The Canon EOS R3 will be 24mp, confirmed by EXIF data

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
LOL. I'm a fan of cogent, logical arguments even when I disagree with them, but I admit that I have a low tolerance for inane arguments that are clearly not well thought out. Ridiculous statements engender ridicule.

I don't think we know that they've chosen to use a 24 MP sensor in the R3. This 'conclusion' is based on the EXIF data from images from a single photographer using a Chrome plugin that is reporting the EXIF info on a camera that has not been released, and other EXIF viewers, e.g. the excellent exiftool, do not show that information on downloaded images.

But, assuming Canon is releasing an R3 with a 24 MP sensor, presumably it's because they believe that's what they perceive as the major market for it wants. The 1D X was 18 MP, and both the MkII and MkIII versions of that camera were 20 MP. If the R3 is targeting that market, 24 MP is very logical.
... and why exactly would the major market for the R3 demand a 24MP sensor, when a 45MP sensor could be fitted?

It's not at all "logical" to assume that because previous generation models were limited by technology to lower resolution, that the R3 should be limited (if it is) to 24MP. Using *your* logic, that would mean we should never have advanced beyond 5MP, or welcomed any other advances such as AF! :ROFLMAO:

There is no noise benefit or DR benefit to having a lower resolution sensor, and no reason why the R3 processor can't cope with 30fps, given that the R5 can cope with 45MP RAWs at long 20fps bursts.

... and in any case, if a 45MP sensor was fitted, users could select a lower res if and when they required it.

Only the ridiculous are tempted to ridicule. Cogent logical replies are welcome, condescending ones are indicative of general intolerance and an unjustified superiority complex, which comes across in almost all of your replies, not just to my "inane argument".
 
Upvote 0

unfocused

Photos/Photo Book Reviews: www.thecuriouseye.com
Jul 20, 2010
7,184
5,483
70
Springfield, IL
www.thecuriouseye.com
...Is your photography for your own use, rather than for instant publication on the media?
I shoot for a small college. My photos are posted to social media, added to a website gallery, submitted to the local media and saved for use in college publications. I am the "picture desk." No staff, so everything is my responsibility. After a game, I cull through the photos and process the best ones. I generally do that the night of the game. The next morning I post them to social media and add them to the athletics website.

I wait until the next morning because a social media or website post at 10 p,m. or midnight is not going to get the same viewership as one posted mid-morning. I also upload them to Dropbox by morning so the sports information staffer can share that link with the local media. These days "local media" in a small town means one newspaper with one editor and a lag time of at least a day, plus a couple of online publications which post things as soon as they get them. Later I'll add them to the college's server so the marketing staff has them for publications. For a typical game I'll process 10-40 images out of the thousands of frames I will have shot.

I can process a raw file faster than messing with jpgs because that's what I'm used to. I can see how those in big organizations with support staffs might do things differently, but remember, they represent just a small fraction of working photographers.
 
  • Like
Reactions: 1 users
Upvote 0
Jul 21, 2010
31,099
12,863
... and why exactly would the major market for the R3 demand a 24MP sensor, when a 45MP sensor could be fitted?

It's not at all "logical" to assume that because previous generation models were limited by technology to lower resolution, that the R3 should be limited (if it is) to 24MP. Using *your* logic, that would mean we should never have advanced beyond 5MP, or welcomed any other advances such as AF!
Why was the 1D X III 20 MP, then? It uses a Digic X just like the R5, so clearly faster data throughput was possible. You still haven’t addressed this basic question. I suspect that’s because you realize the 1D X III was not limited to 20 MP for technological reasons, but admitting that is tantamount to admitting that your logic has more holes than Swiss cheese.

And again you’re attributing statements to me that I’m not making. Where did I suggest technology shouldn’t advance? It does. It certainly did between the 1D X II, which had dual Digic 6+, and the 1D X III, which has Digic X. The MkII was 14/16 fps (might be the max for dual Digic 6+), the MkIII is 16/20 fps and Digic X runs 45 MP @ 20 fps in the R5, so for data throughput the MkIII could obviously have been 45 MP.


Your logic is that the R3 has lower MP because Canon is limiting it as a marketing tactic to drive (future) R1 sales.

So let me ask you again – why was the 1D X III was limited to 20 MP? If not as a marketing tactic to drive sales of _______ (still avoiding that original question, aren’t you), then why?

A very logical answer is that the market for the 1-series would be satisfied or even would prefer a 20 MP sensor. Canon knew that, because they conduct extensive market research.

Thus, it’s equally logical that they’ve done the same for the R3, and determined that (the presumed) 24 MP is appropriate for the intended market.


There is no noise benefit or DR benefit to having a lower resolution sensor, and no reason why the R3 processor can't cope with 30fps, given that the R5 can cope with 45MP RAWs at long 20fps bursts.

... and in any case, if a 45MP sensor was fitted, users could select a lower res if and when they required it.
All true. All apply to the 1D X III, which Canon fitted with a 20 MP sensor. Dare I ask again…why?
 
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
Why was the 1D X III 20 MP, then? It uses a Digic X just like the R5, so clearly faster data throughput was possible. You still haven’t addressed this basic question. I suspect that’s because you realize the 1D X III was not limited to 20 MP for technological reasons, but admitting that is tantamount to admitting that your logic has more holes than Swiss cheese.

And again you’re attributing statements to me that I’m not making. Where did I suggest technology shouldn’t advance? It does. It certainly did between the 1D X II, which had dual Digic 6+, and the 1D X III, which has Digic X. The MkII was 14/16 fps (might be the max for dual Digic 6+), the MkIII is 16/20 fps and Digic X runs 45 MP @ 20 fps in the R5, so for data throughput the MkIII could obviously have been 45 MP.


Your logic is that the R3 has lower MP because Canon is limiting it as a marketing tactic to drive (future) R1 sales.

So let me ask you again – why was the 1D X III was limited to 20 MP? If not as a marketing tactic to drive sales of _______ (still avoiding that original question, aren’t you), then why?

A very logical answer is that the market for the 1-series would be satisfied or even would prefer a 20 MP sensor. Canon knew that, because they conduct extensive market research.

Thus, it’s equally logical that they’ve done the same for the R3, and determined that (the presumed) 24 MP is appropriate for the intended market.



All true. All apply to the 1D X III, which Canon fitted with a 20 MP sensor. Dare I ask again…why?
I'm still waiting for you to provide a believable theory why Canon hasn't put a 45MP sensor in the R3. To say or imply that it's just because 1Dxiii users are happy with 20MP and don't want more, is just plain daft, as they'd have the option of using a lower res than 45MP when they needed it, just by selecting it in the menu. And at the same time they'd have a much more flexible camera that provided far greater cropping options - very important to many users, particularly as the extra cropping ability allows them to use shorter, lighter and cheaper lenses to achieve the same end result.

There simply is no advantage to having a low res sensor if comparing same-generation sensors of the same size. DR is identical, and noise levels are *lower* with a high res sensor if the output is printed at the same size. I don't believe that the R3 would have any problems maintaining long 30fps bursts with 45MP either. And it's unlikely that a 45MP sensor costs much more to produce than a 24MP sensor - certainly a bit more due to lower yield rates, but as a percentage of the cost of what is likely to be a $5500 camera, the difference in cost would be pretty insignificant.

So I'll ask you again, exactly what advantage is there for a user to be limited to 24MP? Other than unfounded and uninformed fear about differences in noise levels - https://www.dpreview.com/videos/794...esolution-sensors-are-not-better-in-low-light

And answer me honestly, if you had the choice between 2 otherwise identical cameras, one with 45MP (with an option to select 24MP), and the other limited to a maximum of 20 or 24MP, which would you choose?

There may well be an alternative explanation that neither you or I have thought of, but Canon more than any other manufacture is renowned for using product segmentation and so-called "crippling" to protect sales of other cameras in its range (including models under development).

Perhaps also the reason why they didn't put 45MP in the 1Dxiii was simply because they knew that the 1Dxiii was an interim camera and that the truth is that they would much rather steer people to the RF system and sell a bucketload of new lenses.

They would also be very aware that Sony and Nikon were working on new top of range pro models, and are putting out the R3 with a deliberately lower spec (and probably more affordable price) while they continue to develop the R1.
 
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
B
How are noise levels lower? I'd agree to a statement that they were identical--especially now with the electronics on the back, meaning the entire front is gathering photons. With the electronics on the sensor front, as its been until now, a fixed amount per pixel cannot gather photons, so as you quadruple resolution you quadruple the non-photon-gathering portion of the sensor. In contrast, with the electronics on the back, a 96MP sensor down-sampled 4:1 to 24MP should make literally an identical image to a native 24MP in this case... no worse for sure... but also, I'd have guessed, no better??
Because if the print size is the same for both, the individual noise elements from the high MP camera would be smaller, due to lower magnification. Effectively you'd be looking at e.g. 2 small noise "specks" rather than one big and more conspicuous one.
 
Upvote 0
D

Deleted member 381342

Guest
I don't think so, for telephotos.

Basically, if you come up with a natural lens design for a long lens, it won't typically have elements especially near the sensor. In contrast as angles get wider, the natural place to put the rear elements gets closer and closer to the sensor. Some wide angles for Leicas and so on were mere millimeters from the film.

As a rule of thumb, if the rear element is flush with the back of an EF lens, then maybe just coincidentally that was the natural place to put it, but more likely they had to constrain the design to keep the lens that far from the sensor. The EF85/1.2L was spilling out the back of the mount but I don't recall any longer lens being like that.

However, another problem is the location of the teleconverter. If the TC is a lot closer to the sensor--and it is--then the rest of the lens design has to allow for that TC position. So I imagine that even though the last EF generation were seriously good lenses, I bet the RF versions have substantially different formulas just to account for the moved TC, even if the specs aren't improved any.

This is all very well known. But the design of a lens is more than just where the elements are placed and how close to the sensor they are. I am talking more about button placement, control rings, build in TC's (of note the Nikon 400mm f/2.8 is rumoured to have a built in 1.4x TC), and other value add. At the very least a lens build natively for RF hopefully won't have a naff looking overly long silver mount. Once some new design RF super tele lenses come out I think it'll be much more clear what to expect from the 400 and 600 mk2.
 
  • Like
Reactions: 1 user
Upvote 0
Jan 29, 2011
10,675
6,121
B

Because if the print size is the same for both, the individual noise elements from the high MP camera would be smaller, due to lower magnification. Effectively you'd be looking at e.g. 2 small noise "specks" rather than one big and more conspicuous one.
If you have twice as many noise pixels but they are half the size each the result is the same amount of noise. It will be just as visible as less’ larger‘ noise.
 
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
This is all very well known. But the design of a lens is more than just where the elements are placed and how close to the sensor they are. I am talking more about button placement, control rings, build in TC's (of note the Nikon 400mm f/2.8 is rumoured to have a built in 1.4x TC), and other value add. At the very least a lens build natively for RF hopefully won't have a naff looking overly long silver mount. Once some new design RF super tele lenses come out I think it'll be much more clear what to expect from the 400 and 600 mk2.
Yes control ring design etc is very important. My EF 24-105mm was recently stolen so I've just replaced it with the RF 24-105mm F4. In some ways it's a much better lens - sharper, smaller, MUCH faster AF. But in other ways it's inferior to the old EF model - e.g. the zoom ring is much stiffer, and the close spacing and identical rubbers of the zoom and focus rings mean that I often grab the focus ring when I want the zoom ring, and vice versa. Hopefully the muscle memory will adapt soon...
 
  • Like
Reactions: 1 users
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
If you have twice as many noise pixels but they are half the size each the result is the same amount of noise. It will be just as visible as less’ larger‘ noise.
I disagree - the total "volume" of noise will be the same in both cases, but beyond a certain threshold (dependent on viewing distance) the half-size noise elements will effectively disappear, but the larger (i.e. more highly magnified) noise elements will still be conspicuous. This was actually demonstrated on dpreview a couple of days ago - https://www.dpreview.com/videos/794...tm_medium=marquee&utm_campaign=traffic_source
 
  • Like
Reactions: 1 user
Upvote 0
Jan 29, 2011
10,675
6,121
I disagree - the total "volume" of noise will be the same in both cases, but beyond a certain threshold (dependent on viewing distance) the half-size noise elements will effectively disappear, but the larger (i.e. more highly magnified) noise elements will still be conspicuous. This was actually demonstrated on dpreview a couple of days ago - https://www.dpreview.com/videos/794...tm_medium=marquee&utm_campaign=traffic_source
Not in my experience, when images are optimally post processed, that is not processed the same but as well as each can be done, the noise is constant for same generation same sized sensors irrespective of pixel count.

I have been saying this since the 7D was released and I compared it to my 1DS III, and even DPReview is now catching up!
 
Upvote 0
Jul 21, 2010
31,099
12,863
I'm still waiting for you to provide a believable theory why Canon hasn't put a 45MP sensor in the R3. To say or imply that it's just because 1Dxiii users are happy with 20MP and don't want more, is just plain daft, as they'd have the option of using a lower res than 45MP when they needed it, just by selecting it in the menu. And at the same time they'd have a much more flexible camera that provided far greater cropping options - very important to many users, particularly as the extra cropping ability allows them to use shorter, lighter and cheaper lenses to achieve the same end result.
Not sure why you cannot grasp this, it's not complicated. Canon released the 1D X III with a 20 MP sensor, even though it could have had more MP. If the R3 is announced at 24 MP as the topic of this thread suggests, then Canon will have released it with a 24 MP sensor, even though it could have had more MP.

So, either Canon is daft and releasing cameras with lower-than-possible MP sensors for no conceivable reason, or they have a reason and you cannot understand it, cannot accept it, or both.

The onus is not on me to provide a theory as to why. You can choose to believe they're daft and have no reason. But if you think that large, successful companies are capricious like that, it's you who's daft.

They made a choice with the 1D X III, they may have made an analogous choice with the R3. I'm certain they have good, logical reasons. Since we seem to agree that those reasons are not technical limitations, they are most likely market-driven, as I suggested some time back.

You suggest the idea that 20-24 MP is sufficient for many people is daft, but honestly, how much real market research have you conducted? Sent out surveys to a few thousand CPS members around the world? Canon does that regularly. Tracked the camera and lens buying habits of individuals and organizations via product registrations over decades? Canon does that. Assessed what gear photographers at sporting events around the world use? Canon does that. Logged the equipment that professional photographers of all genres evaluate on loan, compared to the equipment they actually buy? Canon does that.

So please, tell me what market research you've sponsored or personally conducted to provide data on how many MP people would find sufficient. Read a few posts on the internet? You know a few people with cameras? LOL.


There simply is no advantage to having a low res sensor if comparing same-generation sensors of the same size. DR is identical, and noise levels are *lower* with a high res sensor if the output is printed at the same size. I don't believe that the R3 would have any problems maintaining long 30fps bursts with 45MP either. And it's unlikely that a 45MP sensor costs much more to produce than a 24MP sensor - certainly a bit more due to lower yield rates, but as a percentage of the cost of what is likely to be a $5500 camera, the difference in cost would be pretty insignificant.

So I'll ask you again, exactly what advantage is there for a user to be limited to 24MP? Other than unfounded and uninformed fear about differences in noise levels - https://www.dpreview.com/videos/794...esolution-sensors-are-not-better-in-low-light
There isn't any. So what? My point is that Canon is making what they believe to be rational decisions on these matters, with real data to drive those decisions. Data you clearly don't have.

And answer me honestly, if you had the choice between 2 otherwise identical cameras, one with 45MP (with an option to select 24MP), and the other limited to a maximum of 20 or 24MP, which would you choose?
I have to choose between an R5 with 45 MP and (presumably) an R3 with 24 MP. More MP, or better performance and better (for me) ergonomics. Would I choose a 45 MP R3 over a 24 MP R3? Yes, I would. Do I have that choice? No, I don't. Is 24 MP sufficient for me? Yes, which is why I'll be buying the R3.

There may well be an alternative explanation that neither you or I have thought of, but Canon more than any other manufacture is renowned for using product segmentation and so-called "crippling" to protect sales of other cameras in its range (including models under development).
Canon's so-called 'crippling' of cameras really boils down to Canon not making the specific camera with the specific features that internet complainers and forum dwellers claim they would buy (when in reality, most would not).

Actually, I'm pretty sure that's what this whole discussion is really about – you want an R3 with 45 MP, and Canon is apparently not giving you one, so you're repetitively and petulantly demanding an explanation.
 
  • Like
Reactions: 1 user
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
The point is that Canon is letting Sony and Fuji walk away with a market that Canon created and once dominated. And it's not because the others outmaneuvered them, Canon simply gave up on it.

I don't believe I said anything about the R3 specifically in this thread, but you're right, at this point it's not a package that appeals to me. What is interesting is that 24MP, if true, indicates that Canon isn't being drawn into another pitched megapixel war.

Maybe Canon isn't walking away from the low end market so much as almost the entire low end market is walking away from all the major camera manufacturers?
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
Note the single quotation marks around old. However, they are clearly not RF native designs. I don’t believe like some that making all new designs so soon after release is necessary nor required, but I also expect designing for RF from the get go will produce lenses that better match the RF design.

How, exactly, are the "clearly not RF native designs?"
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
Canon has said that these lenses were not designed for RF. They have also said that they released the RF versions of these lenses by popular demand, but that this was not the original plan. Presumably that means that more telephotos that were planned are still being actively worked on.
There was a thread covering this on Canon Rumors, and here is the source:

Thanks, I hadn't seen that as I was busy with a large shooting assignment that week.

Based on the DP Review writer's response, I don't think it was unreasonable to ask if those lenses had been developed with the RF system in mind.

"It was surprising to learn from Mr. Tokura that the EF 400 F2.8 and 600mm F4 professional telephoto lenses were not developed with future native RF support in mind originally, although it does at least explain the absence of a control ring from the new RF-mount versions."
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
It's often stated that among the prime requirements of sports and reportage photographers, is the ability to have small file sizes that can be rapidly transmitted straight to the picture desk - and that means JPEGs (or possibly low-res CR3 RAWs). That's what I hear and read regularly, so I'm guessing it's true.

I think you must be one of the few who shoot RAW for sports (I'm more than happy to be corrected :) ).

If you do your own processing in ACR, it implies that your work is non-urgent....
Is your photography for your own use, rather than for instant publication on the media?

Not all sports and reportage shooters have "instant" deadlines. The guys whose clients/employers want it five minutes before it happened are under increasing pressure to get images out as quickly as possible. But as data transmission rates have increased dramatically, more news/sports shooters are using raw in certain situations where it increases the odds of a better result.

In well lit major sporting venues, though, one who knows what they are doing can get JPEGs straight from camera that are so close to what one could do spending a lot more time processing raw files that it isn't worth the extra processing time. The lighting in major college and pro stadiums is like being inside a giant light box. It's bright, full spectrum, and no flickering. It's in times and places where the lighting is not so great that the benefits of raw make a difference. Crappy high school stadiums and gyms, for instance. The dimmer, more limited spectrum flickering light one finds in those places needs more raw color correction than bright, full spectrum, non-flickering light that most major sports venues now use.

Some news shooters have actually seen things go the other way in terms of deadlines. My hometown newspaper no longer publishes a print edition on Saturday. As a result, their staff photographer no longer has to file his photos of Friday night games by 10:30 p.m. to make the printing deadline. As long as his images are up by early Saturday morning when subscribers wake up and look online to see photos from Friday night's game, everyone is happy. Back when they still had two staffers (at the same paper only twelve or fifteen years ago they had four full-time staff photogs and a photo editor that rarely had to cover a shooting assignment), the newer, younger guy shot raw and processed in his car on his laptop during halftime and after the game. The older guy kept shooting JPEG. The newer guy's photos tended to have better color when his images were viewed online, but it made very little different in newsprint, which is incredibly lo-res and color limited. The newer guy is now "the photo staff" at a sister publication in a town fifty miles away since the last old-timer there retired a couple of years ago. The other guy is still in my hometown and still shooting JPEG most of the time, as far as I know. He has always used raw for feature photo illustrations and the like sometimes, too.

I used to shoot sports with JPEG due to handling speed issues. The buffers of the cameras I was using couldn't keep up when saving raw files. But as the capabilities of cameras improved, I moved to raw a long time ago. I don't have a hard deadline 20 minutes after the game is over. I'm not looking for a handful of images to publish in a newspaper or online and then forget it and move on to the next of several assignments the following day. I'm looking to get a lot more images that feature as many of the different participants from the team I'm covering as possible and put them in a place where parents, grandparents, aunts and uncles, etc. can buy prints or digital downloads of "their" player. Sometimes I see as many sales of high school football and marching band images in the spring near graduation time as I do during the fall, especially for seniors. Their parent are using them to create displays for graduation parties and the like.
 
  • Like
Reactions: 2 users
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
Keep in mind that you are comparing the R3 to a 10 year old camera. That doesn't make sense. Certainly a 1DX is a great camera but technology doesn't stop and if that's what you want buy a 1DX.

The nine years old 1D X was 18 MP. The 20MP 1D X Mark III was introduced in March of 2020, barely over one year ago.
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
I'd buy a new 1992 Integra, but Acura rather small-mindedly doesn't make them any more.

Cars aren't improving nearly as rapidly as cameras (and in many respects are un-improving).

That aside the point you were making is a good one with respect to cameras.

Honda/Acura is brining back the Integra, reportedly in 2022.
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
It is well documented and obvious from their design. We'll see what native RF super-tele's look like with the 300, 500, or maybe even the 200-400.

You still haven't said what, exactly, has been "well documented.

What specific design features make it obvious?

Lens design advances with time. The II and III versions of many Great Whites were better than the older, original versions. Saying a native Super Telephoto RF lens created in the future is sharper because it is RF, rather than because it is a seven years newer design, proves nothing about the difference in mount of the older lens.
 
  • Like
Reactions: 1 user
Upvote 0