Dynamic range testing of the Canon EOS R3 is complete

[SereneSpeed]

Am I reading it correctly that 200 iso for the R5 has less DR than 400 iso? If so, That suits my studio shooting style beautifully!

Anyone want to give me the Coles Notes version to explain that?

 

[MythPlayer]

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 ?

If raw image source from dpreview's publish，you will findout [High ISO Noise Reduction - Standard] in EXIF data.
IS same setting in dpreview's EOS R raw publish.
BUUUUUT~I remember this setting can set as DISABLE in my EOS R!

 

[SereneSpeed]

And...

Is ISO 50 actually that good? I thought I read that IQ suffered when using the expanded ISOs?

 

[AlanF]

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

Actually, Bill, your website has:
"PDR is the dynamic range you would expect in an 8x10" print viewed at a distance of about arms length." But, I presume you adapt this size as necessary.

Sensor Analysis Primer –Photographic Dynamic Range Summary

 

[bclaff]

Am I reading it correctly that 200 iso for the R5 has less DR than 400 iso? If so, That suits my studio shooting style beautifully!

Anyone want to give me the Coles Notes version to explain that?

For the R5 ISO 400 tests slightly higher in PDR than ISO 200 in part due to something called Dual Conversion Gain. But the difference is quite small.

 

[bclaff]

And...

Is ISO 50 actually that good? I thought I read that IQ suffered when using the expanded ISOs?

You would expect ISO 50 to test identically to ISO 100 since extended ISO 50 is identical to ISO 100 with 1 stop of exposure compensation.

 

[Sporgon]

I didn’t realise that in the PDR the results are normalised, so I can now understand why the reduced 45 mp competes with the 24mp and NR. However that leaves me surprised that there’s such a difference between FF and crop.

 

[bclaff]

Actually, Bill, your website has:
"PDR is the dynamic range you would expect in an 8x10" print viewed at a distance of about arms length." But, I presume you adapt this size as necessary.

Sensor Analysis Primer –Photographic Dynamic Range Summary

Sure, or maybe as 8"x10" trimmed from an 8"x12"; the important point is 8" on the shorter side.
Regards

 

[bclaff]

I didn’t realise that in the PDR the results are normalised, so I can now understand why the reduced 45 mp competes with the 24mp and NR. However that leaves me surprised that there’s such a difference between FF and crop.

The larger area sensor requires less enlarging to get to the standard viewing conditions (8" side short for example) so noise is less apparent and PDR is higher.

 

[docsmith]

Am I reading it correctly that 200 iso for the R5 has less DR than 400 iso? If so, That suits my studio shooting style beautifully!

Anyone want to give me the Coles Notes version to explain that?

And it is interesting to see the R3 having three steps, indicating potentially "triple gain conversion"....

As I understand it, dual gain conversion implies that you have separate circuitry optimized for different ISO ranges, instead of one size fits all.

 

[docsmith]

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

Thanks for this feedback. Until today, I had missed the fact that Canon was applying NR to RAW files entirely. Do we know what this looks like in practice?

 

[raptor3x]

If raw image source from dpreview's publish，you will findout [High ISO Noise Reduction - Standard] in EXIF data.
IS same setting in dpreview's EOS R raw publish.
BUUUUUT~I remember this setting can set as DISABLE in my EOS R!

It's not, Bill's data comes from volunteers who go through a protocol he defines that define a series of images from the camera in question.

 

[bclaff]

And it is interesting to see the R3 having three steps, indicating potentially "triple gain conversion"....

As I understand it, dual gain conversion implies that you have separate circuitry optimized for different ISO ranges, instead of one size fits all.

Dual Conversion Gain is an extra transistor and capacitor in the pixel. The transistor controls whether the capacitor is in the circuit or not.

That 2nd bump is probably increased noise reduction; it's not triple conversion gain.

 

[slclick]

Never known a client or visitor at a gallery to be wowed by spec sheets so.....

R3 shooters, how are your images? I'd like to think you'd be pleased?

 

[Billybob]

Never known a client or visitor at a gallery to be wowed by spec sheets so.....

R3 shooters, how are your images? I'd like to think you'd be pleased?

I don't have an R3, but I've seen plenty of images and they're beautiful. I think most are simply fascinated to see this data and are happy that the DR really doesn't drop in ES mode.

Bill, your work is most definitely appreciated.

 

[slclick]

I don't have an R3, but I've seen plenty of images and they're beautiful. I think most are simply fascinated to see this data and are happy that the DR really doesn't drop in ES mode.

Bill, your work is most definitely appreciated.

Well, to me and many others I know not into the left brain side of art and photography it comes across as mental masturbation and incredibly dull but hey, I'd never ask for everyone to be just like me so carryon.

 

[Billybob]

Well, to me and many others I know not into the left brain side of art and photography it comes across as mental masturbation and incredibly dull but hey, I'd never ask for everyone to be just like me so carryon.

For me, it's force of habit. I had a 40D but picked up a Nikon D90 after seeing its much higher DXOMark score. Both shot beautiful images at base ISO, but I soon realized that the Nikon files were indeed more malleable. Plus, back then the Canon crop-sensor cameras really started falling apart as ISO went up. It wasn't so much increased noise as it was a loss of color fidelity. Higher ISO images--as low as 1600 on my D7--looked almost faded. The Nikon, by contrast, retained color integrity at much higher ISO values (e.g., D90 ISO 6400 was quite useable with a good dose of NR). There were things about Canon that I loved too much to switch--like the lenses for example--so I shot both for years.

However, today there is very little difference in DR and noise levels between brands. Hence, I no longer make buying decisions based on those characteristics (however, I guess I am still haunted by MP levels). Nonetheless, I always look forward to seeing Bill's analysis of a major camera release. To me, it's purely entertainment. Now I'm sitting back and waiting to see how the Nikon Z9 fares on these fronts. I have my popcorn ready.

 

[H. Jones]

Never known a client or visitor at a gallery to be wowed by spec sheets so.....

R3 shooters, how are your images? I'd like to think you'd be pleased?

The EOS R3 is easily the nicest camera I've ever used. Just overall the speed and stability of everything in the camera makes it an absolute pleasure to use. Nothing ever slows down, no frames ever get skipped in the EVF, no blackout at all, never hit the buffer, it's just super responsive and a pleasure to take photos with. The EVF unit is an upgrade that's hard to explain, since it has more latitude latitude to show highlights and shadows than the R5. Battery life is pretty close to double my R5's battery life, too. I never notice rolling shutter, far better performance than the R5's rolling shutter, and it's really noticeable.

24mp or not, it's a better experience than my R5 or my 1dx mark II combined. Hard to imagine how cameras could get much better than this.

 

[Chris.Chapterten]

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 believe super fast readout, stacked sensors like the R3 has are not quite as good as a slower readout sensor when it comes to DR and noise performance.

Compare Sony A9ii to Sony A7Riii or iv and I believe the slower reading, higher MP sensors also have more dynamic range.

It’s actually a great achievement that the sensor of the R3 can match the image quality of the R5 IMO.

I’m sure someone will correct me, if these assertions are not quite right?

cheers

 

[perplex1]

are these DR tests for stills only? Would this also apply to DR in video as well?