The Canon EOS R6 Mark III is Canon’s Next Full-Frame Release

I’m worried. If the R5ii is 45 Mpix and the R6iii is gonna be >30Mpix, what is gonna be the differentiator? Now there’s no way that the R6iii would get a top down LCD :(
The R5II way more usable electronic shutter (also affects subject recognition, so better AF), 3x quicker readout (6ms vs 18ms) and less dynamic range loss vs mechanical shutter. And there are many other little cost saving measures in the R6 series.
But those who care about resolution the most will be happy to hear an R6III getting close to an R5II.
 
  • Like
Reactions: 1 user
Upvote 0
The R5II way more usable electronic shutter (also affects subject recognition, so better AF), 3x quicker readout (6ms vs 18ms) and less dynamic range loss vs mechanical shutter. And there are many other little cost saving measures in the R6 series.
But those who care about resolution the most will be happy to hear an R6III getting close to an R5II.
I agree with what you are saying, except the sensor in the R6III is likely to be in region of about 14ms readout speed. The other two factors are the resolution of the EVF and the buffer size.
 
  • Like
Reactions: 1 user
Upvote 0
I don't believe that there is extra pixel data in optically corrected lenses of the same size as the digitally corrected one. We may think that the "light" is more accurate that digital manipulation but the physics still means aberrations.
Can you point to any study/research to show that optical is better than digital correction?

Note that severe vignetting is an issue for single frame astro shots but much less so for panoramas. This is due to the corners being discarded when the panorama is generated - say with a 1/3 frame overlap. The side bits are for the panorama stitching control points but are discarded in the final image. The exception is the top left/right corner of the final merged image where this is not the case.
I do not have any study / research to prove that digital correction loses pixels (as I assume you don't have them to prove that they don't as well, since you haven't offered any).

But this is the way I look at it: it depends on the image circle the lens projects on the sensor. If the image circle of a digitally corrected lens covers the entire sensor then there is no a priori pixel loss, but the lens that requires the most geometric correction will result in a (slightly) less image quality since more pixels will be stretched.

If it doesn't then there is a further (small) loss of quality: an optically corrected lens may still require stretching, but the data used to create the corrected image is based on the full mp count of the sensor, while with digitally corrected lens whose image circle does not cover the full sensor, the stretching will be done using less data (less pixels), therefore more pixels are "created" with digitally corrected lenses.
 
Last edited:
  • Like
Reactions: 1 users
Upvote 0
I do not have any study / research to prove that digital correction loses pixels (as I assume you don't have them to prove that they don't as well, since you haven't offered any).

But this is the way I look at it: it depends on the image circle the lens projects on the sensor. If the image circle of a digitally corrected lens covers the entire sensor then there is no a priori pixel loss, but the lens that requires the most geometric correction will result in a (slightly) less image quality since more pixels will be stretched.

If it doesn't the there is a further (small) loss of quality: an optically corrected lens may still require stretching, but the data used to create the corrected image is based on the full mp count of the sensor, while with digitally corrected lens whose image circle does not cover the full sensor, the stretching will be done using less data (less pixels), therefore more pixels are "created" with digitally corrected lenses.
Good points. Corner sharpness should be reasonable either way and there are few use cases where tack sharp corners are mandatory or at least advantageous for lens selection eg astro.

I can only rely on that Canon has done the comparison and decided that the difference is not worth the extra physical lens component cost. They would be able to have prototypes for both options to compare whereas reviewers wouldn't.
Digital would be cheaper and keeps the lens lighter and smaller which are key features for new lenses.

If there was a significant difference then you could imagine that all the punters would be up in arms about it with youtubers blowing their collective stacks but that doesn't seem to be the case. It could also be that most OEMs are doing the same thing as it makes commercial sense with almost no downside. Having no ecosystem keyboard warriors arguing for their side also helps.
 
Upvote 0
We don't know that yet
It's Canon, they are careful not to make any hybrid "full-featured". It just won't get features that higher-end models like R5II already don't have.

I'd even say that based on R5 vs R6, it is not impossible that it will stick to 16:9 video to segment it further from a C50, would be 1.07x crop like an OG R6. Maybe no All-I codec either, but I do except C-Log 2 to be retained, probably the most important. Wouldn't bet on these things, just wishful thinking based on what they may remove for a lower price (despite adding shutter, EVF, IBIS, and maybe tilt/flip LCD).
 
Upvote 0
Good points. Corner sharpness should be reasonable either way and there are few use cases where tack sharp corners are mandatory or at least advantageous for lens selection eg astro.
Thanks! :)
Stretched corners can be tack sharp, it depends on the algorithms used to do the stretching... those are still not necessarily "real" pixels... but at the very least it's easier to do geometric corrections with more data (pixels) to work with.
I can only rely on that Canon has done the comparison and decided that the difference is not worth the extra physical lens component cost. They would be able to have prototypes for both options to compare whereas reviewers wouldn't.
Digital would be cheaper and keeps the lens lighter and smaller which are key features for new lenses.
I'd much rather they offered both options at specific focal lengths: one smaller, lighter, cheaper, digitally corrected lens and one bigger, heavier, more expensive, optically corrected lens. I'd love for them to do it for 35mm, like they have already done for 50mm and 85mm. I have serious envy for the ones that can use the new Sigma 35 1.2... :cry:
You say "key" features... they are not key for me.
If there was a significant difference then you could imagine that all the punters would be up in arms about it with youtubers blowing their collective stacks but that doesn't seem to be the case. It could also be that most OEMs are doing the same thing as it makes commercial sense with almost no downside.
We agree that there is a difference and we agree that the difference is small. Whether that difference is significant or not it depends on individual photographers.
But even if the difference was significant, why would the punters be up in arms about it at all? wouldn't price, size and weight (key features, aren't they?) more than make up for it? having multiple lenses at the same focal length at different price points with different characteristics is not a new thing.
Having no ecosystem keyboard warriors arguing for their side also helps.
? not sure I understand this last sentence? :unsure:
 
  • Like
Reactions: 1 user
Upvote 0
I'd much rather they offered both options at specific focal lengths: one smaller, lighter, cheaper, digitally corrected lens and one bigger, heavier, more expensive, optically corrected lens. I'd love for them to do it for 35mm, like they have already done for 50mm and 85mm. I have serious envy for the ones that can use the new Sigma 35 1.2... :cry:
You say "key" features... they are not key for me.
I guess that the Z (24-105 & 70-200) are the only ones that have duplicates within that RF ecosystem that are heavier, internal zoom and better quality at a cost. I am not sure if they are digitally corrected or not though.
There aren't other duplicates that I know of but RF versions of EF tend to have "better" features to tempt people to upgrade. Weight/size are some of them especially without the adaptor.
But even if the difference was significant, why would the punters be up in arms about it at all? wouldn't price, size and weight (key features, aren't they?) more than make up for it? having multiple lenses at the same focal length at different price points with different characteristics is not a new thing.
? not sure I understand this last sentence? :unsure:
We see some comments in this forum about users preferring optical correction without clear justification but feelings matter.
The ecosystem keyboard warriors are mostly the Sony fan boys stating that their system is better than everyone else. With Sony also making digitally corrected lenses then they don't have a platform to throw barbed comments at everyone especially the market leader.

The only real argument against Canon/FF lenses now is the lack of 3rd party options which is pointed out ad nausem. For new cost-conscious buyers of full frame systems then E mount may be the best for them but I suggest that most FF buyers are migrating from a APS-C system (as I did from 7D) and their experience, menus and layout plus reusing their existing lenses will be an advantage.
There will always be switchers who will tell everyone that they are changing but it is a big cost to do it.
 
Upvote 0
I guess that the Z (24-105 & 70-200) are the only ones that have duplicates within that RF ecosystem that are heavier, internal zoom and better quality at a cost. I am not sure if they are digitally corrected or not though.
The RF24-105 f2.8 Z lens requires digital corrections: “This is another lens that requires correction of barrel distortion. I did not check, but I'm sure it's 'forced' in camera. The corners are black at 24mm, however the distortion resolves pretty quickly since the black corners are gone by 28mm”.
The quote is from this post by @neuroanatomist.

Personally I prefer the small size of the RF 10-20mm f4 lens above the bulk of the EF 11-24mm f4 lens. The EF lens was optically corrected to fill the corners, but the image quality in the corners was not great. I’d say the quality of the ‘stretched’ corners of the RF lens are comparable to those of the EF lens.
 
  • Like
Reactions: 1 users
Upvote 0
Perhaps, but I wouldn't expect such a camera before early 2027, when the R8 will be four years old.
I don't expect the R8mkii before 2027 or later. 2027 would be fine with me. I want to add an R8 as a second lightweight and compact camera body. Since I just purchased the RF 50mm F1.4 VCM :love: (currently waiting for it) I am in no hurry and I could/ would be in the market for R8mkii in 2027 or later.

Based on past history, the refresh cycle for lower tier models, such as the R8, are shorter than for higher tier models such as the 1-Series and 5-series.

The R6 Mark II came a mere two years after the R6, though it has now been three years and still counting since the R6 Mark II debuted. Split the difference and that's a 30-32 month cycle if the R6 Mark III comes by the end of January, 2026. The R8 has been on the market for 30 months...
 
  • Like
Reactions: 1 user
Upvote 0
‘Wild’ wide angle prime….
Wild for canon or just in general?? They already have the 10-20/4 following their ef11-24/4 which are wild for zooms.

14/1.4 and 14/1.8 in different mounts and samyang’s veritable 14/2.8 and 14/2.4 have been available for some time.
Canon’s original EF14/2.8 in 1991 and updated in 2007 were expensive and clearly not a priority since. Their 16/2.8 is the token wide prime today.

‘Wild’ would mean either wider than 14mm and/or very fast. Either or especially both would equal very expensive

1991 to 2007 was 16 years. It's been 18 years since 2007, so maybe the RF 14mm L is about due? :)
 
  • Like
Reactions: 1 user
Upvote 0