but 30 second exposures have noticeable trails when really looking. Maybe 30mp on a sharp lens is just a bit too long for the rule of 500.

By the rule of 500 you should be able to get 14 seconds from a 35mm lens, and I should be able to get >30 seconds on the 14. In practice, however, it seems that from my experience maybe the rule of 350 is more accurate.
 

Joules

EOS RP
Jul 16, 2017
380
304
Hamburg, Germany
It's definitely an astrophotography version of the EOS R: Canon published a leaflet about it a few days ago:
Great find. For how tight lipped they are about leaks a lot of Material slips through from official Canon sources lately. First we get the full specs of the M6 II and 90D from Canon Australia and now this. Maybe they're taking a page out of Googles books?
 

tron

EOS 5D SR
Nov 8, 2011
4,092
390
I do some astrophotography with my Canon EOS R. Apart from the lenses used, the most important thing is the HDR of the sensor. With my EF 35mm f1.4L II, any exposure slower than 15 seconds is too long (unless I'm using a clock drive), so I have to ramp up the ISO and then squeeze every bit of HDR I can in post.
I believe even 15sec is too much with a 30Mp camera at 100%. I do some Landscape astrophotography during summers. The late summers I use my 5DIV which is also 30Mp. If I want to see a star ... a star at 100% I have to be restrained around 15 sec with my ... 14mm lens. Fortunately now I have the 14 1.8 Sigma lens. But yes sometimes I use up to 20 or 25 sec to allow a little trailing in order to achieve a little lower ISO. So I understand your 15 seconds and the reason you select so.
 
  • Like
Reactions: RobbieHat

tron

EOS 5D SR
Nov 8, 2011
4,092
390
I'm not surprised about the trailing, if I'm honest. I use a 5D IV for astro occasionally, and I've found that when I have proper focus, the rule of 500 is not enough. I would commonly use a 14mm f/2.8 open for 30 seconds on a 6D and be satisfied with the results (albeit maybe with less than perfect focus), but I've found that with the same lens on the 5D IV I am hesitant to even go for 25 seconds - much better results at 20 seconds when focus is bang on. I'm pretty confident that I don't have any movement anywhere, but 30 second exposures have noticeable trails when really looking. Maybe 30mp on a sharp lens is just a bit too long for the rule of 500.

By the rule of 500 you should be able to get 14 seconds from a 35mm lens, and I should be able to get >30 seconds on the 14. In practice, however, it seems that from my experience maybe the rule of 350 is more accurate.
Yes this is my experience too. In fact think rule of 250!
 

amorse

EOS 7D MK II
Jan 26, 2017
487
488
www.flickr.com
Yes this is my experience too. In fact think rule of 250!
You might be right! I was so mad when I figured out that 30 seconds was too long on the 5D IV - I've since resorted to median stacking on really high ISO exposures as a solution. It's such a pain, but the results are SO much cleaner than further underexposing, or just using a higher ISO (in many but not all circumstances).
 

amorse

EOS 7D MK II
Jan 26, 2017
487
488
www.flickr.com
I believe even 15sec is too much with a 30Mp camera at 100%. I do some Landscape astrophotography during summers. The late summers I use my 5DIV which is also 30Mp. If I want to see a star ... a star at 100% I have to be restrained around 15 sec with my ... 14mm lens. Fortunately now I have the 14 1.8 Sigma lens. But yes sometimes I use up to 20 or 25 sec to allow a little trailing in order to achieve a little lower ISO. So I understand your 15 seconds and the reason you select so.
How do you like that Sigma 14/1.8? I've been eyeballing that for a while now as an upgrade to my trusty (but light weight) Rokinon 14/2.8.
 

tron

EOS 5D SR
Nov 8, 2011
4,092
390
You might be right! I was so mad when I figured out that 30 seconds was too long on the 5D IV - I've since resorted to median stacking on really high ISO exposures as a solution. It's such a pain, but the results are SO much cleaner than further underexposing, or just using a higher ISO (in many but not all circumstances).
Interesting but let me ask: I do landscape astrophotography in which case the land remains steady (unless there are leaves and there is wind) and the stars which move (OK earth rotates but you get the point). How do you apply the algorithm? Do you break the picture in two parts and process seperately?
 

Joules

EOS RP
Jul 16, 2017
380
304
Hamburg, Germany
Interesting but let me ask: I do landscape astrophotography in which case the land remains steady (unless there are leaves and there is wind) and the stars which move (OK earth rotates but you get the point). How do you apply the algorithm? Do you break the picture in two parts and process seperately?
There is software that can stack the sky and ground separately if you draw a border between them on one reference frame. Sequator is a good example of this, as it is free and provides really good results.

The alternative to getting better astro images is using a sky tracker, but in that case you still have to do processing because you have to merge a shot of the sky (where the ground is blurry) and one of the ground (blurry sky) in post. No free lunch.

So far I've used the former method but I purchased a tracker (Fornax Lightrack II) recently and this weekend looks like I'll have some time and good weather so that will get a chance to shine.
 
Last edited:
  • Like
Reactions: tron

tron

EOS 5D SR
Nov 8, 2011
4,092
390
There is software that can stack the sky and ground separately if you draw a border between them on one reference frame. Sequator is a good example of this, as it is free and provides really good results.

The alternative to getting better astro images is using a sky tracker, but in that case you still have to do processing because you have to merge a shot of the sky (where the ground is blurry) and one of the ground (blurry sky) in post. No free lunch.

So far I've used the former method but I purchased a tracker (Fornax Lightrack II) recently and this weekend looks like I'll have some time and good weather so that will get a chance to shine.
I was thinking about the latter because I have an old Astrotrac model. But the place I was taking pictures didn't allow me to see the polar star due to rocks! So up to now I have done just the half: Took a picture with enough minutes as the shutter speed in order to achieve a low (3 digit) iso and I combined it with one another picture. Not enough though. The method you mentioned seems very interesting :)
 

unfocused

EOS 1D MK II
Jul 20, 2010
5,049
1,434
66
Springfield, IL
www.mgordoncommunications.com
I think it's kind of a stretch to assume that this is an astrophotography camera. It could be, but it seems like Canon would have a lot of other priorities before releasing such a body.
I like to admit when I'm wrong (sometimes. :) ) and clearly I was wrong about this.

It's definitely an astrophotography version of the EOS R: Canon published a leaflet about it a few days ago:
It still surprises me, but I guess it's a way to pick up a few extra sales without much investment.
 
  • Like
Reactions: SteveC and Sharlin

Sharlin

EOS 6D MK II
Dec 26, 2015
1,066
579
Turku, Finland
EOS Ra. A camera special-purposed for solar photography, with a built-in non-removable 20 stop ND filter.
Forgot to say that it also has IBIS, QPAF, 16 stops of DR, shoots 20fps 16-bit RAW, and records 24/30/60p uncropped oversampled 10-bit 4:2:2 4K without 30min limitation. But the ND filter is glued to the sensor and cannot be removed without damaging it.
 

amorse

EOS 7D MK II
Jan 26, 2017
487
488
www.flickr.com
Interesting but let me ask: I do landscape astrophotography in which case the land remains steady (unless there are leaves and there is wind) and the stars which move (OK earth rotates but you get the point). How do you apply the algorithm? Do you break the picture in two parts and process seperately?
I do the same type of photography - typically I'll have numerous frames for the sky, single very long exposure for the foreground with low ISO all taken without moving the camera or changing composition, and then blend them together - if that's all you're curious about then you probably don't need to read the rest - a lot of detail to follow if you're curious.
All the sky frames need to be properly aligned because of the planetary rotation, so I'll typically choose the first/middle/last sky exposure to align to as not not feel like I'm cheating in putting the milky way somewhere it wasn't or changing composition (personal guideline - otherwise I feel like it's not that different than a sky swap from another location or night, which I'm not into). Occasionally I'll put a lit tent into the frame, and in those cases I'll also need to add some exposures to manage the brightness of the tent, and light leak from the tent on to the foreground. I've included a couple examples below with what I did to get it done as examples. Admittedly, these are not for everyone's taste, but I do like them.

night glow 2 - web.jpg

5D IV, Rokinon 14mm f/2.8.
Sky - 8 frames, ISO 8000, 20 seconds. All frames manually aligned using warp (that distortion is nasty - this takes forever), then converted to a smart object and set to median blend in PS.
Landscape - 2 minutes (speaking from memory here) at around ISO 2000, brought back to darker luminance in photoshop
Tent - two frames with a light on - one to illuminate the ground around the tent appropriately, one to not blow out the luminance in the tent (I can't remember the settings of each for the life of me)

singing to the stars - web.jpg

This one didn't work as well because of the tree overlap with the sky, and a lot of haze in the sky requiring me to push ISO higher than usual. If you look closely near branches, you can see spots which couldn't median stack (because when you align the stars, the tree will appear to be moving) and had to be included with only noise reduction from a single exposure. I've included this example because I think comparing those less than ideal spots near the branches against the open sky gives you an idea of how much noise reduction and increases in detail you can get from an 8 image median stack relative to a single frame in one instance.

5DIV, Rokinon 14/2.8.
Sky - 8 frames, 20 seconds, ISO 10000 (honestly, increasing to 16 frames would have been a good idea here)
Foreground rocks - 1 frame, 2 minutes ISO 2000 (speaking from memory).
 
Sep 6, 2019
2
4
23919AD8-8681-4EF4-89F7-5F90CCC9F677.jpeg


Took this with the R and Sigma 20/1.4 Art at 1600 ISO/1.8 for 20 seconds. No trails (I don’t think?), and lots of star detail, a little distortion from the planets. Single shot, not a composite, just a little Lightroom post-processing to get the colors up (and dampen the glow from Phoenix).

I’m super amateur at Astro, but what would be different in the end result with something like the EOS Ra?
 
  • Like
Reactions: tron

blackcoffee17

EOS RP
Sep 17, 2014
228
188
I think it's kind of a stretch to assume that this is an astrophotography camera. It could be, but it seems like Canon would have a lot of other priorities before releasing such a body.
It's probably pretty easy and cheap for them to develop it. It's essentially an R with different filter on the sensor and some software tweaks.