R5+100-500+1.4x or R7+100-500?

drsnoopy wrote:

I have the R5 and 100-500, and very happy with the combination. I also have an R7 on order but now I'm wondering whether anyone has done a comparison between the R5 with the 1.4xTC (giving 700mm max FL) and the r7 with the bare lens (giving 800mm effective max FL). I do understand the issue with the TC restricting the short end of the zoom range. I've also tried crop mode on the R5 (17MP) and the results are mostly good but would like the extra sensor resolution. My feeling is that the r7 should have the edge, bearing in mind of course the differences in noise, which I am happy to deal with in RAW conversion (DXO) or in post (Topaz). However the TC would be less expensive...

Thanks for any examples or pointers towards other sources.

The R7 has less noise than a 1.6x crop from the R5.  It may seem otherwise due to two sources of illusion; 100% pixel views, and the bias that converter defaults have towards accentuating noise with smaller pixels, by sharpening them more, even when the final display target can't even show original pixel-level detail.  Even the R6, which has less noise than the R5, is no better in crop mode for noise than the R7.  The R7 has Canon's best recording in an APS-C sensor area, period.  It has about the same visible noise as the R6, but about 4x the pixels-on-subject, and less noise than the R5, with about 2x as many pixels-on-subject.

A TC on the R5 or R6 would get more pixels-on-subject, but the noise differences between the cameras don't really change in any positive way like that, because to get the same pixels-on-subject with the FF cameras, you need a 1.4x on the R5, which will double the ISO, and a 2x on the R6, which will quadruple the ISO (with the same shutter speed and pupil size).  The TCs add a small bit of their own aberration, and that makes details punch through the noise a little less than with the R7 with no TC.  Also, any benefit of the R5 or more so R6 for "low light AF sensitivity" is lost to the TCs solutions.  The fact is, when normalized for pixels-on-subject with TCs, the R5 and R7 have the same Canon-claimed sensitivity, and the R6 is a 1/2 stop less sensitive.  IOW, -6.5EV, -6EV and -5EV become -4.5EV (R6), -5EV (R5), and -5EV (R7).  So, the fact that the R7 will AF more slowly in low light with the same lens becomes irrelevant when you start biasing things with TCs.

As far as the R3 is concerned, despite its amazing speed by 2022 standards, it is also noisier than the R7, similar to the R5, when cropped or when using TCs.

R2D2 wrote:

John Sheehy wrote:

I subscribe to this assessment, esp when birding.

However I've noticed that whenever I'm not maxxing out the crop, the R5 clearly pulls ahead of the R7 in the IQ department.

If the R7's entire frame has less noise than a 1.6x crop from the R5, then any smaller sensor area used by both will also have the same ranking. Noise per unit of sensor area does not vary with sensor area.

The R7's advantage exists only in a very narrow set of usage cases (and that's only considering IQ, and not the other performance factors). In short, the R5 is clearly turning out to be the better wildlife/birding body for me (now that I've gotten to shoot the R7 more). YMMV

Or my mileage may be your kilometerage.

There's another illusion that I didn't mention, because it does not involve exact comparisons of usage. There is also the confounder effect of which camera you use for which job; if you shoot the R5 whenever you can fill the frame well, but shoot the R7 when you are more focal-length-limited, then most of your R7 images will be of smaller and/or distant subjects, which will tend to be noisier for reasons that have nothing to do with sensor efficiency with noise.

Same kind of thing with TCs; people will use the TC more often when the subject is smaller and/or more distant, and blame the TC, when all the TC did was add a slight amount of aberration. Distance and small subjects are a PITA for IQ, and I don't think that we need to blame that on the tools that we use to make the most of them.

The farther away the subject is, and the smaller it is, the faster the shutter speed we need, and the less light from the subject we get with the same shutter speed and pupil size.

The only way the R7 forces more noise on you is if your lens is too long, and there is no TC to take off, so you have to back away from the subject because of the narrower angle of view.

drsnoopy wrote:

I have the R5 and 100-500, and very happy with the combination. I also have an R7 on order but now I'm wondering whether anyone has done a comparison between the R5 with the 1.4xTC (giving 700mm max FL) and the r7 with the bare lens (giving 800mm effective max FL). I do understand the issue with the TC restricting the short end of the zoom range. I've also tried crop mode on the R5 (17MP) and the results are mostly good but would like the extra sensor resolution. My feeling is that the r7 should have the edge, bearing in mind of course the differences in noise, which I am happy to deal with in RAW conversion (DXO) or in post (Topaz). However the TC would be less expensive...

Thanks for any examples or pointers towards other sources.

I have been trying to answer this question for several weeks, making several thousand frames in the process, mostly of birds in and out of trees and bushes, but also set static landscape type subjects.

My first conclusion is that either R7 + 100-500 or R5 + 1.4x  + 100-500 can be used to make excellent pictures and that getting experience with the equipment is probably more important than  the this-or-that choice.

Price: the R7 + 100-500 obviously wins but only if one does not also have an R5.

Convenience: The R7 kit wins here also. It is  smaller and the full zoom range of the lens is available.

Image quality: On static subjects the R7+100-500 combination wins by a very small margin, observable only by pixel peeping at 100% on screen.

Autofocus: This is by far the most important issue and also unfortunately where  things get complicated. Compared to R5 + 100-500 without the extender, both the R7 + 100-500 and R5 + 1.4x + 100-500 have less reliable AF. The loss of AF capability is worst in low light and in dappled light when both can struggle to acquire focus on the subject which in my tests is usually a bird.

Having acquired the subject it seems to me that the R5 holds onto it more tenaciously than the R7 which is prone to drifting off focus. The R7 is also more prone to focus drifting in bright light and with relatively easy subjects. It will usually come back during a burst but not always.

I use Servo AF Case 1 with tracking sensitivity and Accel/decel tracking at default for both cameras.

I suspect more insights on this question will emerge over time as more users have experience with both kits.

Andrew

R2D2 wrote:

John Sheehy wrote:

R2D2 wrote:

John Sheehy wrote:

I subscribe to this assessment, esp when birding.

However I've noticed that whenever I'm not maxxing out the crop, the R5 clearly pulls ahead of the R7 in the IQ department.

If the R7's entire frame has less noise than a 1.6x crop from the R5, then any smaller sensor area used by both will also have the same ranking. Noise per unit of sensor area does not vary with sensor area.

I'm not talking about a smaller sensor area, but a larger one, and filling up a larger one.

That isn't what you wrote, at least not clearly.  What you wrote, in gold text above, is that you're maxing out the R7 frame and comparing it to the R5, and you find the R5 to have less noise, but you didn't state anything that would change other than the camera, and the context was "focal length limited", so the only way I can read it is that you meant that the R5, in crop mode, and AOTBE, has less noise than the R7.  That is demonstrably false; every exposure-controlled comparison shows that the R7 has about 1/2 stop less visible read noise in the shadows of high ISOs.  If you're talking about the fact that you get less noise when you fill the R5 frame than the R7 frame with the same exposure and ISO, then of course that's true, but that is really about different photo-ops, isn't it?

The R7's advantage exists only in a very narrow set of usage cases (and that's only considering IQ, and not the other performance factors). In short, the R5 is clearly turning out to be the better wildlife/birding body for me (now that I've gotten to shoot the R7 more). YMMV

Or my mileage may be your kilometerage.

There's another illusion that I didn't mention, because it does not involve exact comparisons of usage. There is also the confounder effect of which camera you use for which job; if you shoot the R5 whenever you can fill the frame well, but shoot the R7 when you are more focal-length-limited, then most of your R7 images will be of smaller and/or distant subjects, which will tend to be noisier for reasons that have nothing to do with sensor efficiency with noise.

Same kind of thing with TCs; people will use the TC more often when the subject is smaller and/or more distant, and blame the TC, when all the TC did was add a slight amount of aberration. Distance and small subjects are a PITA for IQ, and I don't think that we need to blame that on the tools that we use to make the most of them.

The farther away the subject is, and the smaller it is, the faster the shutter speed we need, and the less light from the subject we get with the same shutter speed and pupil size.

The only way the R7 forces more noise on you is if your lens is too long, and there is no TC to take off, so you have to back away from the subject because of the narrower angle of view.

You may call the entirety of my experience (with both cameras) an illusion, but "that's why they play the game."

I only suggested that you could have come to the conclusion that the R5 in crop mode has less noise than the R7 without anything that could prove it.  If you were talking about using the entire R5 sensor, then you should have said so.  This is a focal-length-limited thread.

R2D2 wrote:

You missed the important word “not” in my (now purple) post above. When NOT maxxing out the crop.

See, we’ve both actually been in agreement.

I'm still not sure what you're saying.  If you're not "maxing the crop," with the R7, that should mean that even the R7 has an excessively wide angle of view, and will be cropped, with the R5 losing a greater percentage of its sensor area for the final crop of the same photo-op.

Are you, or are you not, saying that the R5 has less final-image-level noise using the same ISO and exposure, when using the same sensor area in mm x mm to form an image?

If so, I don't think that it can be shown in a controlled test, unless it is done wrong, like over-sharpening the R7 at the pixel level, or viewing at 100% (which is a different subject magnification for the two cameras).

Gaber1 wrote:

Hallo !

I have done a test with a chart r5 with 100 500 at 500 r5 with 100 500 and 1.4 x and R7 with 100 500

All shots were taken from a sturdy tripod in electronic mode shooting with the phone app to avoid vibration. focal lenght 500 1/25 maximum aperture same WB and tint

ON my opinion the results are:

R5 crop vs R7 wins R7 by a thin hair

R5 crop vs R5 + 1.4 x wins R5 and 1.4 x

R5 + 1.4 vs R7 to me negligible difference maybe a litle better R7 but i'm not shure

Yep, I have both combinations, and I mostly use the R7. It's more convenient because it has the full zoom range, and in real world use there is little or no advantage in choosing the R5. The one opposing issue is rolling shutter of course. Plus with the R5 you have the option of removing the TC for the full frame advantage, if the subject is within range.

For anyone who is buying and has to choose between them, the cost difference is massive, for very little performance difference.

You can check the image that I took in this other post in dpreview

https://www.dpreview.com/forums/post/66894370

Ciao

Gabriele