A high-megapixel EOS R camera is still on the roadmap [CR2]

tron

CR Pro
Nov 8, 2011
5,222
1,616
Simple. The R for some bizarre reason uses Canon cheaper remote control and I do lots a long exposure shots. The 6D MKII is a vastly underrated camera.
This is true. I do not understand why they did that. So my tc-80n3 is useless for EOS R. But I am sure there is a ton of similar 3rd party cheap controls out there.
 
  • Like
Reactions: 1 user
Upvote 0
Feb 28, 2013
1,612
272
70
Looks like I'm not going to shoot until next week, so having fun here instead!
The tests are from DPReview so I don't think they're pointless btw, I just extracted some samples to check what I wanted to check.
Ive spent a big chunk of my life looking at film / digital tests on a movie screen including up to pixel level. Most were pretty pointless because tests are a zillion miles from real world shooting where you dont have a controlled environment. In theory film should be trounced every time by digital from a purely technical perspective. From an artistic point of view that's rarely the case in the film industry.
Lenses are a different kettle of fish they do have the ability to completely change the look & character much more than the camera straight out of the camera.
If you want a perfect technical match then pixel pitch to line pairs per mm (nyquist) i.e. matching lenses to sensors is the way to go not randomised sensor tests.
 
  • Like
Reactions: 1 user
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
IBIS - assuming it works as expected - would be nice but I do not have an issue with my 5DsR. I use big white lenses with IS. Up to now I have used my 400DOII at 1/160 handheld and the result was excellent. I had even forgotten to chenge it afterwards for sometim but even so I got some keepers. And the ones that showed movement were from the treecreeper and not me.


For me, IS in all forms is a win as it lets me get away with longer shutters and keep the ISO down to earth. I'm frequently in tough lighting with my camera (indoor event no flash allowed, museum/church interior no flash / no tripod, nighttime walk handheld, etc.). So for me, more stops let's me take an ISO 6400+ shot back down into a higher IQ sort of place (ISO 400-1600). I'm a fan.

But with a supertele like yours I'm assuming (perhaps incorrectly) that you are dealing with moving subjects, so cranking down your shutter speed even further with IBIS might be possible but not very practical. It may not help you with portraiture or the odd bird on the tree once 1/160 (without IBIS) becomes 1/30 (with IBIS) shutter -- branches and leaves may move, the subject may move, etc.

- A
 
  • Like
Reactions: 1 users
Upvote 0
But that's what I was trying to check, the differences between the sensor tech on per-pixel level.
If you look carefully you'll see 90D and D850 with smaller pixels outperform 5Ds and 5DIV respectively.
That shows the difference and progress in noise control.

90D may have the same pixel density as the rumoured RS, so one of the questions is if such a noise level will be satisfactory for an 80Mp monster.

If you are comparing noise on a pixel basis then the test are pointless to the extent that unless there are radical differences between the sensors, the larger pixels must win, with some exceptions, over the iso range, depending on the iso. At base iso, circuitry noise can dominate as photon flux is so high and a poor FF can be beaten by an APS-C. At higher isos, the S/N of a pixel depends the number of photons hitting and hence on the square root of its area and so an APS-C has a S/N 1.6x lower than that of FF. The earlier Canon sensors were weak at low iso, but from the 5DIV and 80D, they have been very good. The 90D is very good at base iso, and does as well as can reasonably be expected.
 
Upvote 0
If you want a perfect technical match then pixel pitch to line pairs per mm (nyquist) i.e. matching lenses to sensors is the way to go not randomised sensor tests
Very good points and lenses are important, but I was checking just one specific aspect of sensor performance. I'm a pixel-peeper and I spend quite some time in 1:1 view mode when editing. More noise means less room for certain things such as shadow lifting, sharpening, noise reduction, heavy or not so heavy crop etc.
 
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,351
22,524
But that's what I was trying to check, the differences between the sensor tech on per-pixel level.
If you look carefully you'll see 90D and D850 with smaller pixels outperform 5Ds and 5DIV respectively.
That shows the difference and progress in noise control.

90D may have the same pixel density as the rumoured RS, so one of the questions is if such a noise level will be satisfactory for an 80Mp monster.
If you output to the same size as a low resolution sensor, then a very high resolution sensor will give you similar noise to the low level one with greater detail, at the expense of a larger initial file. This is why posters here are telling you that comparing pixel level noise is irrelevant for that case. If you are cropping, then you put up with the extra noise for the sake of extra detail. If you don't need the detail, then get the lower resolution sensor.
 
  • Like
Reactions: 2 users
Upvote 0

tron

CR Pro
Nov 8, 2011
5,222
1,616
For me, IS in all forms is a win as it lets me get away with longer shutters and keep the ISO down to earth. I'm frequently in tough lighting with my camera (indoor event no flash allowed, museum/church interior no flash / no tripod, nighttime walk handheld, etc.). So for me, more stops let's me take an ISO 6400+ shot back down into a higher IQ sort of place (ISO 400-1600). I'm a fan.

But with a supertele like yours I'm assuming (perhaps incorrectly) that you are dealing with moving subjects, so cranking down your shutter speed even further with IBIS might be possible but not very practical. It may not help you with portraiture or the odd bird on the tree once 1/160 (without IBIS) becomes 1/30 (with IBIS) shutter -- branches and leaves may move, the subject may move, etc.

- A
Exactly but IBIS on another camera like an R5 or 5D5 equivalent would help. I bought the R system to have access to RF15-35 and RF24-70 due to IS. Similarly with you I like shooting interiors like museums, churches, etc. So there IBIS would be welcome.
 
  • Like
Reactions: 1 users
Upvote 0
Feb 28, 2013
1,612
272
70
Very good points and lenses are important, but I was checking just one specific aspect of sensor performance. I'm a pixel-peeper and I spend quite some time in 1:1 view mode when editing. More noise means less room for certain things such as shadow lifting, sharpening, noise reduction, heavy or not so heavy crop etc.
Never had a problem heavily cropping with the 5DS that's one of its strengths thanks to the resolution. I only shoot at ISO100 generally at most ISO200. We have a wall with a 50ft x 20ft poster from the 5DS and its as sharp as a tack with great tonal range. I know at least two fashion photographers that use the 5DSr in the studio (along with Phase One) and their shots end up on billboards.
 
  • Like
Reactions: 1 user
Upvote 0
I have been wondering if any of my older L glass (first gen IS) will hold up to a 70MP+ sensor. Did you have to upgrade any L glass for the 5DSR, and if so what?
I upgraded the 28-70 and 70-200 as well as replaced the 400 5.6 with a 100-400. I added a number of others along with way. I felt the upgrades were worth it and needed with the higher resolution revealing softness in those older versions.
 
  • Like
Reactions: 1 user
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
Never had a problem heavily cropping with the 5DS that's one of its strengths thanks to the resolution.

I've got a beach volleyball shot that's cropped in more than 2x and still makes a sharp 16x20. High resolution has its advantages. I hope the rumor is true that the R5 is 45mp and I'll still be looking forward to an 83mp monster. Canon just might squeeze two camera sales out of me this year.
 
  • Like
Reactions: 1 user
Upvote 0
I'm on board for the EOS R3 name. I'd love to see it fit in to the roughly 5D price point and push the R series cameras that are replacing the 5D down in price a bit but that's probably asking for too much.

The EOS 3 was my first major EOS camera and I have fond memories of it. Should charge a battery or find the AA carrier and run a roll or two through it.
 
  • Like
Reactions: 1 user
Upvote 0
As a landscape photographer I've been waiting for over 2 years through hundreds of rumours of the upcoming Canon High Megapixel Camera:
(CANON-RS)
touted to be 75 to 100 Megapixels
this would be a huge
benefit for me as I print 40x60 inches and larger mostly

If Canon doesn't announce a high megapixel camera body in February 2020 I will be reluctantly jumping ship to Sony and buying the a7riv or the Nikon D900 (with the Sony 64 Megapixel Sensor) or even possibly the upcoming 100 megapixel Fuji GFX Lite Body

I would rather stick with Canon I love their hardware
and have been a Canon Aficionado since 1980 when I got my Canon A1 film camera.

I think Sony can be consistently relied upon to upgrade their camera bodies every 24 months and they have more resources in their sensor fabrication to stay on the
cutting edge

I don't give a Rat's Ass about video.
A 45 megapixel catch-all generalist body is not what I'm looking for

Please Canon if you're listening give the studio and landscape photographers a reason to stay with you.
 
  • Like
Reactions: 1 user
Upvote 0
May 11, 2017
1,365
635
Tripod landscape shooters still want low noise and good dynamic range. The more pixels Canon puts into this beast, the more I worry about the noise and overall performance. Time will tell but the 45Mp one may happen to be a sweeter spot for landscapes, especially low light/astroscapes.
It looks like we are going to see a 45mp F5 before we see a high res mirrorless, which may well be an indication where Canon thinks the market sweetspot is. On the other hand, Canon seems to think that there is room in the pricier part of the FF mirrorless space for 20mp, 45mp and high res cameras.
 
Upvote 0
If you output to the same size as a low resolution sensor, then a very high resolution sensor will give you similar noise to the low level one with greater detail, at the expense of a larger initial file. This is why posters here are telling you that comparing pixel level noise is irrelevant for that case.

But I don't want to resize to the smaller sensor's resolution. For a start I want to check how the sensor performs at its native resolution. Next step I might normalise and compare at the same size, but why cannot I check the noise at the native resolution?
 
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
That wasn't magnification, that was viewing 1:1. You can click on Comp button and see similar results, slightly imroved for 5Ds because of downscaling.



I don't see shadows in 5DIV any brighter tbh. Moreover, to me it feels like 5Ds image is actually a bit brighter than 5DIV on DPR, same in your sample.



It is sharper, but if you apply any noise reduction (LNR) you'll loose a lot of detail.

What actually disappoints me though is this modified comparison with 90D https://www.dpreview.com/reviews/im...t=1&x=0.8574631217392756&y=0.2192025092867554

Again at 1:1 90D also produces messy shadows, worse than 5DIV. If this is what new-gen Canon sensors are capable of, I'm concerned. They still haven't caught up with Nikon D810 from 2014.

!!!!NEWS FLASH!!!!

1:1 is a different enlargement.magnification ratio for sensors with differently sized pixels.

If you are enlarging 4.14µm x 4.14µm pixels (5Ds) to the size of one pixel on your monitor it is being magnified/enlarged more than enlarging 5.36µm x 5.36µm pixels to the size of one pixel on your monitor.

In the case of an FHD 23.5" monitor with 96 ppi:

The 5Ds image is being enlarged to an equivalent 90.5" x 60.3" size at 1:1
The 5D Mark IV image is being enlarged to an equivalent 70" x 46.7" size at 1:1
 
  • Like
Reactions: 2 users
Upvote 0

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
4,722
2,655
Please correct me if I have misunderstood you. If it's viewed 1:1 with no downscaling, does that mean you are looking at the same number of pixels for all of them? If that is so, then the 5DIV with its larger pixels must win by a real margin. If they are all viewed at the same output size, I would expect it to be much closer but with a small advantage to the 5DIV because of edge effects as the 5DIV has a truly excellent sensor. Sony's sensor act in a similar way with the lower res 7DRIII slightly better than the newer 7DRIV.

7DRIII and 7DRIV? What are those? Do you mean the α7rIII and α7rIV?
 
Last edited:
  • Like
Reactions: 1 user
Upvote 0
It's all true but as I said we were talking about different 'enlargement ratios'.
I was talking about digital conversions. When you view 1:1, there's no compression and no enlargement of the data from your raw file. Anything other than 1:1 requires additional digital conversion (resampling). Most of the time it's downsampling so that you can fit large size image into the screen or a window on the screen.
Because there's no resampling at 1:1, you see the sensor noise as it is and that was important for my particular test.

!!!!NEWS FLASH!!!!

1:1 is a different enlargement.magnification ratio for sensors with differently sized pixels.

If you are enlarging 4.14µm x 4.14µm pixels (5Ds) to the size of one pixel on your monitor it is being magnified/enlarged more than enlarging 5.36µm x 5.36µm pixels to the size of one pixel on your monitor.

In the case of an FHD 23.5" monitor with 96 ppi:

The 5Ds image is being enlarged to an equivalent 90.5" x 60.3" size at 1:1
The 5D Mark IV image is being enlarged to an equivalent 70" x 46.7" size at 1:1
 
Upvote 0