Dynamic range testing of the Canon EOS R3 is complete

[Canon Rumors]

The folks at Photons to Photos have completed their dynamic range testing of the Canon EOS R3 sensor. The dynamic range is pretty comparable to the Canon EOS R5.
Canon EOS R3 vs Canon EOS R5 Dynamic Range
Head on over to Photons to Photos to learn more.

Continue reading...

 

[Niko Todd]

In ES mode, the R3 is much better:

https://www.photonstophotos.net/Charts/PDR.htm#Canon EOS R3(ES),Canon EOS R5(ES)

 

[H. Jones]

The best part of contributing data to this is finding out the results! Very very cool to see that the ES is very, very close to the mechanical shutter. That matches my real-world experience with shooting in ES, where I haven't noticed notable differences in dynamic range. ISO 100 Electronic shutter has more dynamic range than ISO 159 mechanical shutter, this is really not a difference I'm all that worried about whatsoever.

This is a camera I am totally happy to leave in ES and never click the physical shutter once. Excellent all around.

 

[Sharlin]

So… rather than a single big jump in gain, we get two smaller ones, which gives the R3 an edge at very high ISOs (which makes sense in a sports/WL body). At around third of a stop it’s not an insignificant edge either!

 

[Sporgon]

Is anyone able to quantify the noise reduction on the R5 ? And in time the R3 which looks like it uses it all the way up the ISO. Kind of jars a bit with me; I can't understand a 24 mp sensor having the same high ISO with NR applied as a 45mp one without ? Given equal generations.

Bill Claff himself suggested around two thirds of a stop. Is it possible to achieve the same improvement in DR through noise reduction in post ?

 

[Nelu]

Continue reading...

"The dynamic range is pretty comparable to the Canon EOS R5."
So, a 24MP sensor has pretty much the same dynamic range as a 45MP sensor.
Good job, R5!

 

[DBounce]

This falls inline with what I have noticed with my R3 also. Excellent blend of dynamic range, great high ISO performance and speed.

 

[Tremotino]

The best part of contributing data to this is finding out the results! Very very cool to see that the ES is very, very close to the mechanical shutter. That matches my real-world experience with shooting in ES, where I haven't noticed notable differences in dynamic range. ISO 100 Electronic shutter has more dynamic range than ISO 159 mechanical shutter, this is really not a difference I'm all that worried about whatsoever.

This is a camera I am totally happy to leave in ES and never click the physical shutter once. Excellent all around.

Why should dynamic range be lower when using ES?!

 

[Billybob]

Not a surprise. There are no major improvements in IQ with current tech. It's good to see that the DR didn't drop significantly at low ISO like the original Sony A9 does. I see that Canon is still "cooking" RAW files at low ISO. I suspect that Nikon's Z9 will be similar in this respect. Anyway, no story here. Since IQ is pretty similar across cameras and companies, let's see how AF performance compares.

 

[Tremotino]

Does anyone know the DR of Canon's new 3MP SPAD Sensor? I would love to see a sample photo too!
It should go to mass production next year for autonomous cars and surveillance.

Canon rumors has brought no rumors at all in the last view weeks and months. Just news that are official or pure wishthinking or speculation.
Where are the 2 new lenses comming until the end of this year?

 

[H. Jones]

Why should dynamic range be lower when using ES?!

This is a question for someone with far more technical knowledge than I have, but I believe from past things I've read, the reason for it is that without the mechanical shutter closing after each exposure, stray light can impact the image following the exposure, introducing slightly more read noise. That's why you see the option for "Electronic First Curtain shutter," which still has a secondary mechanical shutter, and does not impact DR at all. That way, with the mechanical shutter closed, there is no additional light to generate read noise/heat on the sensor circuits for the brief milliseconds it takes to process the exposure.

Obviously there are ways to manage this, since Canon has improved the ES dynamic range in every release of a camera, but it's still slightly visible in the EOS R3 charts. It really doesn't make a difference at this point, though. Previously, Canon was also limiting Electronic shutter on the R5/R6 to 12-bit readout, but I'm sure with all the marketing around the uncompressed raw at 30 fps, they aren't limiting that anymore.

 

[privatebydesign]

"The dynamic range is pretty comparable to the Canon EOS R5."
So, a 24MP sensor has pretty much the same dynamic range as a 45MP sensor.
Good job, R5!

When the result is normalized, which it is, why wouldn’t they be the same?

 

[angelisland]

In ES mode, the R3 is much better:

Photographic Dynamic Range versus ISO Setting

View attachment 201598

I would call that slightly better at some ISOs, not "much better."

 

[Joules]

Why should dynamic range be lower when using ES?!

Most implementations of electronic shutter we've seen so far use 12 bit instead of 14 bit to increase the sensor read speed. Or reduce rolling shutter if you prefer.

But a file with 12 bit per channel loses two stops of dynamic range if the data is linear, compared to a 14 bit one. That's purely in terms of the number range - in the real world noise means the DR is less than the file permits. So it is a tradeoff.

The stacked sensor from the R3 and Z9 provide better read out speeds by design, so they don't have to rely on this tradeoff to allow the photographer to take advantage of the benefits of electronic shutter.

 

[AlanF]

When the result is normalized, which it is, why wouldn’t they be the same?

Absolutely true: the DR is calculated for the same output size, roughly equivalent to a 10"x12" print held at arms length.

 

[bclaff]

Why should dynamic range be lower when using ES?!

Reading pixels in electronic shutter (ES) mode has more noise than in mechanical shutter (MS) mode therefore dynamic range is reduced.
FWIW, unlike certain other cameras, ES mode is still 14 bit it's not reduced to 13-bit or 12-bit.

 

[bclaff]

Is anyone able to quantify the noise reduction on the R5 ? And in time the R3 which looks like it uses it all the way up the ISO. Kind of jars a bit with me; I can't understand a 24 mp sensor having the same high ISO with NR applied as a 45mp one without ? Given equal generations.

Bill Claff himself suggested around two thirds of a stop. Is it possible to achieve the same improvement in DR through noise reduction in post ?

I wish there was a better way to quantify the effect of signal processing but I have not found one.

The 2/3 stop figure you cite is an estimate for the R5 at base ISO.

 

[bclaff]

Absolutely true: the DR is calculated for the same output size, roughly equivalent to a 10"x12" print held at arms length.

Small point, I use 8" on the short side. So in this case 8"x12"

 

[bclaff]

... I see that Canon is still "cooking" RAW files at low ISO. I suspect that Nikon's Z9 will be similar in this respect. ...

We will see but I doubt it. No Nikon to date has exhibited this type of signal processing at low ISO settings.

 

[AlanF]

Small point, I use 8" on the short side. So in this case 8"x12"

Great to see you here!