Canon R7 owners : are you happy with your high iso shots?

Younes B

Senior Member
Messages
1,257
Solutions
2
Reaction score
475
Location
Montreal, CA
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
thanks for your input, I'll check the site you mentioned.

I was hoping one could get away with NR in Lightroom only, without an additional NR software.

I need to see more pics.

The concern I had was that given the current lack of fast teles in RF-s mount, you have to shoot with consumer teles with a smaller AV and hence higher ISO if you need speed to freeze action. High iso performance becomes key....

I also have to admit that I do look at my pics at 100%....and you buy 33 MP to be able to crop as well so 100% pixel peeping is not just a fantasy.
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
Are you shooting RAW or CRAW?

CRAW tend to add more noises depending the conditions.
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
Are you shooting RAW or CRAW?

CRAW tend to add more noises depending the conditions.
Pssst. I don't think the OP has the R7.

R2
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
thanks for your input, I'll check the site you mentioned.

I was hoping one could get away with NR in Lightroom only, without an additional NR software.
I guess one could, though with the R7 in particular I've become wedded to 3rd party software.
I need to see more pics.
Agreed.
The concern I had was that given the current lack of fast teles in RF-s mount, you have to shoot with consumer teles with a smaller AV and hence higher ISO if you need speed to freeze action. High iso performance becomes key....
I use the RF mount and yes, the lack of long fast teles has been a problem Supposedly, a bunch are on the way. It would be helpful to understand what you shoot?
I also have to admit that I do look at my pics at 100%....and you buy 33 MP to be able to crop as well so 100% pixel peeping is not just a fantasy.
Yes, one does. Since I shoot a lot of WL/BIF, more pixels on target and reach are critical. The R7 has proved to be a performer and while it's not my first choice, it is a great choice for travel, hiking, etc. where I can pair it with the compact 100-500. On a fast prime, I prefer my R5. Here's an interesting read for you from Bryan over at The Digital Picture which speaks more to resolution than noise. Let's Talk About Reach. Does an APS-C Format Imaging Sensor Increase Reach? Specific Canon EOS R7 vs. R5 Comparison (the-digital-picture.com). Have you looked at the various studio rendering comparisons between different bodies?
 
Are you shooting RAW or CRAW?

CRAW tend to add more noises depending the conditions.
Pssst. I don't think the OP has the R7.

R2
Indeed, I don't have it yet but I am seriously considering it as it is within my budget and has outstanding AF, but I have concerns wrt high ISO perf.

I am wandering if a lower count MP R7 may have been better ( less noise ). After all, the R6 is only 20 MP..
 
Last edited:
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
Are you shooting RAW or CRAW?

CRAW tend to add more noises depending the conditions.
I have R7 and R5. I am delighted with the ISO noise from the R7. It is slight worse than the R5 of course but it seems to be lest than on f stop more noise. In any case the DXO and Topaz denoising work well so I am happy.

Before the R7 in my own hands I saw the ISO horror show of some the sample images but real world use along side the R5 showed how decently the R7 performs IMO. It is an APS-C sensor but a good one. I try to stay below ISO 6400 when possible. I think it is very similar to the 7DII.
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
I try to keep my ISOs down when using the R7 (I save lower light for FF generally).

I did however test the R7 out in bad weather/light conditions just after I got it to see how it would do (esp the AF). ISOs were from 800 - 3200. Samples here are 2160px (click on "original size")...

https://www.dpreview.com/forums/thread/4668613

R2
 
I am wandering if a lower count MP R7 may have been better ( less noise ). After all, the R6 is only 20 MP..
downsampling will get you there easily. Then you can look 100% at the result
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
I have noticed the same thing.
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
I don't care for how these programs handle noise in many shots I have seen. I don't use them so maybe some people get too aggressive with the settings. Some shots have weird artifacts and the bokeh looks abnormal.
 
Last edited:
It’s about a stop behind full frame in real world use. Exactly what you’d expect, given the size difference. I’m comparing it to the R. With good processing, ISO 12800 is eminently usable. I don’t know what samples you’ve looked at, but if they’re unacceptably noisy at 1600 or 3200, it’s almost certainly down to bad exposure and/or bad processing.
 
Are you shooting RAW or CRAW?

CRAW tend to add more noises depending the conditions.
Pssst. I don't think the OP has the R7.

R2
Indeed, I don't have it yet but I am seriously considering it as it is within my budget and has outstanding AF, but I have concerns wrt high ISO perf.
Just from curiosity. Did you try to check in detail other 30Mpx+ cameras, if you like their high ISO performance?
I am wandering if a lower count MP R7 may have been better ( less noise ). After all, the R6 is only 20 MP..
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
I don't care for how these programs handle noise in many shots I have seen. I don't use them so maybe some people get too aggressive with the settings. Some shots have weird artifacts and the bokeh looks abnormal.
Yes, all of your points are valid in some respect though the software has saved images which otherwise were unusable. For example, I shot this Sandhill Crane with severe backlighting and though I used 1 2/3rds stops of EC had to do a fair amount of PP. Mind you this was shot at an ISO of 10,000 so yes, the original was noisy. Will I make 24x36 images of any of these series and put them on the wall? No the subject nor the execution warrant it, yet the results are fine for web posting (this is a low resolution 1900k, 1.5MP image).

a11a5f68d9b6457fbd3e0777e17cf2d1.jpg
 
Last edited:
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
It's not noisier than expected. Go to photons and compare the SNR, DR, etc.
Are these outliers due to bad use, atmospheric conditions or else?
More likely underexposure. This sensor demands ETTR.
I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Have you seen the output from a Sony A7RIV or R5? For the unaccustomed, the R7 does produce a bit of noise, however it is easily mitigated in Topaz, Noise AI, Deep Prime, etc. Obviously, it isn't my first choice of bodies for a studio, though it performs really well where one wants a cropped sensor and relatively many pixels on target. What additional information do you need?
I don't care for how these programs handle noise in many shots I have seen. I don't use them so maybe some people get too aggressive with the settings. Some shots have weird artifacts and the bokeh looks abnormal.
Yes, all of your points are valid in some respect though the software has saved images which otherwise were unusable. For example, I shot this Sandhill Crane with severe backlighting and though I used 1 2/3rds stops of EC had to do a fair amount of PP. Mind you this was shot at an ISO of 10,000 so yes, the original was noisy. Will I make 24x36 images of any of these series and put them on the wall? No the subject nor the execution warrant it, yet the results are fine for web posting (this is a low resolution 1900k, 1.5MP image).
The image you provided is one of the good examples and impressive for the ISO it was shot at. As for the R7's noise levels, this might be the tradeoff for the pixel density of the 32mp sensor along with tech Canon is using to manufacture it. I am curious to see how the upcoming Fuji 40mp sensor fares in this regard.
 
I'm very happy with high ISO shots, e.g. above 3200, from my R7. They are noisy of course but respond very well to Topaz DeNoise AI.

Bramble9
 
Adam2; I really like the colors in this shot.

Bramble9
 
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.

Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).

Any input?
The trick increasingly with high density sensors seems to be to avoid using Lightroom for RAW conversion at high iso. Some shots we see won’t be processed optimally:

DPP, in camera jpeg and particularly DXO Deep prime (with non excessive setting in PL5) work wonders vs Lightroom in my view as of late 2021 on modern sensors. I don’t like PureRAW though as can’t control it and can look plasticy.

TopazAI sharpen again with care then can tease out great high iso pictures - but care needed so not to look unnatural- and generally hard to get results as dramatic as on their web examples:
 
Last edited:
Hi guys,

Not trying to offense anyone here but I have seen quite a few pics coming out of the R7 at ISO 800 or higher than are noisier than one would expect from a current generation sensor.
Well with such a small pixel size you can hardly expect better results.
Are these outliers due to bad use, atmospheric conditions or else?

I know it is comparing different sizes but we have come to expect ISO 3200 clean shots nowadays from full frame sensors
R7 is not a FF.
and even my 7 years old K5 has clean output at that ISO ( though only 16MP ).
Mimd the pixel size,
Any input?
 

Keyboard shortcuts

Back
Top