Is the Canon EOS R7 the next camera to be announced? [CR2]

[dtaylor]

Nikon has just officially discontinued the D500, arguably the best crop DSLR ever. https://nikonrumors.com/2022/02/01/the-nikon-d500-is-now-officially-discontinued.aspx/ I used one for over a year, incredible AF and superb IQ. I sold it because I prefer the R5.

I would prefer the R5 to. But someone could spend the next 10 years filling album after album with sports and wildlife shots from the D500. The shot they might miss for not having an R5, R3, Z9, A1, etc. would be comparatively rare.

It's not like the 2000's where sensor and feature upgrades seemed massive because the technology was young.

 

[Skyscraperfan]

As a crop camera uses a smaller part of the image circle, IBIS has much more space to move. So IBIS should be very strong when used with a full frame lens.

 

[David - Sydney]

It wouldn't have to be a fifth lens mount.

They can simply produce lenses that trigger a full-frame camera's crop mode, just like an adapted EF-S lens will. (I want them adequately labeled, though.)

You are right that it wouldn't need to be a new physical mount but it is still another "special" lens in the RF lens ecosystem. Appropriate labeling is still a 5th lens product in marketing terms,

It would apply only to a focal range of 10-22mm for instance though. The roadmap RF 18-45mm f/4-5.6 IS STM would be used on ff as is as well as APS-c sensors in crop mode.

 

[bbasiaga]

you'll be waiting a long time then. the battery life on MLCs is generally worse than their DSLR cousins, the prices of RF glass is generally more than EF equivalents and the RF cameras have also been priced higher than their DSLR equivalents.

All that said, I wouldn't give up my R5 Its been a fantastic improvement over the 5D IV.

MILC will get better with respect to battery life, but will never match DSLR. They are just doing so much more computing. Eventually batteries of higher capacity will get smaller, which will help, along with incremental improvements in chip efficiency. But honestly, battery life isn't as bad as it gets a rap for, unless you are using all the features at once like GPS, WIFI, etc.

If it did all this and took the R5/6 battery grip. I probably couldn't preorder it fast enough lol

Could you imagine the nightmare that would occur on this and other forums if the R7 DIDN'T take a page from the other APSc bodies and make a slightly more compact body (which would preclude the use of the BG-R10 grip)?!?!?! It would be a bloodbath!

The only thing I can think of that could achieve that level of critical scorn would be if it DID take a page from the other APSc cameras and make a slightly more compact body!


-Brian

 

[AlanF]

I would prefer the R5 to. But someone could spend the next 10 years filling album after album with sports and wildlife shots from the D500. The shot they might miss for not having an R5, R3, Z9, A1, etc. would be comparatively rare.

It's not like the 2000's where sensor and feature upgrades seemed massive because the technology was young.

Very true.

 

[cgc]

An f/1.8 lens will provide the same light intensity per unit area on the image at any focal length, which is why f-number is what we use for getiing exposure right. If you move back to 16 ft, the size of the duck decreases by 1.6x horizontally and vertically and so the total amount of light hitting it during exposure has gone down 1.6x1.6x. Blow up the image of the duck taken at 16 ft to the same size for viewing as the duck taken at 10 ft, and it will appear noisier.

The duck is the same size on both sensors and is being illuminated by the same lens. So, if both sensors are equally efficient, then there should be the same S/N for both.

Said with other words: when we "blow up" a FF image by 1.6x we are in fact doing a crop, ending on a FF sensor run as an APS-C one, with 2.56x less total light and the corresponding additional noise. So best to keep the FF body saving the steps back and getting the full light

The former poster tried to argument that the amount of light was equal, once accounted for the changed depth of field and angle of view. Yes, the exposure is the same (e.g. same Av values) but the sensor, as others have noticed, is 2.56 times smaller, so the effective light (once it is expanded to the print size) is 1.3 stops less. That also affects to midtones signal/ratio and overall IQ.

Theoretically it there exists the "systems equivalence" so, for example, we could build a 12-35 F1.4 four thirds lens (please forget about aspect ratio) providing those cameras the exact equivalent of a FF 24-70 F2.8 in all regards, including light capture ability (you simply keep your position and shot the 4:3 lens/camera halving the FF zoom and aperture values to get the exactly same photo with same amount of light and IQ). But wait a moment... it there exists such a lens? Will there someday exist a four thirds 14-35 F1 equivalent to the RF 28-70 F2?. And if somebody takes the challenge... will it be feasible?. I always use the same REAL example: the "L-series" Olympus 14-35 F2 (more than 900 gr), which was the world's first F2 zoom, compared to a FF close peer EF 24-70 F4L IS, which only weights 600 gr and on top of that adds IS and more wide angle plus 1:2 macro. On the other side, digital medium format does not keeps a big advantage over FF because what it seems to be truly difficult is to manufacture big aperture lenses. Perhaps a 0.5 crop medium format would be more efficient mimicking FF F1.2 lenses (F2.4 there) and of course to make exotic F0.7 glass (only F1.4 there) but maybe not so much from F2.8 and on (F5.6 there).

 

[Blue Zurich]

Oh, sorry for not knowing that, as I haven't followed any R-series developments since I tried the EVFs in the R and Rp and found them unusable.

Give the EVF in the R5/R6 a chance, it's light years better than tech from say 2016 (Pen F, I'm talking to you here)

Refresh rates, contrast...much more like WYSIWYG. Some say it is fully there, some would say close but not quite. No matter that debate, all in all these are great cameras, even in less than competent hands.

 

[maboleth]

In my case I have worn out multiple 7 series bodies. My current 7d2 is nearing 800k shots. It simply cannot go on forever. The top dial is getting jumpy but other than that it is still working well. Quite a testament to how well Canon made that body. Hopefully it lasts until a suitable R comes out.

I'm not sure what is holding you to upgrade to 90d, in my case it's better in every possible way, except two things:

1. Body is not all metal (though metal-plated), but looks sturdier than R6 IMO.
2. You cannot deep-program AF point buttons. For example, both on my 7D/mk2 and 5d4 I mapped back AF-ON to be One Shot and * button to be Servo with separate AF points (Spot AF and Zone). You cannot go that deep on 90d. That was a massive blow for me, but I learned to overcome it by using shutter button for AF (default) and AF-ON mapped as a Servo switch.
3. No dual card slots (with CRAW and high-capacity cards, I don't care for that at all)

But that's where drawbacks end for me. Speaking of its AF module, I find it improved over 7D2 and much improved over 7D. It's on par with my 5D4, slightly less precise with 50mm f1.8 STM, but then again that lens isn't the most precise one and it's sharper on FF overall as it's not pushing the optics beyond normal levels. LiveView on the other hand rivals mirroless cameras. Super fast, super smart, face recognition with eye-find focus and Focus Peaking for manual lenses.

All in all, I have effectively moved from 5d4 to 90d. With a price of 1k euros, it's a steal for such a great and modern aps-c camera.

 

[maboleth]

Give the EVF in the R5/R6 a chance, it's light years better than tech from say 2016 (Pen F, I'm talking to you here)

Refresh rates, contrast...much more like WYSIWYG. Some say it is fully there, some would not close but not quite. No matter that debate, all in all these are great cameras, even in less than competent hands.

EVF of R6 is great, but I guess as someone that grew up on mechanical Nikon FM2/3a cameras, I cannot treat EVF as "normal" viewfinder. What I see in EVF is not the scene I see with my eyes (OVF). That can have ups and downs, of course. One of the ups is that what you see on EVF is 99% what you will get. But in a sense, EVF is sort of virtual-reality-viewfinder and I totally understand people that just dislike it no matter what.

 

[josephandrews222]

I want to jump in here but don't know exactly where to start...let's try these two:

=====

maboleth, you write:

"One of the ups is that what you see on EVF is 99% what you will get. But in a sense, EVF is sort of virtual-reality-viewfinder..."

I need some help understanding your comment here! (no snark intended); is it just habit that draws you to OVFs?

=====

In this thread and one or two other rather recent threads here, there's been lots of talk about size and ergonomics and sensors etc.

Posters talk about the 90D and its crop sensor...there's even a table or three listing this property or those data (again, no snark here--I like tables full'o data!)...somewhere I saw one or two posts where someone talks about the 90D and mirrorless as well and wonders about the R7 etc...with ZERO mention of Canon's M format!

If anybody cares, I've posted here on CR about the M format...and lately, in particular, about the M6MkII/(adapted)EF 70-300 IS II combination.

I've done the vast majority of my wildlife shooting with a 5DMkIII body and various telephoto zooms (including the Canon 100-400 IS II)...and indeed its old-school viewfinder is great and comfortable at the same time.

But the M6MkII/(adapted)EF 70-300 IS II combination, with either of the EVFs attached...is pretty darn good.

Pretty darn good.

And of course the sensors inside both the M6MkII and the 90D? The same.

...smh

 

[unfocused]

You seem to be saying that if you change the sensor from FF to APS-C but not the lens, zoom setting, or distance to subject then you get A resulting image with a narrower FOV (corresponding to a focal length 1.6x longer) AND a narrower aperture A little more than a stop you state, the math would put it at 1.3 stops). The former is true, the latter is not fact, it’s completely false.

Field of view is narrower because you’re sampling a smaller area of the image circle. Focal length doesn’t change.

Aperture doesn’t change. If distance doesn’t change, the depth of field is actually (very) slightly shallower (because the smaller sensor has a smaller circle of confusion). It is only if you increase the distance to subject to match framing that the depth of field corresponds to a 1.3 stop narrower aperture.

The smaller sensor means less total light collected, which means effectively 1.3 stops more noise.

Hopefully I am misunderstanding your post. if not, it’s a good example of how pressing a shutter button for 40 years doesn’t confer technical understanding, anymore than someone who flips a light switch their whole lives necessarily knows how wiring actually works.

I can't believe that after all these years and endless debates we are on the same page. I'm glad you finally came around.

Seriously, I think we have both been correct just not necessarily communicating clearly.

However, I do question one statement (or at least want to clarify it).

"The smaller sensor means less total light collected, which means effectively 1.3 stops more noise."

I would say that this is only true because the smaller sensor is likely to have the pixels more tightly packed into the sensor. In other words, the size of the sensor isn't the determining factor, it's the size of the pixels. The 45mp R5 does not gather any less light when cropped to APS-C size. Of course it is no longer 45mp, but about 17.5 mp. Whether you are shooting in 1.6 crop mode or full frame mode and cropping to 17.5 mp, the light gathering is going to be the same.

Now, I readily admit that I don't speak the same techno-speak as others, but you get the idea.

I think this is worth clarifying because some of the "experts" on this forum make fuzzy statements that can easily imply that the mere size of the sensor means more light gathering.

 

[dtaylor]

I would say that this is only true because the smaller sensor is likely to have the pixels more tightly packed into the sensor. In other words, the size of the sensor isn't the determining factor, it's the size of the pixels. The 45mp R5 does not gather any less light when cropped to APS-C size.

Of course it does. Whether you use a lens which only projects an APS-C imaging circle, or crop to APS-C in post, you are not gathering/using as many photons. And this changes the final SNR for the image and the appearance of noise for a given view size.

I think this is worth clarifying because some of the "experts" on this forum make fuzzy statements that can easily imply that the mere size of the sensor means more light gathering.

The size of the sensor does impact light gathering, and a larger sensor means more light gathering, assuming the same light intensity per square mm (i.e. same exposure). So long as gapless microlenses are present, pixel size is practically irrelevant in the sizes we see on DSLRs and MILCs. (I haven't investigated, and cannot rule out, the possibility that size becomes a factor at even smaller pixel pitches.)

 

[Deleted]

I'm not sure what is holding you to upgrade to 90d, in my case it's better in every possible way, except two things:

1. Body is not all metal (though metal-plated), but looks sturdier than R6 IMO.
2. You cannot deep-program AF point buttons. For example, both on my 7D/mk2 and 5d4 I mapped back AF-ON to be One Shot and * button to be Servo with separate AF points (Spot AF and Zone). You cannot go that deep on 90d. That was a massive blow for me, but I learned to overcome it by using shutter button for AF (default) and AF-ON mapped as a Servo switch.
3. No dual card slots (with CRAW and high-capacity cards, I don't care for that at all)

But that's where drawbacks end for me. Speaking of its AF module, I find it improved over 7D2 and much improved over 7D. It's on par with my 5D4, slightly less precise with 50mm f1.8 STM, but then again that lens isn't the most precise one and it's sharper on FF overall as it's not pushing the optics beyond normal levels. LiveView on the other hand rivals mirroless cameras. Super fast, super smart, face recognition with eye-find focus and Focus Peaking for manual lenses.

All in all, I have effectively moved from 5d4 to 90d. With a price of 1k euros, it's a steal for such a great and modern aps-c camera.

I have a 90D, well I own one I should say. My daughter borrowed it some time back.

For me, the ergonomics do not match up to the 7D2. With a longish lens nor does the AF. I spent time using them side by side and for much of the subjects the 90D struggled more than the 7D2. The 90D is a great all rounder, but the 7D2 is better for fast subjects like swift, terns etc imo. If I ever do get my 90D out of the clutches of my youngest, it will replace my 80D as a macro body.

I do own a D500 too, won in a competition. The comparison with the 7D2 is a hard one. They are much closer than the specs suggest. Both have their advantages and drawbacks. In certain conditions with certain subjects the AF worked better on one vs the other. Ergonomics of the 7D2 were better, but I have been a canon shooter for many moons so that in part could be muscle memory. The 7D2 is certainly more programmable and having three buttons on the back for AF control vs the one on the D500 makes a big difference.

Why Nikon decided their lens mounts should not follow lefty loosey, righty tighty is beyond me. Parobably some marketing twonk trying to be different from the flock. Nothing a damn good flogging would not have sorted out.

If rumours are true I hopefully will not have long to wait for a suitable RF crop body. Hopefully it will live up to expectations. It really is a tough segment with very critical users. One huge advantage of an EVF is for video. I have shot some handheld video (some terrible examples here - https://youtube.com/channel/UCVfyNa6n6eaxlQAC6Ye40Sw showing I have a lot to learn) and it is not easy using a rear screen. I look forward to being able to use the EVF for that alone. When I used an R5 last year there were a few situations where it made life trick for stills. For instance tracking fast birds (hobby in this case) from a bright background to dark reeds the EVF was effectively unusable until it adjusted. You had yo sort of imagine where the bird was and hope it hadn’t strayed too much off path. Amazingly, despite the EVF being black, the focus tracking kept up as did the metering. Also found my shooting style of tracking things for a long time, awaiting very specific behaviour, really was a huge battery drain.Anout 4-1 ratio R5 - 7D2. I can deal with that though.

 

[Blue Zurich]

EVF of R6 is great, but I guess as someone that grew up on mechanical Nikon FM2/3a cameras, I cannot treat EVF as "normal" viewfinder. What I see in EVF is not the scene I see with my eyes (OVF). That can have ups and downs, of course. One of the ups is that what you see on EVF is 99% what you will get. But in a sense, EVF is sort of virtual-reality-viewfinder and I totally understand people that just dislike it no matter what.

I completely agree with that statement as I too grew up with those fantastic Nikon film bodies and even older RB67s. It's not true wysiwyg but my point was how they are better than ever and not the hold up for some as a reason to avoid mirrorless like I and others had in our disappointment with lesser EVF's.

 

[Lee Jay]

What he has done is to calculate the size of the Airy disk using the standard formula d/2 = 1.22*wavelength*f-number. Using green light of 500nm wavelength gives the values in his table.

Diffraction limit for the Rayleigh criterion:
cycles/mm = 1/(1.22 * wavelength * f-number)

Converting from number of cycles per unit of size to size of each cycle:

1/cycle size = 1/(1.22 * wavelength * f-number)

Solving for f-number:

f-number = cycle size/(1.22 * wavelength)

Normally, it would take 2-pixels to resolve one cycle (Nyquist). However, because the sensors we use have pixels that are not of zero size, because they are Bayer sensors and thus require demosaicing, and because they usually have AA filters, I like to use 3 pixels per cycle, rather than 2 pixels per cycle, to compensate for the loss of resolving power from these sources.

f-number = (3 * pixel size)/(1.22 * wavelength) for the Rayleigh criterion

If we plug in your numbers (500nm light and the pixels from the 90D), we get:

f-number = (3 * 3.2 microns)/(1.22 * 0.5 microns) = 15.7 (f-number for the Rayleigh criterion on 3.2 micron pixels using 3 pixels per cycle)

If you prefer extinction instead of Rayleigh, this formula works:
f-number = (3 * pixel size)/(1.00 * wavelength) for MTF=0 (extinction)

 

[Lee Jay]

When I used an R5 last year there were a few situations where it made life trick for stills. For instance tracking fast birds (hobby in this case) from a bright background to dark reeds the EVF was effectively unusable until it adjusted. You had yo sort of imagine where the bird was and hope it hadn’t strayed too much off path. Amazingly, despite the EVF being black, the focus tracking kept up as did the metering. Also found my shooting style of tracking things for a long time, awaiting very specific behaviour, really was a huge battery drain.Anout 4-1 ratio R5 - 7D2. I can deal with that though.

Dang. Those things, the lack of dynamic range in the viewfinder, and viewfinder lag are exactly what make EVFs unusable for me. Sounds like those are not fixed in the R5, despite the claims of others.

 

[David - Sydney]

MILC will get better with respect to battery life, but will never match DSLR. They are just doing so much more computing. Eventually batteries of higher capacity will get smaller, which will help, along with incremental improvements in chip efficiency.

There hasn't been a significant change from Li-ion battery technology in small devices for some time unfortunately. Canon has been able to fit a little more capacity in the same LP-E6 form factor but hasn't done the same for the LP-E19 form factor with the R3 release which is surprising.

Chip (and algorithm) efficiency is the main way to improve battery life as Apple's M1 and RISC/Arm have demonstrated. Sony does appear to be in the lead here for camera systems as for approximately the same battery capacity, they have greater CIPA ratings than Canon does (with smaller bodies either managing heat better or prematurely cooking their chips).

Mirrorless will be the same or better battery life than DLSR using live view opposed to OVF only. You may ask why someone would use a DLSR predominately via live view but the 1DXiii does have better better AF tracking via live view and 20fps vs OVF even if the ergonomics aren't great.

 

[dtaylor]

Normally, it would take 2-pixels to resolve one cycle (Nyquist). However, because the sensors we use have pixels that are not of zero size, because they are Bayer sensors and thus require demosaicing, and because they usually have AA filters, I like to use 3 pixels per cycle, rather than 2 pixels per cycle, to compensate for the loss of resolving power from these sources.

I doubt it's that high based on extinction resolutions published by Imaging Resource when they used to do that level of testing. But I do agree it has to be >2. In my own estimations I tend to use 2.2, though I admit I grabbed that value from the domain of audio sampling. I have not performed, nor seen, any tests that would firmly establish if it should be 2.2 or 3 or 2.5...

Whatever the exact numbers are, there's something to this because in the real world the diffraction limit is observably not a hard and fast limit. Higher pixel pitch cameras continue to resolve more detail a couple stops beyond their supposed limit. I'm sure target contrast, and how steeply it falls off towards the airy disk edge, comes into play as well because if there's enough contrast the detail can still be resolved.

 

[unfocused]

Of course it does. Whether you use a lens which only projects an APS-C imaging circle, or crop to APS-C in post, you are not gathering/using as many photons. And this changes the final SNR for the image and the appearance of noise for a given view size.

The operative term here is "for a given view size."

 

[neuroanatomist]

I can't believe that after all these years and endless debates we are on the same page. I'm glad you finally came around.

Seriously, I think we have both been correct just not necessarily communicating clearly.

Well, I haven’t changed my correct statements to false ones and the concept of equivalence hasn’t changed. So if you think we’re on the same page now……

However, I do question one statement (or at least want to clarify it).

"The smaller sensor means less total light collected, which means effectively 1.3 stops more noise."

I would say that this is only true because the smaller sensor is likely to have the pixels more tightly packed into the sensor. In other words, the size of the sensor isn't the determining factor, it's the size of the pixels. The 45mp R5 does not gather any less light when cropped to APS-C size. Of course it is no longer 45mp, but about 17.5 mp. Whether you are shooting in 1.6 crop mode or full frame mode and cropping to 17.5 mp, the light gathering is going to be the same.

Now, I readily admit that I don't speak the same techno-speak as others, but you get the idea.

Ladies and gentlemen, we now return you to your regularly-scheduled programming.

No. It has nothing to do with pixel density. However, you have changed the terms. I was talking about comparing a FF picture to an APS-C picture. You are talking about cropping a FF image to the area of an APS-C image. In that case, the images, FoV, DoF and noise are identical.

I think this is worth clarifying because some of the "experts" on this forum make fuzzy statements that can easily imply that the mere size of the sensor means more light gathering.

There’s nothing fuzzy about it, it’s basic physics. A larger sensor gathers more light. The oft-used analogy is water – put a teacup and a bucket out in the same rainfall (the light coming from the scene), leave them out for the same amount of time (shutter speed), and the bucket will collect more water than the teacup. As I said to someone else, if you argue with physics, physics will win. Every. Single. Time.