It looks like 2021 will be the last year for the EOS M lineup [CR2]

[hachu21]

Well, on EOS M and first EOS R there is no in camera stabilisation for photography, only for video. So I guess that's the subject here.

Following this technical post, the 2 additional pins seems related to lens identification only (rather strangely I would say).
The EF lens mount
Maybe because of the collapsible design with the in-lens barrel swich (11-22 and 15-45)? The body display an error if you turn the camera on with those lenses retracted.

 

[Bob Howland]

M7 is bound to be much smaller and thus less capable than R7.

Why? The RP, M5 and SL1/2/3 are all about the same size and much smaller than the 5-series and 7-series EF bodies and slightly smaller than the R5 and R6 bodies. A lot of the "capability" is in firmware. which is relatively/completely independent of camera size. It basically all marketing decisions.

 

[EOS 4 Life]

Why? The RP, M5 and SL1/2/3 are all about the same size and much smaller than the 5-series and 7-series EF bodies and slightly smaller than the R5 and R6 bodies. It basically all marketing decisions. A lot of the "capability" is in firmware. which is relatively/completely independent of camera size.

R7 is supposed to be in the R6 body.
It is supposed to be a mirrorless 7D and not meant to be small.

 

[gektor]

The EOS M system has never been intended for those who would even think of wanting an option to upgrade. That's never been who the EOS M system was aimed at by Canon.

I started using the M system when the very first EOS-M was still up-to-date. It was a great backup camera for my DSLRs. Now I've somehow switched completely to the EOS-M, for bodies and lenses, not knowing that it's not for enthusiasts. I think, the users like me are those who are complaining about the dead end situation. M system is very capable and it attracted many enthusiasts.

M is the first and exciting mirrorless system by Canon, it supports EF and EF-S and was a perfect transition from canon DLSRs. Nobody thought that canon would ever fall into the trap of developing a "dead system".

 

[EOS 4 Life]

I started using the M system when the very first EOS-M was still up-to-date. It was a great backup camera for my DSLRs. Now I've somehow switched completely to the EOS-M, for bodies and lenses, not knowing that it's not for enthusiasts. I think, the users like me are those who are complaining about the dead end situation. M system is very capable and it attracted many enthusiasts.

You were not wrong to complain but Canon's focus was to make a camera system that was small, light and priced modestly.
Whether are not it is a dead-end remains to be seen.

 

[gektor]

Canon and all other camera manufacturers need to think thoroughly about the future of digital photography, regardless of the system. The power of computational photography is just beginning to shine. Smartphone sensors are still quite small, but they are starting to get slightly bigger. Eventually, AI and image processing will produce perfect full-frame-sensor-like images from small lenses and small sensors. Without computational photography and AI, camera manufacturers will lose more and more markets, including professionals. Who needs a professional camera when wedding guest images straight from a smartphone look better than a professional's manually developed RAW images (not to speak of out of camera JPGs)? It's getting harder and harder to justify the cost of a professional or enthusiast camera for most users. Canon needs to innovate as much as possible, especially in the area of computational photography. All of the Canon, Nikon, and Sony cameras are still pretty "dumb" by today's computational photography standards. How will they ever catch up? And if they don't, they will die out. Just as film photography lost to digital photography, classic digital photography will lose to computational photography.

So essentially, all Canon systems as they are today are a dead end.

 

[Bob Howland]

Canon and all other camera manufacturers need to think thoroughly about the future of digital photography, regardless of the system. The power of computational photography is just beginning to shine. Smartphone sensors are still quite small, but they are starting to get slightly bigger. Eventually, AI and image processing will produce perfect full-frame-sensor-like images from small lenses and small sensors. Without computational photography and AI, camera manufacturers will lose more and more markets, including professionals. Who needs a professional camera when wedding guest images straight from a smartphone look better than a professional's manually developed RAW images (not to speak of out of camera JPGs)? It's getting harder and harder to justify the cost of a professional or enthusiast camera for most users. Canon needs to innovate as much as possible, especially in the area of computational photography. All of the Canon, Nikon, and Sony cameras are still pretty "dumb" by today's computational photography standards. How will they ever catch up? And if they don't, they will die out. Just as film photography lost to digital photography, classic digital photography will lose to computational photography.

So essentially, all Canon systems as they are today are a dead end.

First, film photography lost to digital photography because it used film. The difference is qualitative, not quantitative. The last three years I used film, I digitized all my negatives. The results were 11mp digital images which were good at the time but much worse than the 12.8 MP native images from my 5D and absolutely pathetic compared with my 5Ds images.

Second, if computational photography can perform miracles to small lens and small sensor images, imagine what it can do with premium lens and large sensor images.

Third, when I hear "computational photography", I think the eye detection algorithms in the R5 and R6, not changes to the images themselves.

 

[SnowMiku]

Canon and all other camera manufacturers need to think thoroughly about the future of digital photography, regardless of the system. The power of computational photography is just beginning to shine. Smartphone sensors are still quite small, but they are starting to get slightly bigger. Eventually, AI and image processing will produce perfect full-frame-sensor-like images from small lenses and small sensors. Without computational photography and AI, camera manufacturers will lose more and more markets, including professionals. Who needs a professional camera when wedding guest images straight from a smartphone look better than a professional's manually developed RAW images (not to speak of out of camera JPGs)? It's getting harder and harder to justify the cost of a professional or enthusiast camera for most users. Canon needs to innovate as much as possible, especially in the area of computational photography. All of the Canon, Nikon, and Sony cameras are still pretty "dumb" by today's computational photography standards. How will they ever catch up? And if they don't, they will die out. Just as film photography lost to digital photography, classic digital photography will lose to computational photography.

So essentially, all Canon systems as they are today are a dead end.

I've been thinking the same things, the only thing the Smartphone manufacturers haven't worked out yet is good telephoto, but I can imagine they will work this out with computational photography that takes up no extra space and goes equivalent 300mm plus and will look as good as a camera and lens of today. The only thing the smartphone will lose out on is ergonomics and dedicated buttons for everything. I can only see this replacing entry level and enthusiast cameras in the future but I think the professional level cameras and lenses will stay for a very long time.

 

[Stu_bert]

I've been thinking the same things, the only thing the Smartphone manufacturers haven't worked out yet is good telephoto, but I can imagine they will work this out with computational photography that takes up no extra space and goes equivalent 300mm plus and will look as good as a camera and lens of today. The only thing the smartphone will lose out on is ergonomics and dedicated buttons for everything. I can only see this replacing entry level and enthusiast cameras in the future but I think the professional level cameras and lenses will stay for a very long time.

Cheaper for them to put zoom optics inside as many have done...

 

[SnowMiku]

Cheaper for them to put zoom optics inside as many have done...

As far as I know they can go to around equivalent 50mm or can they go further these days? I don't keep up to date with smartphones so I have no idea haha.

 

[koenkooi]

As far as I know they can go to around equivalent 50mm or can they go further these days? I don't keep up to date with smartphones so I have no idea haha.

Some phones have a 'periscope' style design, where the periscope is layed flat inside the phone.A quick google suggests that 135mm equiv. has been announced by at least one vendor.

 

[Stu_bert]

As far as I know they can go to around equivalent 50mm or can they go further these days? I don't keep up to date with smartphones so I have no idea haha.

Yep as Koenkooi said, between 5x and 6x are available, starting around 24mm equivalent and going up to 120-135mm equivalent. Typically these are alongside a wide angle Lens. Couple that with fast (f1.4 and maybe some faster), and combined with the computational side, they have probably captured the rest of the compact camera specifications, as well as exceeding them in ease of use.

Having developed the tech, dropping in a larger one or even two seems to be something we can expect in a few years.

 

[EOS 4 Life]

Third, when I hear "computational photography", I think the eye detection algorithms in the R5 and R6, not changes to the images themselves.

Yeah, "computational photography" is usually image processing on a smartphone that I would rather do myself on a computer in post.

 

[gektor]

Yeah, "computational photography" is usually image processing on a smartphone that I would rather do myself on a computer in post.

There are a plethora of techniques that can be grouped under the term "computational photography." Image stacking for greater dynamic range or noise reduction, for example, that cannot be achieved in post-production. The other category of techniques can be achieved in post-production and is similar to human "developing" and retouching of images. For enthusiasts, editing your own image is part of the hobby. For professionals, any automation that gives better results faster is essential. As a professional, being able to deliver perfect results fast is important. There are virtually no customers who want to get RAW images of their wedding or other event. Out of camera JPGs are considered inferior and tinkering with every single RAW is not profitable in most cases. So presets are made, which are a form of automation already. But why are the out of camera JPGs need to be so inferior?

Several years ago, articles like these would be totally unthinkable: Shooting an Entire Wedding Day with the iPhone 11 Pro (petapixel.com), Pro Wedding Photographers Compare iPhone 11 Pro to Canon 5D Mark IV (petapixel.com). The fact that we can now seriously compare a phone (!) and a pro-class camera is completely out of this world by my photographic standards. Still, phones will always be very limited in terms of sensor and lens size. Imagine what can be achieved with modern smartphone technology combined with large camera sensors. As a professional, you could get perfect pictures and deliver each one without any editing.

Today, the two worlds are very separate. There is a professional workflow that, although camera JPGs are getting better, requires expertise and time for post-processing. And there is an automated workflow in a phone that is able to deliver a perfect result in the same second a photographer takes a picture. I think professionals would also like to benefit from the level of automation that smartphone computational photography enables today.

So it could be that not only the M system dies out, but the entire market of classic digital cameras, if someone figures out how to integrate a performance of computational photography into "real" cameras, which in some cases are even inferior to today's phones. What a huge shame, actually.

 

[TravelerNick]

Today, the two worlds are very separate. There is a professional workflow that, although camera JPGs are getting better, requires expertise and time for post-processing. And there is an automated workflow in a phone that is able to deliver a perfect result in the same second a photographer takes a picture. I think professionals would also like to benefit from the level of automation that smartphone computational photography enables today.

So it could be that not only the M system dies out, but the entire market of classic digital cameras, if someone figures out how to integrate a performance of computational photography into "real" cameras, which in some cases are even inferior to today's phones. What a huge shame, actually.

Making things easier just makes it easier to replace the photographer. The last people to benefit from any of this are the people being replaced.

BTW you're not asking for something "perfect" you're asking for something the software thinks is "perfect".

The other issue is it's easier to do what you want with tiny little sensors. It's not just that they need it more it's just easier. Stacking multiple images for every shutter click means taking multiple images in a very short period of time. A smaller sensor is always going to be faster than something bigger.

But image stacking also imposes limits. If we assume 10 frames per image your max shutter speed is now 1/10th of an instant. Make that instant too long and you're now dealing with motion.

 

[gektor]

Making things easier just makes it easier to replace the photographer. The last people to benefit from any of this are the people being replaced.

Absolutely. It's not a good thing for photographers. Still, photographers and their tools need to evolve with progress.

Many professions have become extinct because of automation, and that's not necessarily an exclusively good thing. But it's also well known that you can't stop automation. If the process of carefully "developing" RAWs by hand is replaced by automation, then this slower kind of work associated with classical tools will be pushed out into and become exclusive to the hobby and luxury markets.

If the level of images produced by a professional photographer can be achieved more easily and quickly by anyone, the benefit of hiring a professional diminishes even more. If the hired professional is using hardware and processes that are inferior to modern standards, then it's a tragedy. That's why especially the devices used by professionals urgently need to evolve.

The M system is partly aimed at hobby photographers who have time to tinker with RAWs and edit their images manually. Consumers, who are also the target group of the M system, are more critical because they might want a good JPG image straight from the camera and are not willing to accept that their dedicated camera takes worse pictures than their phone. The "consumer" group of users may indeed be disappointed by their brand new canon M or R.

The R system is aimed at professionals who have their workflows. But I think that professionals today, in 2021, may still need more modern technologies and automation to stay competitive. What they get is the same hardware as always, but with improved specifications. It's like developing better and better gasoline cars in a world that is moving to EVs. Sooner or later, gasoline cars will become a niche market. Just like classic photography tools as opposed to intelligent tools. Today, there is no known highly intelligent camera with a large sensor, good optics, and a high degree of AI and automation. The need for such system will become clearer in the years to come.

So I conclude that neither the M nor the R series nor anything else from Sony, Nikon, etc. can be considered a truly modern tool for photography. All these shiny new cameras like R5, R6, etc. are the culmination of classic camera design that will probably come to an end in the future. On the other hand, the dynamic range of sensors may eventually become so good that some computational techniques are no longer necessary. But even then, if smartphones continue to produce more appealing images through some kind of computation, people will prefer these modern tools.

 

[stevelee]

Pros I have talked with use Lightroom to automate processes. They don't have time to do the tinkering with each picture that I do with ACR and Photoshop or how I similarly treat Lightroom. Just as there are presets in Lightroom, there are picture styles in making JPEGs in camera. With a bit more operating memory, they could set up user settings in picture styles, or maybe they already have and I just don't use them.

 

[EOS 4 Life]

Image stacking for greater dynamic range or noise reduction, for example, that cannot be achieved in post-production.

Why not?
A lot of cameras do exposure bracketing.
Those images can be combined in post into a single image with a high dynamic range.

 

[mdcmdcmdc]

Why not?
A lot of cameras do exposure bracketing.
Those images can be combined in post into a single image with a high dynamic range.

Agreed.
Image stacking has been a standard post-processing technique in digital astrophotography for 20 years. You can do it in conventional post processing SW like Photoshop, or there are many astrophotography-specific programs available.

 

[stevelee]

ACR automates HDR by stacking, for example.