Canon confirms 8K capable EOS R camera in development

jolyonralph

Game Boy Camera
CR Pro
Aug 25, 2015
1,423
944
London, UK
www.everyothershot.com
The Sony shooters like to brag about under exposing by 4 or 5 stops and recovering a photo that would otherwise be trash. Personally I think they should just learn to become better photographers.

What an ignorant statement! Not sure if you're just inexperienced or simply a Canon fanboy who can't handle that the competition has a better sensor.

There are plenty of situations where the shot you want to take has extreme variations in lighting, especially when you're shooting outdoors (or in a room with a sunlit window) on a sunny day. Being able to work with images that have portions either underexposed or overexposed is something REAL photographers do too, you know...
 
  • Like
Reactions: 1 users
Upvote 0

jolyonralph

Game Boy Camera
CR Pro
Aug 25, 2015
1,423
944
London, UK
www.everyothershot.com
I would, for the TS-E 17/4L and the EF 11-24/4L. The front filtering options for those lenses are a massive PITA, the TS-E has no rear gel slot and though the 11-24 does have the rear slot, gel filters don’t deliver the best IQ and polarizing isn’t an option. Even if Canon delives RF versions of those lenses, it seems unlikely they’d replicate the rear CPL/ND option, doing so would defeat the claimed advantages of the RF’s shorter flange distance

Specialist EF-mount lenses such as the TS-E range are likely to be the last EF lenses to stay in production in the future when RF has entirely taken over for this very reason. I would not be surprised though if they do future RF T/S lenses that they keep the same flange distance and add in a filter slot for the same CPL/ND that fits the adaptor - basically the existing EF lens 'welded' to an adaptor. The advantages of the shorter flange distance in this case being outweighed by the advantages of using that space for the filter.

I think the EF 11-24 won't be directly replaced, I expect something like an RF 12-24 f/2.8 lens at some point which won't have the filter adaptor.
 
Upvote 0
Am I wrong, or is the statement that Canon is developing something really just mean that they are looking at things in R&D and likely nothing is coming in any of the next generation of EOS R cameras that will likely have 8K video. This may be looking towards 2 or 3 generations in the future or maybe for pro cinema/broadcast cameras soon.

I can't wait for them to be able to do so, mostly because that should also mean they can process higher resolution stills faster and maybe we can get a 5DS(R) type mirrorless sooner that can shoot 10 frames per second for more than a few shots in a row.
 
Upvote 0
This was discussed above. 8K is 7680×4320, which is 33 MP. Those dimensions taken from a 3:2 aspect ratio sensor would require 7680x5120, which is 39 MP. In other words, the 5Ds/R sensors are already fully capable of delivering 8K video from a resolution standpoint, no steroids required.

Yes, Canon definitely already has the sensor capable of 8K in FF. I think a 16:9 crop of the 5DS/r sensor yields 8688x4888 which would allow a 1:1 pixel readout at 7680x4320 with a small 13% crop which would be excellent for video with no pixel binning at all and a 4.14um pixel pitch and still be able to take 50MP stills. It will be interesting to see where Canon takes this. I think their upcoming 8K cinema camera will give us some hints of EOS R 8K's future.

The only other player right now dabbling in 8K within the pro-sumer space is Sharp. Announcing at NAB at under $5. But 8K on a M43 sensor combined with 2.25um pixels (Even my Iphone XS Max has 1.4um pixels) on a H.265 codec writing to SD cards is a scary thought...
 
  • Like
Reactions: 1 user
Upvote 0
The Sony shooters like to brag about under exposing by 4 or 5 stops and recovering a photo that would otherwise be trash. Personally I think they should just learn to become better photographers.

Wow that hurts.. I guess I'm one of those sadists that have my exposure dial set at -3 all day long so I can come back from a wedding shoot and sit in my iron chair with my leather harness in front of Lightroom sweating and seeing how much of the shoot I can salvage...
 
Upvote 0
Yes, Canon definitely already has the sensor capable of 8K in FF. I think a 16:9 crop of the 5DS/r sensor yields 8688x4888 which would allow a 1:1 pixel readout at 7680x4320 with a small 13% crop which would be excellent for video with no pixel binning at all and a 4.14um pixel pitch and still be able to take 50MP stills. It will be interesting to see where Canon takes this. I think their upcoming 8K cinema camera will give us some hints of EOS R 8K's future.

The only other player right now dabbling in 8K within the pro-sumer space is Sharp. Announcing at NAB at under $5. But 8K on a M43 sensor combined with 2.25um pixels (Even my Iphone XS Max has 1.4um pixels) on a H.265 codec writing to SD cards is a scary thought...

They're the only player that has made a public statement that they're playing with 8K. Sony is gearing up for 8K and will probably use their line of cameras to show what can be done with their 8K sensors, like the Sony sensor that Sharp is using. They have an 8K sensor that Nikon will soon be using and have a few others that'll be sold to other companies. The sensor in the Sharp is a tweaked version of the Panasonic GH5S. I'm curious if Panasonic will offer a service to swap out the sensors for that one, similar to what some cinema companies do.

The technology in the Sharp camera is the very same that's used in the new 48 megapixel cell phones coming out with quad bayering. The bayer mode can fortunately change to increase low light performance by dropping the resolution down to 12 megapixels.
 
Upvote 0
How are you defining ideal?
Video resolution that is a whole-number multiple of existing industry standards paired with stills resolution that captures the majority of resolution available in 135 format sensors sounds pretty ideal to me. 4k provides lower stills resolution than most users require. By the time video resolutions reach 16k current ILC camera formats will be about as relevant as a Kodak Brownie is today.

edit: Currently I have 5 digital camera that I use and every one has a different resolution. That is ridiculous. I'd like to see standardization on 8K so sensor designers can focus on attributes of sensors that add actual value. I don't expect this to happen right away but I would hope that's where we will be in five years or so.
 
Last edited:
Upvote 0

cellomaster27

Capture the moment!
Jun 3, 2013
361
52
San Jose - CA
8K.. since 4K is cropped 1.25x currently, then 8K would be 2.5x crop? HAHA This is cool and all but what if we just acknowledge that Canon is behind and give us a good 4K just to catch up? Maybe they can down sample from 8K, that would probably be better. True 8K can be made for cinema line? Who here would shoot actual 8K with a mirrorless camera, such as a future R, for production? Not that you can't.. it's just, not the best tool for that type of job imo.
 
Upvote 0
Mar 2, 2012
3,188
543
Video resolution that is a whole-number multiple of existing industry standards paired with stills resolution that captures the majority of resolution available in 135 format sensors sounds pretty ideal to me.

The highest resolution 135-format sensors commercially available are 70MP; that’s far higher. I don’t think there will ever be such an intersection. When video goes 16k, stills will be 250MP+. Video necessitates more resources, and holding stills to video resolution is an artificial limitation which IMO is non ideal.
 
Last edited:
Upvote 0
Talking about finalizing a 2019 roadmap. Sure would be great if Canon would reveal it to us mortals.

That caught my eye too. Canon hasn't been big on "roadmaps." He might have been referring to an internal roadmap, but he seemed to suggest there was to be a public one. It does make business sense for them to do so, given that they've left the market uncertain that the pro body will come to match the amazing lenses they've put out. Nikon addressed this same issue (having caused the opposite problem - pretty respectable bodies with mediocre lenses) by putting out a roadmap. Problem for Nikon is that the roadmap they released reveals that the lenses they're going to push out for the Z mount still won't come close to Canon's ambitions.

If Canon's 1DX-like R mount camera is going to come out 1.5-2 years from now, then they might do better with uncertainty, rather than revealing the chasm of time before release.

Couple points of interest:
1 - I find it interesting that multiple people in the comments don't consider the 1DX2 to be the be-all of pro cameras. Its resolution and fps and size makes it a bit of a niche camera, versus perhaps a more common pro use case of high resolution and lower fps.
2 - Following on that, I consider myself to be in a bit of a temporary place with my 5D4 cameras. They don't have the fps needed, but they're better than the 1 series for my purposes, due largely to resolution. The slow fps was the main weakness of the 5D4 release, in my opinion, although I like the cameras very much. I bring this up because Canon can't release an R mount camera that is sort of like the 5D4 that has slightly lower or same fps and still inspire a massive upgrade cycle among us 5 series owners. I don't think they'll inspire quite the upgrade cycle unless they give pro users an upgrade in some important metrics, such as sensor quality/ISO, fps, although features such as IBIS and a few other nifty new mirrorless tricks will certainly provide some bait.
3 - If Canon indicated a 5DSR upgrade in R mount in 2019 on a roadmap, I'd by an R right now and start shifting lenses, selling off some EF glass in particular for the 50 and the 28-70. But, if precedent holds, we all have some time, and maybe the current R won't be the backup body I want when the real pro body comes out in 2020-2021.

Random side note. I'm very curious what's going to come out from Panasonic and Sigma in terms of L mount bodies. As it is now, I have 6 or 7 Art lenses on Canon mount, and find them in most cases superior to the Canon equivalent, with a few exceptions (35 & 85 f/1.4s come to mind). Sigma hasn't exactly blown everyone away with their bodies, as they've hitched their wagon to their very innovative-but-perhaps-not-the-most-practical sensor tech. Panasonic, on the other hand, is impressive as heck on the body side of things. I bring this up because at some point in the next couple of years we're going to be able to compare Sigma's Mark 1 Art series to a perhaps-redesigned set of primes for the new L mount they're supporting. Yes, they could just add L mount to existing designs, but I suspect Sigma will want to exploit the design flexibility of the newer, shorter flange distance, much like Canon has done of late. This may give us the option of having better-than-Canon glass from Sigma on Panasonic bodies that use Sony sensors. You may all flame me now.
 
  • Like
Reactions: 1 user
Upvote 0
The highest resolution 135-format sensors commercially available are 70MP; that’s far higher. I don’t think there will ever be such an intersection. When video goes 16k, stills will be 250MP+. Video necessitates more resources, and holding stills to video resolution is an artificial limitation which IMO is non ideal.
I'm aware that it's possible to make a sensor with higher resolution than 8K for stills. I am talking about hybrid cameras optimized for both stills and video.

If you are working in a field that requires resolutions greater than what can be provided by an 8K sensor I'm sure you will be able to find someone to sell you one. Larger sensors and sensor arrays will provide much better results than adding smaller and smaller photosites on 135 sensors IMO but that's not what we're discusing.
 
Upvote 0
Mar 2, 2012
3,188
543
I'm aware that it's possible to make a sensor with higher resolution than 8K for stills. I am talking about hybrid cameras optimized for both stills and video.

If you are working in a field that requires resolutions greater than what can be provided by an 8K sensor I'm sure you will be able to find someone to sell you one. Larger sensors and sensor arrays will provide much better results than adding smaller and smaller photosites on 135 sensors IMO but that's not what we're discusing.

I just trying to understand how 30ish MP constitutes capturing the majority of the resolution from 135 format sensors (given that such sensors with double that resolution are on the market), and why 8k is ideal and, say, 16k (another even interval) would not be.

Edit: on the first point, did you instead mean capture the resolution available from the majority of 135 format sensors?
 
Last edited:
Upvote 0
I just trying to understand how 30ish MP constitutes capturing the majority of the resolution from 135 format sensors (given that such sensors with double that resolution are on the market), and why 8k is ideal and, say, 16k (another even interval) would not be.

Edit: on the first point, did you instead mean capture the resolution available from the majority of 135 format sensors?
I may have caused some confusion by using the word "Resolution" to mean two different things. "Sensor Resolution" being the number of pixel/photosites on the sensor and "Subject Resolution" being the ability of a camera/sensor system to record subject detail. They are related but not the same.

My point is there is a declining benefit in regard to improving "subject" resolution by increasing photocells (sensor resolution) at a fixed sensor size (135). Adding more photocells than would be required to achieve 8K would somewhat increase subject resolution but at the expense of significantly increased computation and processing for video. Thus, I consider a sensor that can read out 8K at 1:1 and accommodate 40-45Mp for stills to be an optimum (ideal) solution for Stills/Hybrid cameras. The goldilocks solution with 4K being too small and 16K being too large.

In my opinion other disruptive technologies (AI driven computation) will replace contemporary camera systems before 16k systems/sensors roll out but that's further out there so who knows?

Hope that clarifies what I was trying to say.
 
Upvote 0

Hector1970

CR Pro
Mar 22, 2012
1,554
1,162
I'm sure there are plenty of users who use DSLR's / Mirroless for video and are probably delirious at the thought for 8K video on a camera.
For most users though surely even 4K is a nice to have and rarely used.
The processing power required to edit 4K is big, 8K must be a nightmare.
It just ups the ante completely in terms of memory cards required, long term storage and laptops to work with it.
An uninteresting HD video is no better in 4K or 8K.
Its different for real video specialists / movie makers I'm sure it would be very handy to get such detail in such a relatively small formatt.
I just find it hard to see that it move major numbers of cameras if it were on the spec of 8K.
I wonder can normal people spot the difference between 4K and 8K.
It's like a visual diminishing returns.
 
  • Like
Reactions: 1 users
Upvote 0
Mar 2, 2012
3,188
543
I may have caused some confusion by using the word "Resolution" to mean two different things. "Sensor Resolution" being the number of pixel/photosites on the sensor and "Subject Resolution" being the ability of a camera/sensor system to record subject detail. They are related but not the same.

My point is there is a declining benefit in regard to improving "subject" resolution by increasing photocells (sensor resolution) at a fixed sensor size (135). Adding more photocells than would be required to achieve 8K would somewhat increase subject resolution but at the expense of significantly increased computation and processing for video. Thus, I consider a sensor that can read out 8K at 1:1 and accommodate 40-45Mp for stills to be an optimum (ideal) solution for Stills/Hybrid cameras. The goldilocks solution with 4K being too small and 16K being too large.

In my opinion other disruptive technologies (AI driven computation) will replace contemporary camera systems before 16k systems/sensors roll out but that's further out there so who knows?

Hope that clarifies what I was trying to say.

Thanks for clarifying. For what worth, in taking your judgment of ideal as static (i.e., with current technology).
 
Upvote 0
I would assume that an 8K capable image sensor will also benefit still photographers.
I think both Nikon and Sony have higher resolution sensors (6K) that down sample to output better quality 4K, maybe what canon is thinking (start with 8K and get an even better 4k). and without the crop ;-)
 
Upvote 0
While I do like more pixels for the odd time I need to crop and having better video options is a bonus, I would prefer canon develop sensors with better dynamic range and maybe use AI like in smartphones to help improve images. Honestly when I compare the slow processing of the in-camera processing of HDR images on my Canon R, compared to the HDR on my cellphone, I prefer the shots on my cellphone more, as my phone does a better job of getting skies the blue I see and good detail in shadows at the same time. And I've tested out all HDR in-camera processing options to try to replicate the great results that are instant on my phone. Even at low light now, cell phones are processing images nicer. Yes I can do everything later in post, but there's a lot to be said about getting a good shot right away. I upgraded from the Canon 6D (mark 1), and the dynamic range is strikingly better and I would guess that part of the difference is the lens as well. I am comparing the 6D with EF 24-105 and R with RF 24-105. I was happy with this improvement, it has encouraged me to do more street photography as I prefer to do minimal processing and minimal HDR processing. Though on a cellphone HDR is an instant picture no different than having it turned off. It's no different on jpeg or raw for processing time for HDR, it's just very very slow. I assume this is a processor issue not a sensor issue, so I would hope that Canon look at using third party sensors that are outpacing their own to keep up with this aspect of photography.
 
Upvote 0