Gordan Laing Review: The Canon EOS R5 for photography

Canon Rumors Guy

Canon EOS 40D
CR Pro
Jul 20, 2010
10,779
3,158
Canada
www.canonrumors.com
Gordon Laing from Camera Labs has completed his review of the Canon EOS R5. Instead of reviewing the video performance of the camera, his review focuses on stills photography. The Canon EOS R5 looks to be a great one.
From the autofocus to the competitive dynamic range, the Canon EOS R5 will likely find its way into a lot of photographers’ bags.

Continue reading...


 
  • Like
Reactions: 2 users
Here's a chart showing the dynamic range compared to other cameras. The R5 is excellent. https://www.photonstophotos.net/Charts/PDR.htm
Yeah it beats even A7rIV, there was a discussion in another thread. There's a caveat though: the low-ISO metrics are coming from images with noise reduction applied. It looks like the R5 does some noise reduction on raw files in camera.
 
  • Like
Reactions: 2 users
Upvote 0
That's a decent review, although I was surprised with noise comparison made on jpegs. It's basically useless.
Not really IMHO. I'm not interested in the amount of noise, I'm interested what I can get out of low light situations. And while actually seeing less noise in a comparison might be satisfying in some way, knowing that I can have an ISO 12.800 picture processed with no noise visible benefits me more.

Why there is a jump around 400ISO?
I remember hearing that the algorithm changes at ISO 400

...looks like the R5 does some noise reduction on raw files in camera.
Ehm, it's not RAW any longer then.
 
  • Like
Reactions: 3 users
Upvote 0
Not really IMHO. I'm not interested in the amount of noise, I'm interested what I can get out of low light situations. And while actually seeing less noise in a comparison might be satisfying in some way, knowing that I can have an ISO 12.800 picture processed with no noise visible benefits me more.


I remember hearing that the algorithm changes at ISO 400

You can apply arbitrary amount of NR on raw files depending on your taste and target image dimensions, so the noise comparison is most valuable if done on raw files before any noise reduction is applied.
Out of camera jpegs have some unknown amount of NR applied in camera, so noise comparison on jpegs actually shows how nice NR algorithms are in different cameras.
 
  • Like
Reactions: 1 users
Upvote 0
Why there is a jump around 400ISO?
They think Canon used some kind of dual gain in this sensor and ISO 400 is where it kicks in.

I don't have the camera yet but I downloaded some raw samples and they really look fantastic in the shadows even at ISO 400-800, compared to 5DIV.
 
  • Like
Reactions: 4 users
Upvote 0
Ehm, it's not RAW any longer then.

It may (and most likely is) very moderate amount of NR. Interestingly it's only applied up ISO 640.

Sony also tamper with the raw files (see star eater effect).

Yes, raw sometimes isn't completely raw but slightly cooked.
 
  • Like
Reactions: 6 users
Upvote 0
For those who were talking about R5's Dynamic Range...
Now add the specs (top AF, IBIS, 12/20fps, 8k Raw, 4k HQ), RF Glass and the colors Canon provides and then come and tell me that Canon has not the the best hybrid system out there!! Expect a few firmware updates that will extend record times and reduce cooldowns. Im telling this mostly to myself who cancelled the preorder and now I have to wait for sometime...
 

Attachments

  • R5 DR.png
    R5 DR.png
    182 KB · Views: 766
Last edited:
  • Like
Reactions: 4 users
Upvote 0

Sharlin

CR Pro
Dec 26, 2015
1,415
1,433
Turku, Finland
that makes zero sense, a raw file is not completely raw? sure, ok

It's as raw as you can get. Whatever processing the camera may or may not do is a black box to the user. Complaining that CR3 is not "raw" is just semantic masturbation as long as it's not possible to make the camera output anything more raw than what we already get.
 
  • Like
Reactions: 3 users
Upvote 0

LDS

Sep 14, 2012
1,763
293
Ehm, it's not RAW any longer then.

Depends. If the camera can actually have a measure of actual sensor noise and subtract it from the data the image is still RAW, in many ways a better RAW.

Anything not based on real noise measurement but on algorithms trying to detect what is noise and what is not is a different thing.
 
  • Like
Reactions: 3 users
Upvote 0