R7 vs. R6 II - Sensor Size

[DNBush]

Just in terms of the sensors, 32.5mp APS-C (R7) and 24.2mp FF (R6 II), which would generally be better in terms of IQ (low light, high ISO, DR, etc.)?

 

[EduPortas]

Your loss. I have recently bought the Sigma 10-18 and 18-50 for my R7. They are both really nice lenses. Better than Canon's RF-S 10-18 (which is surprisingly good for the price) and 18-45. I have kept my RF-S 18-150 for the extra reach, but the Sigma 18-50 is better for everything up to 50mm. Also, the Sigma 56 F1.4 is a simply superb lens. For a while it was my most used lens on my M6II. I might get the RF version for my R7. The 16 F1.4 is a great lens too, but the size meant I didn't use it much on my M6II.

I'm sure all of them are technically good lenses. Clinical, but good.

I try to keep my lenses first-party just for convinience. The idea of Canon updating some firmware and making third-party lenses obsolete or wonky is always a real posiblity.

 

[BirdShooter7]

Your loss. I have recently bought the Sigma 10-18 and 18-50 for my R7. They are both really nice lenses. Better than Canon's RF-S 10-18 (which is surprisingly good for the price) and 18-45. I have kept my RF-S 18-150 for the extra reach, but the Sigma 18-50 is better for everything up to 50mm. Also, the Sigma 56 F1.4 is a simply superb lens. For a while it was my most used lens on my M6II. I might get the RF version for my R7. The 16 F1.4 is a great lens too, but the size meant I didn't use it much on my M6II.

I'm sure all of them are technically good lenses. Clinical, but good.

I try to keep my lenses first-party just for convinience. The idea of Canon updating some firmware and making third-party lenses obsolete or wonky is always a real posiblity.

Maybe I’m wrong but it was my understanding that the Sigma RF mount lenses are licensed by Canon so this shouldn’t be an issue anymore.

 

[Alastair Norcross]

Your loss. I have recently bought the Sigma 10-18 and 18-50 for my R7. They are both really nice lenses. Better than Canon's RF-S 10-18 (which is surprisingly good for the price) and 18-45. I have kept my RF-S 18-150 for the extra reach, but the Sigma 18-50 is better for everything up to 50mm. Also, the Sigma 56 F1.4 is a simply superb lens. For a while it was my most used lens on my M6II. I might get the RF version for my R7. The 16 F1.4 is a great lens too, but the size meant I didn't use it much on my M6II.

I'm sure all of them are technically good lenses. Clinical, but good.

I try to keep my lenses first-party just for convinience. The idea of Canon updating some firmware and making third-party lenses obsolete or wonky is always a real posiblity.

Maybe I’m wrong but it was my understanding that the Sigma RF mount lenses are licensed by Canon so this shouldn’t be an issue anymore.

That's correct. And I have no idea what EduPortas means by "clinical, but good". It sounds like the kind of thing you say when you want to insult a lens, but have no basis for doing it. I've seen other people criticize lenses as "clinical", and it never makes any sense. Perhaps it's the flip side of saying that a lens has "character" when it's actually not a very good lens.

 

[EduPortas]

Maybe I’m wrong but it was my understanding that the Sigma RF mount lenses are licensed by Canon so this shouldn’t be an issue anymore.

Thanks, that's a good point.

Still won't change my personal policy towards third-party options. It's OE or nada.

 

[EduPortas]

That's correct. And I have no idea what EduPortas means by "clinical, but good". It sounds like the kind of thing you say when you want to insult a lens, but have no basis for doing it. I've seen other people criticize lenses as "clinical", and it never makes any sense. Perhaps it's the flip side of saying that a lens has "character" when it's actually not a very good lens.

That's correct, I much prefer lenses with their flaws. This also makes them smaller, generally.

Every brand is guilty of this new trend and Sigma is one of the worst offenders. Huge lenses that are almost perfect in sharpness but posses no magic.

 

[John Sheehy]

I think there are some confusion regarding FF sensors and the statement that they capture 2,56 times more light. (1,6 x 1,6)

That is for the same exposure, a conditional statement.

-Yes FF sensors are 2,56 times bigger (area)

-exposure is always the same when settings are the same. (And the sensors are equally sensitive to light)

Exposure has nothing at all to do with a sensor's QE or the photosite capacity. Exposure is about what the lens projects before it goes through the sensor's surface

-Its all about pixel size. If you have a crop sensor and a FF sensor with the same individual pixel size. They would be equally as good. (If made the same way)

Not really. Pixel-level noise should be similar, but in practice, we have mostly seen higher pixel-level noise with the same pixel size on FF cameras, vs APS-C cameras. Some examples would be the 7D2 is cleaner than 5DSR, Nikon D500 is cleaner than D850.

The problem is that crop sensors are always packed tighter compared to FF sensors when regarding individual pixel size. This is why FF sensors generally perform better.

"Per unit of sensor area", there is no generic noise benefit to larger pixels, whatsoever, other than with very low f-numbers with front-side-illuminated sensors. In fact, they are more vulnerable to downstream noises because downstream noises are independent of sensor charge and any first-stage amplification, and one bad noise impulse (or hot pixel) ruins a larger sensor area, with larger pixels.

Almost all of the differences and wildcards that we still see when normalizing noise per unit of sensor area, are due to readout technology and pixel clocks, and have nothing to do with photon efficiency, which really does not vary much at all between sensors of the same general time period, despite widely varied pixel densities and sensor sizes.

The underlying rules implied by the facts on the ground of existing sensors, are that larger sensors especially are difficult to read out at a fast pixel clock without introducing extra noise, and larger sensors' noise character tends to be more spatially correlated, and therefore easier to see, compared to smaller sensors.

Larger sensors only have better potential for noise at a given exposure level and ISO because of a larger sensor area, to capture more total light. To benefit from this in real world practice requires either full exposure at base ISO with no shutter speed challenge, or a lens that has a larger pupil than what you would use with a smaller sensor, when shutter speed needs prevent base ISO.

Here are four different sensors with different sensor sizes and pixel densities, all receiving approximately the same number of photons in the exposure, in each of the four windows:

Studio Comparison Tool

 

[DNBush]

Thanks to all who replied. I've decided to go with the R6 II.