General Full-Frame R3 / R5 MK II differences in image quality

[BobU]

Hard to title this question / thread.

I have had an R3 4 years. Love it. Nothing I cannot do with it. Even if I want to take images that look just like what I see in terms of color, exposure, etc, though often add my own signature to the images in Lightroom.

I needed a second body so bought an R5 MK II. Today was really its first outing after setup and test images in my office. Resolution is great. Tack sharp. But a couple of qualities hard to describe. Let me say, EVERYTHING is set identical to the R3. Picture styles, white balances, etc.

First, the whites seem "chalky" to me for lack of grasping a better word! Not a different color temp. Not blown out, just a different quality of white. Second, I would have an incredibly difficult time accurately getting the colors I see in a final image. Greens are more intense. Unsure what else, but I just could not easily make it like a journalistic shot capturing what all eyes see. Its embellished. Now that could be a benefit in concert work, but not landscape or business environments or architecture.

So I am at a loss for where to look. I would like them to produce similar images. I realize will never be "identical" as two different sensors. But...

What do you folks think?

 

[BobU]

OK, just had a fabulous and long conversation with Claude about this issue. I can post the most pertinent parts of it but it will be VERY long. So asking if you would like that before I do.

 

[Dean Wilson]

I don't have the R3 (hint, hint) but I was wondering if you set the R3 on a tripod and took a perfectly exposed photo and then switch camera bodies with the R5 MK II (using the same lens).

Assuming all camera settings are identical

I would be curious if you edited one photo and saved the Preset from one and used it to the other...what the differenced would be. Especially under daytime lighting conditions (low ISO) and viewed at 100%. I know comparing 24MP and 45MP isn't "fair".

I remember a Fro Know Photo YouTube video comparing the cameras, but it has been awhile.

HOWEVER, I have 70+ YO eyes V1.0 and they have seen better days >snicker<.

 

[PKM-UK]

Hi Bob,

Could you post example jpgs (same subject, lighting etc…), or - better still - links to the raw files so people could download and have a look at them (e.g. via Dropbox)? This may help them to be able to suggest things you could consider trying.

Colour can be very difficult as we all see it differently, but - for example, you can take an image of a passport colour checker card and then create a custom profile with DxO PhotLab Elite. With this software you can also apply any of the DxO profiles to any camera - so I could apply an R3 (or Leica, Fuji-film etc..) profile to one of my R5 II raw files. (If you don’t have PL E but would like to try it, they offer free trials).

One other thought, LrC uses Adobe profiles. Perhaps try the files in Canon’s (admittedly clunky) DPP software?

(Finally, I’m not the best person to ask about colour as I like to fiddle around with our raw files - and prefer ‘poppy’ edits, but good luck on finding a solution!).



Phil

 

[Photofarmer]

interesting thread watching learning

 

[BobU]

Thanks folks. I'm working through this.

Just looked at the jpgs and better than expected in terms of "true to life" colors. I find the R5 has an "odd" (cannot find the word for it) white and a green push. To me, less natural than the R3. Reds might even be a little muted in comparison.

I have tried all Adobe profiles, all Canon profiles and adaptive color and all different and even though just starting points they reach different end results. Adobe Color, Adobe Standard, Canon Faithful, Canon Standard seem to be best for me, but also change with each image. Adaptive color can be interesting. I did post a few processed from today and some interesting results. But none would qualify for identical to what I saw which I think needs to be a known place before creativity gets added in to the development process.

I'll get through it

 

[BobU]

My conversation with Claude is PAGES but some interesting points mentioned:

The R5 II specifically has a reputation for more aggressive processing choices, not just color

One detailed DPReview comparison found the R5 II exposes images "more intelligently" and its auto white balance gives images extra "oomph" compared to older Canons even without touching color sliders — visible in things like blue shirts and green grass, and found it harder to improve on in post than to match the in-camera automatic result. That tracks with your "chalky whites" observation — it's very likely the R5 II's tone curve/highlight rendering handling near-white tones differently, not a color temp issue at all, which is why it doesn't feel like a WB problem to you.

Claude then talked in depth about color profiles and I reminded him that I shoot raw which is unaffected by color profiles in camera.

You're right, and I overstated that — let me correct it precisely.

Picture Style and RAW: Confirmed, Picture Style only bakes into the embedded JPEG preview/thumbnail and camera LCD/EVF rendering. When you open a RAW file in Lightroom, Adobe never reads that Picture Style tag. Lightroom builds its own conversion using Adobe-authored "Camera Matching" profiles that mimic each Picture Style name (Camera Standard, Camera Neutral, Camera Portrait, etc.), but Adobe creates these itself by shooting charts and reverse-engineering each camera's JPEG output — camera by camera, model by model. So even though you've set identical custom Picture Styles in both bodies, Lightroom is applying two separate Adobe-built profiles, one calibrated against the R3's sensor/output and one against the R5 II's, and there's no guarantee those two Adobe profiles behave the same, even under the same profile name. That mismatch is a well-documented sore point — Adobe's Camera Matching profiles are widely criticized as inconsistent from one camera model to the next, exactly the kind of gap that would produce your "chalky whites, hot greens" experience despite matched in-camera settings.

That's likely your actual root cause, not anything in the cameras themselves.

Where to look, given you shoot RAW:

1. In Develop, check what profile is actually assigned per image — go to Profile Browser and see if the R3 and R5 II files are landing on the same Adobe Camera Matching profile, or if one defaulted to Adobe Color/Adobe Standard while the other is on a Camera Matching profile. That mismatch alone could explain a lot.
2. Try forcing both to Adobe Neutral (not a Camera Matching profile) as a common, camera-agnostic baseline, and see if the gap narrows — that isolates whether it's the profile or something else (like the R5 II's tone curve/DR handling of highlights, which is a separate, sensor-generation issue).
3. If you want closer matching long-term without a ColorChecker session, a purchased profile set built specifically to normalize Canon bodies to each other (e.g., PerfeFilm's Canon-matching profiles) exists and is built for exactly this cross-body consistency problem — worth a look if the built-in profiles don't get you there.

These turn out to match FILM profiles only, and only 2 Canon EOS cameras, no R's.

So a good sampling of what AI can do with these kind of questions.

 

[PKM-UK]

Hi Bob,

One other thought, have you considered using Linear Profiles?

(I find them a bit of a pain, but they may be worth considering).

Have a look Here.

Phil

 

[BobU]

Thanks Phil. I had not heard of this before. Got some reading and playing around to do!

 

[ACEkin]

I wonder whether a Pantone or XRite gray card, as well as their ColorChecker Passport, may act as mediators. Some years ago, I wrote an article on creating camera and light profiles rather than using the built-in system components. In case anyone may want to give it a shot, here is the link:

Apexel 100mm Macro Lens Kit with a sturdy and flexible grip

A macro lens adaptor kit may give a new life to your phone camera. Apexel makes a nice kit that grabs the phone tightly with an adjustable mount

www.keptlight.com