This is the possible Canon RF mount camera roadmap [CR2]

Tangent

EOS 90D
Nov 13, 2015
126
71
For the RP Plus / level up I'd want and expect
  • larger battery, longer battery life
  • still compact, as much as possible given larger battery
  • IBIS
  • Better EVF than RP (the R6 EVF, most likely)
  • Improvements to AF and video
  • dedicated / programmable button to select AF mode; esp eye tracking
  • weathersealed, ar at least more so than RP
  • Still has flip out LCD, of course
  • Moderate MP 30 more or less
  • Higher frame rate than RP
  • Built-in stills intervalometer (!!!)
  • Better DR (!!!)

For the APS c the roadmap rumor didn't say anything about a 7D level of performance. I dunno, hope I'm wrong, but that kinda sounds like wishful thinking...

A guess: The APSc R lenses will mount on any R body -- the R bodies are already set up for a crop mode. Might be useful in a pinch on a FF. The R lenses will mount on the the APS body too, of course, with a crop factor. So all the lenses will be interoperable.

Ummm... things aren't loooking real good for the M line now...
 
Last edited:

Darecinema

Addicted to lenses.
CR Pro
Sep 8, 2018
37
48
Anybody know if there’s gonna be Black Friday or cyber Monday deals on the C70?
I highly doubt it seeing as it hasn’t even shipped yet and Black Friday deals are typically to clear out older stock to make way for newer models and get rid of taxable excess inventory before year ending.
 
  • Like
Reactions: navastronia
Nov 16, 2020
3
4
I suppose there isn't any news on that M-mount camera that was previously reported. I am excited to upgrade from my M50 into an M-mount camera that I can take with me in a tiny sling bag everywhere, especially if said new M-mount camera had IBIS, 4k without line skipping/cropping, and I wouldn't even be mad if it omitted the viewfinder again (provided they still allow the attachable one). Finding another body to use my excellent EF-M 32mm f/1.4 on is proving difficult. RF-mount cameras with those features are rather big.
 
  • Like
Reactions: Bahrd

Darecinema

Addicted to lenses.
CR Pro
Sep 8, 2018
37
48
is it possible to have ProRes internal recording to CFExpress in the future Canon EOS R1?
I think this is mainly a licensing point honestly, Apple licenses the ProRes Codec so Canon would have to have an incredibly strong reason business wise to make that leap and I don’t think they do as much as it pains me. Átomos and Black Magic Design are two companies who have made the leap but they are purely for cinema prosumers and pros so that makes sense in my mind, not so much for primarily stills cameras. I can’t think off the top of my head of any Milc or DSLR that records ProRes internally from any brand but I might be forgetting one. I think both Nikon and Panasonic had paid upgrades on their roadmap and I think you are making that payment to cover the license fees. Of course I could be completely wrong on all of this...
 
  • Like
Reactions: Karel

David - Sydney

EOS RP
CR Pro
Dec 7, 2014
533
440
www.flickr.com
5DV! 5DV! 5DV!

Come on, Canon – just give us one final 5D!

:mad:
I was really hanging out for a 5Dv as well but I can't see it happening. They can't get enough R5 to customers yet either and I think that by the time they meet all the pent-up demand, then there won't be much demand left for a 5Dv especially once current 5Div users get to see how the R5 works for them. Simple things like the flippy screen make a huge difference for tripod users or vlogging or even selfies.
I assume that you mean that a 5Dv will have the same body size/sealing as the 5Div but if we see how the 1DXiii works compared to the R5 then the only benefits I can see are the OVF and battery life. If the R5 innards are put in a 5Dv then maybe there would be better heat management given the size difference but that is about it. The best AF for the 1DXiii and video is via liveview so holding out the camera is not intuitive.
I would expect that a 5Dv would be more expensive than the R5. Losing the EVF cost but adding the OVF/pentaprism/mirror + AF sensor.
 

H. Jones

Photojournalist
Aug 1, 2014
498
745
The reality is after my amazing experiences with the R5, I'm going to sell my 5D/1D and leap on the R1 as soon as it gets announced, no matter the actual specs.

That said, this is my expectation:
  • 20 or 30mp
  • Quad-Pixel AF
  • Probably no IBIS
  • Hopefully global shutter
  • Up to 60 FPS, no-EVF-delay silent shooting with absolutely no rolling shutter whatsoever
  • ~6K raw video internally at 60 FPS
On its own without changing any specs of the 1DX3, Canon could sell me a dozen R1 cameras if they include a global shutter. That would be absolutely friggin amazing, especially when you consider that you could easily anti-flicker at any framerate or banding since the whole frame would be exposed at the exact same time, and Canon could hopefully include the shutter micro adjustment that Sony included in the A9II.

There is absolutely no reason to increase the resolution beyond 30 mp, and I would honestly say 30 is too high if Canon pulls off 60 fps stills, which is a crazy spec that would make 20 megapixels totally acceptable in exchange.

Looking at the R5's raw 8k video, Canon has definitely done some amazing work at increasing their raw output, and I think Canon is really going to move towards photo framerates easily matching video framerates.

As for IBIS, Canon has consistently continued to include it in many of their high-end lenses, so I anticipate that the R1 will avoid IBIS for durability and heat. Cinema cameras don't tend to have IBIS, as most who work in pro video never rely on IBIS and already have their own preferred methods of mechanical stabilization.

On the video front, out of dozens, I don't know anyone that I've ever met with a 1DX3 that has used 5.5k raw video. In the news industry, we're mostly shooting 1080p, and in my commercial work we *sometimes* shoot 4K. I think 6K would be more than enough to satisfy the cinema shooters who use the 1D series while not forcing Canon to increase photo resolution by much.

One of the biggest things to me, though, would be full uncompressed raw at 20 fps with anti-flicker. That would be an absolute dream.
 

navastronia

EOS RP + 5D Classic
Aug 31, 2018
653
754
The reality is after my amazing experiences with the R5, I'm going to sell my 5D/1D and leap on the R1 as soon as it gets announced, no matter the actual specs.

That said, this is my expectation:
  • 20 or 30mp
  • Quad-Pixel AF
  • Probably no IBIS
  • Hopefully global shutter
  • Up to 60 FPS, no-EVF-delay silent shooting with absolutely no rolling shutter whatsoever
  • ~6K raw video internally at 60 FPS
I like your list except for what I've bolded. I think Canon should be able to implement decent IBIS, in a robust and sturdy mechanism, by the time they release the R1.

Re: video specs, I think really nice-looking full-sensor 6K would be terrific and would put the camera's resolution at 24 MP (like the S1H)
 

justaCanonuser

Grab your camera, go out and shoot!
Feb 12, 2014
710
513
Frankfurt, Germany
Just hope the APS-C model is every bit a 7D3 and doesn’t skimp on the features.

There is a HUGE market for a 7D3!!
I'd be ready to get me an R7 Mark I :D Why? I think Canon's dual pixel based AF system is now so mature that an R7 would be a really nice upgrade for birders, since the 7D2's AF performance is not overwhelming. Plus, less noisy sensor electronics would be good, our Nikon D500 really puts shame on Canon in that respect. I don't expect Canon to release any new DSRL model on the level of the 7D, so I'd welcome an R7 - IF its price would be settled well below 2k $.
 

justaCanonuser

Grab your camera, go out and shoot!
Feb 12, 2014
710
513
Frankfurt, Germany
As for IBIS, Canon has consistently continued to include it in many of their high-end lenses, so I anticipate that the R1 will avoid IBIS for durability and heat. Cinema cameras don't tend to have IBIS, as most who work in pro video never rely on IBIS and already have their own preferred methods of mechanical stabilization.
IBIS is a nice example for the fact, that Canon sometimes seems to listen too much to people posting their wishes on photo sites. Without the weak heat links of IBIS, the R5 surely would have no heating issues with long 8k video takes. That's why I expect no IBIS in future pro bodies, be it the R or Cine series.
 

PerKr

EOS 90D
Jul 11, 2018
103
93
Sverige
Thinking the APS-C might be the lower cost RP replacement. R100. Then an R10 to replace the original R inbetween the R100 and R6. The question is though; how will they differentiate between the R6 and a lower priced full frame body? Single card slot, lower resolution viewfinder, 8fps mechanical shutter, less efficient IBIS? I would assume they would be using the same sensor and processor so the AF would be similar.
 

koenkooi

EOS R
CR Pro
Feb 25, 2015
1,491
1,293
Thinking the APS-C might be the lower cost RP replacement. R100. Then an R10 to replace the original R inbetween the R100 and R6. The question is though; how will they differentiate between the R6 and a lower priced full frame body? Single card slot, lower resolution viewfinder, 8fps mechanical shutter, less efficient IBIS? I would assume they would be using the same sensor and processor so the AF would be similar.
The RP lacks a front curtain, so it's EFCS only. Combine that with a smaller body made with cheaper materials, you should have decent cost savings. Personally, I think putting the R6/1Dx3 sensor + digic X in the current RP body would already be a massive upgrade.

When shooting the RP and 1dx3 side by side I liked the 1dx3 results better, especially above ISO 800. The improved AA filter and DR more than compensated for having fewer megapixels.

There are a lot of knobs Canon to twist to get to a certain price point: body, shutter, sensor, IBIS, card slots and more. I also hope 'software' features won't be capped by marketing, only by engineering. So no RAW video or 20 fps stills on an camera with a single UHS-1 slot.
 
  • Like
Reactions: navastronia

padam

EOS R
Aug 26, 2015
1,106
740
The RP lacks a front curtain, so it's EFCS only. Combine that with a smaller body made with cheaper materials, you should have decent cost savings. Personally, I think putting the R6/1Dx3 sensor + digic X in the current RP body would already be a massive upgrade.

When shooting the RP and 1dx3 side by side I liked the 1dx3 results better, especially above ISO 800. The improved AA filter and DR more than compensated for having fewer megapixels.

There are a lot of knobs Canon to twist to get to a certain price point: body, shutter, sensor, IBIS, card slots and more. I also hope 'software' features won't be capped by marketing, only by engineering. So no RAW video or 20 fps stills on an camera with a single UHS-1 slot.
I would think they might actually keep the same RP sensor in their new cheapest model, the other one in betwen the two might have a new sensor.

From the M50 Mark II people should be aware that new models in the budget segments may not mean massive upgrades to make a case for the higher-end R6 and R5 models.
 
  • Like
Reactions: navastronia

H. Jones

Photojournalist
Aug 1, 2014
498
745
I like your list except for what I've bolded. I think Canon should be able to implement decent IBIS, in a robust and sturdy mechanism, by the time they release the R1.
I don't disagree and I would personally love IBIS, but part of what makes me think that the R1 could ditch IBIS is if it has a global shutter, Canon could brag that they have a pro camera with 0 moving parts, which would be a huge flex for the pro market where one could imagine getting literally 100 million photos out of the camera before something non-mechanical wears out. That would be a big statement piece for a pro body, and I think Canon could go for that in the first generation, before they include IBIS in a mark II.
 

Jasonmc89

EOS 80D
Feb 7, 2019
288
301
UK
I'd be ready to get me an R7 Mark I :D Why? I think Canon's dual pixel based AF system is now so mature that an R7 would be a really nice upgrade for birders, since the 7D2's AF performance is not overwhelming. Plus, less noisy sensor electronics would be good, our Nikon D500 really puts shame on Canon in that respect. I don't expect Canon to release any new DSRL model on the level of the 7D, so I'd welcome an R7 - IF its price would be settled well below 2k $.
Agreeeeeeeee
 
  • Like
Reactions: justaCanonuser

usern4cr

R5
CR Pro
Sep 2, 2018
755
825
Kentucky, USA
They could go to a BSI sensor to improve wide angle corners, but it’s my understanding that the sensor micro lens array can be designed to help even on non-BSI. This is optical science over my head, but all I can say is based on the R5 results I’m getting on the best RF lenses, Canon will need to do something with the sensor to improve the corners on a 90mp full frame setup.

So far, AFAIK, Canon has resisted using BSI sensors in their cameras. Sony, AFAIK, uses them in theirs. BSI is the best way to get high quantum-efficiency sensors, particularly as the size of each pixel gets smaller and smaller. But it costs more to design and manufacture. If Canon continues to use conventional FSI (front side illumination) on higher MP sensors (with the same frequency lithography) then most of the light will be blocked by all the misc. electrical lines before it reaches the sensor (even more so at the edges of the sensor), and thus they will either have to live with it (and have less light sensitivity) or have better micro lenses to funnel light around the blocking circuitry. I don't know if they use refractive microlenses or diffractive ones. Your linked article refers to using diffractive microlenses to get a spot with a diameter as small as 4 um (microns). A probable ~90MP sensor will have around 11.6K x 7.7K pixels, with each pixel being 36mm / 11.6K = 3.1 microns. So their 4 um spot is too big to be of any use. To make matters worse, their use of DP (dual pixel) focusing means each pixel is somehow divided into 2 sensors (left & right) so that you now have (I assume) a 1.55 x 3.1 micron half-pixel sensing area. As you approach the sensor edges & corners, the light hitting those pixels has an off-center average of incidence which will vary depending on the f# of the lens, so any design to adjust the microlenses towards the edges will have to guess which f# to design for, making matters worse. So I don't know what kind of magic they will use to make this work with a FSI design, but they already have done some pretty good magic as they've had to solve this on their 45MP sensor already.

If they surprise everyone and come out with a BSI design for their ~90MP sensor then it will make these issues easier, but they still have to have R,G,B filters as well as left-right lensing for their DP design. So they still have angle of incidence issues as you approach the edges of the sensor. One technique they could do is to use QP (qual pixel) design and then interpolate each quarter pixel in a 2x2 Bayer array as a "pixel" so that their marketing declares they have 2x2 x 2x2 = 16 pixels instead of 4 pixels. Then they'd have a 22.5MP QP sensor which they claim as a 90MP sensor since their software would interpolate the final image as 90MP. That would actually make it easier to produce, but result in more interpolation artifacts than one would expect in a normal 90MP Bayer array.

Despite all this, my guess is that they are going to just use their existing technology to double the number of pixels, but use a 1.4x times smaller lithography manufacturing so that it works as good as it happens to work and call it a day.

But all this doesn't touch on my main question I asked, which is how much of the image trouble you see as you approach the edge of the sensor is caused by the lens itself, and not the sensor or its microlenses. We know the resolution of lenses drops appreciably as you reach the edges and I would guess that this is the main cause of the drop in performance you are seeing and has less to do with the microlenses structure. So I would expect the ~90MP sensor to have the same ratio of loss of resolution towards the edges as you have now.
 
Last edited: