DPReview just put up their sample gallery for the R5

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
Happy to be corrected but having processed a couple of their R6 files and looked at their R5 images are those guys capable of taking a correctly exposed properly lit and sharp photo? honestly they pretty much all look like crap to me, just don't believe that's all those cameras are capable of, very surprised
I am with you there... they are good and bagging Canon cameras though. I suppose, with their skills, the camera (tech) is the only excuse....
However R6 files look noticeably better to me at iso 6400 in comparison to R5. Makes me reconsider my options the only reservation I have is that a 20MP camera would be a poor resale value due to everyone is chasing megapickles these day...
Hard to explain to a general public that this 20Mp sensor beats hands down the mighty R5 at Iso 6400 and above..
 
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
Hard to explain to a general public that this 20Mp sensor beats hands down the mighty R5 at Iso 6400 and above..
Probably because it isn't true, unless Canon really messed something up about the R5 sensor. Using the D850 as a 45 MP placeholder here just to demonstrate that higher resolution does not equal losing 'hands down' to a lower Megapixel camera at ISOs above 6400:

 
Upvote 0

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
Probably because it isn't true, unless Canon really messed something up about the R5 sensor. Using the D850 as a 45 MP placeholder here just to demonstrate that higher resolution does not equal losing 'hands down' to a lower Megapixel camera at ISOs above 6400:

I am not implying 45 vs 20 performance here. This is specifically R5 at 6400 vs R6 at 6400. For an average Joe public the higher megapickles the better camera is. That’s how they have been profiled by various internet sources. :).
 
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
I am not implying 45 vs 20 performance here. This is specifically R5 at 6400 vs R6 at 6400. For an average Joe public the higher megapickles the better camera is. That’s how they have been profiled by various internet sources. :).
Well, without the R5 available in the same apples to apples comparison, I am just withholding judgement about it.

I don't expect the R6 to meaningfully beat the R5 in any aspect of image quality when compared properly. In any case I'm skeptical of any claims about one beating the other at this time, due to lack of coverage.
 
Upvote 0

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
Well, without the R5 available in the same apples to apples comparison, I am just withholding judgement about it.

I don't expect the R6 to meaningfully beat the R5 in any aspect of image quality when compared properly. In any case I'm skeptical of any claims about one beating the other at this time, due to lack of coverage.
Sure just going by the DPR library of R5 images.. see some of these taken at ISO6400. There is not much details left. It’s all looks mashed up... R6 images at ISO6400 however, are quite detailed. Totally expected as we know what 1Dx3 performance t iso 6400 is like.
 
Upvote 0

HenryL

EOS R3
CR Pro
Apr 1, 2020
359
983
Sure just going by the DPR library of R5 images.. see some of these taken at ISO6400. There is not much details left. It’s all looks mashed up... R6 images at ISO6400 however, are quite detailed. Totally expected as we know what 1Dx3 performance t iso 6400 is like.
FWIW, I don't put much stock in the DPR gallery images as indicative of actual output quality - at least not the OOC jpgs that we can view. I can't make heads nor tails out of the comet photos at 6400 for the R5, but the image of the bike hanging on a wall is not horrible even when viewed at 100%. Clearly their is evidence of bad noise reduction causing smearing/blotches, but even ACR would likely handle that just fine and we could end up with a better result.

The R6 gallery doesn't have any images at 6400, most seem to be ISO 100, but the yellow cat at 5000 also shows those same signs of bad noise reduction. There is, however, a good comparison in the R6 gallery with the small bird nest shot at 20000. There is a sample OOC image that again shows poor noise reduction, followed by the same image converted in an ACR beta that is much better. That sample is impressive.

All that said, I wouldn't be surprised that the R6 provides slightly better high iso performance. I don't expect it to be an earth shattering difference at all, certainly not enough to sway me towards the R6. For me personally - the relevant comparison for the R5 is against the 5D4 and 7D2, the camera's that I'm replacing. At first glance, it seems to have them beat. We'll see when I get it in my hands.
 
  • Like
Reactions: 1 user
Upvote 0

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
Yeah I am not talking noise. Look at the details level as in how much details still left at the ISO level in question... I’m looking at the RAW files. unprocessed. Sure noise can be dealt with. Details were already lost however.

p.s. to my eye,The R5 iso 6400 performance is slightly worse than the same of 5D4.. and likely around 1.5 stop better that the same of 7D2, about 1/3 of a stop or so worse than 1Dx3.
I am sure that Photons to photos will provide a reliable data very soon. At this stage it is all perceived and personal.
 
Last edited:
Upvote 0

koenkooi

CR Pro
Feb 25, 2015
3,656
4,235
The Netherlands
Yeah I am not talking noise. Look at the details level as in how much details still left at the ISO level in question... I’m looking at the RAW files. unprocessed. Sure noise can be dealt with. Details were already lost however.

How are you looking at the RAW files?
 
  • Like
Reactions: 1 user
Upvote 0
Jan 29, 2011
10,673
6,120
Sure just going by the DPR library of R5 images.. see some of these taken at ISO6400. There is not much details left. It’s all looks mashed up... R6 images at ISO6400 however, are quite detailed. Totally expected as we know what 1Dx3 performance t iso 6400 is like.
Are you looking at both at 100% or are you normalizing? Are the images you are comparing similar with similar detail and illumination?

At this point I believe the only thing the R5 is suffering from is mediocre sample image availability, no direct comparison images, limited RAW conversion options, and people not comparing like for like.
 
  • Like
Reactions: 2 users
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
Are you looking at both at 100% or are you normalizing? Are the images you are comparing similar with similar detail and illumination?

At this point I believe the only thing the R5 is suffering from is mediocre sample image availability, no direct comparison images, limited RAW conversion options, and people not comparing like for like.
Big thumbs up in that.

Considering how essential the magnification is to any discussion about noise and detail, I find it frustrating how little it is mentioned. Unless comparing side by side, screen size and resolution differences would also be worth noting.
 
Upvote 0

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
Yeah I am not talking noise. Look at the details level as in how much details still left at the ISO level in question... I’m looking at the RAW files. unprocessed. Sure noise can be dealt with. Details were already lost however.
Big thumbs up in that.

Considering how essential the magnification is to any discussion about noise and detail, I find it frustrating how little it is mentioned. Unless comparing side by side, screen size and resolution differences would also be worth noting.
looking at 75% screen resolution, going by the level of details I would expect from my trusty 5D4s.. all subjective. looking at the details level only. in general..
012A0815-20200723-235501-RawDigger-ScreenShot 75.png

p.s. screen resolution: 2560 x 1440, 32"
 
Upvote 0

HenryL

EOS R3
CR Pro
Apr 1, 2020
359
983
Yeah I am not talking noise. Look at the details level as in how much details still left at the ISO level in question... I’m looking at the RAW files. unprocessed. Sure noise can be dealt with. Details were already lost however.

p.s. to my eye,The R5 iso 6400 performance is slightly worse than the same of 5D4.. and likely around 1.5 stop better that the same of 7D2, about 1/3 of a stop or so worse than 1Dx3.
I am sure that Photons to photos will provide a reliable data very soon. At this stage it is all perceived and personal.
I can't open the raw file to compare, but I don't have any reason to doubt your observations. My point though was not necessarily comparing noise levels, but that the detail is likely/possibly being destroyed by the in-camera noise reduction. I should have been more clear.
 
  • Like
Reactions: 1 user
Upvote 0

Quirkz

CR Pro
Oct 30, 2014
297
221
the phobolgrapher posted some ISO 25k and 51k images of night shots through the city. I'm actually very impressed. Colours are excellent with very good detail on the well exposed parts of the image. (though, obviously, plenty noise in the darker parts as you'd expect, though the noise seems to be nicely random without any colour cast.)


As others have said, it's not the pixel by pixel comparison vs the r6 that's important, it's the normalised view.
I personally thought the images on DPreview looked quite fine, but I still really want to be able to view those raws.

I'm actually cautiously optimistic now that this might actually be an *upgrade* over my old 5d4, when normalised to same size.
 
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
The digital pictures tests show no high iso advantage for the r6 when both are compared at the r6 resolution:

No surprise there. I checked at Iso 25600 as well, the only difference I can see is the better contrast on the text in the R5 shots. And maybe the R6 noise is a little bit bit less colorful. Might also just be my tired eyes :LOL:

Thanks a lot for sharing!
 
Upvote 0
Jan 29, 2011
10,673
6,120
the phobolgrapher posted some ISO 25k and 51k images of night shots through the city. I'm actually very impressed. Colours are excellent with very good detail on the well exposed parts of the image. (though, obviously, plenty noise in the darker parts as you'd expect, though the noise seems to be nicely random without any colour cast.)


As others have said, it's not the pixel by pixel comparison vs the r6 that's important, it's the normalised view.
I personally thought the images on DPreview looked quite fine, but I still really want to be able to view those raws.

I'm actually cautiously optimistic now that this might actually be an *upgrade* over my old 5d4, when normalised to same size.
I see banding in all the high iso shots particularly in out of focus backgrounds, to the extent that the images are unusable. Considering these are camera processed jpegs it isn’t what I would want to be seeing.
 
Upvote 0