R1 Reviews

SvO2 · Sunday at 6:33 PM

Whether or not the base ISO is truly 200, it does seem like the dynamic range at 100 is the same as at 200 and more than a full stop less than the R3/R5-I at ISO 100. If they needed to make that sort of sacrifice for the faster readout, it does make me wonder why Canon didn't just go global shutter.

Quarkcharmed · Sunday at 7:09 PM

neuroanatomist said:
If it’s actually 200, only a relative handful of people will know (and probably not all of them will care).

The Canon specs state the native ISO range is 100-102400. Bill Claff shows ISO 100 as a pull from 200. He also indicates NR or other ‘cooking’ based on the energy spectrum, but that really looks no different from the Sony a1…except he plots that with circles instead of triangles on the PDR plot, indicating no NR. Let’s just say I’m not convinced the R1 base ISO is 200.

Normally - basically, in all other R cameras - you'd expect about 1 stop improvement at ISO 100 against ISO 200, something like this:

So if ISO 100 in the R1 is native, it doesn't behave like native. As a last resort, we may envisage a possibility the test shots provided to Bill were flawed.

Quarkcharmed · Sunday at 7:17 PM

frankchn said:
The Sony A9III has a base ISO of 250, so not exactly unprecedented.

The A9III has a high base ISO because it has lower full well capacity -- half the photo site was occupied by a capacitor needed for the global shutter. Maybe the R1 has similar electronics in each photo site that limits capacity.

The R1 shows almost the same performance as A9III, but without a global shutter:

Photographic Dynamic Range versus ISO Setting

So if we believe the P2P data, the R1 sensor is not an engineering marvel (although the camera as a whole can be brilliant in other aspects).

frankchn · Sunday at 7:34 PM

Quarkcharmed said:
The R1 shows almost the same performance as A9III, but without a global shutter:

Photographic Dynamic Range versus ISO Setting

So if we believe the P2P data, the R1 sensor is not an engineering marvel (although the camera as a whole can be brilliant in other aspects).

To be fair, I didn't expect the sensor to be any engineering marvel just by the specs. It is a moderate resolution fast read out sensor (1/400) but the core specifications isn't anything we haven't seen before -- either from Canon or from other manufacturers.

I view the camera as a whole as a refinement instead of a revolution. The R3 to R1 update to me feels like the 1DX2 to 1DX3 update (same res, higher fps, better AF, matched card slots), and yes I've owned all 4 of those cameras.

neuroanatomist · Sunday at 10:50 PM

Quarkcharmed said:
So if ISO 100 in the R1 is native, it doesn't behave like native. As a last resort, we may envisage a possibility the test shots provided to Bill were flawed.

Makes sense. I did notice the lack of a drop in DR going from 100 to 200.

neuroanatomist said:
Bill Claff shows ISO 100 as a pull from 200. He also indicates NR or other ‘cooking’ based on the energy spectrum, but that really looks no different from the Sony a1…except he plots that with circles instead of triangles on the PDR plot, indicating no NR.

Worth noting that Bill just changed the R1 from triangles to circles on the PDR plot.

Quarkcharmed · Sunday at 11:28 PM

neuroanatomist said:
Worth noting that Bill just changed the R1 from triangles to circles on the PDR plot.

He says there's still some processing applied, just too minor to be flagged.
Some Sony sensors also have it.
The threshold first flagging the cooked raws, as far as I understand, is largely arbitrary/empirical.

SvO2 · Sunday at 11:39 PM

Quarkcharmed said:
He says there's still some processing applied, just too minor to be flagged.
Some Sony sensors also have it.
The threshold first flagging the cooked raws, as far as I understand, is largely arbitrary/empirical.

Yeah he did post somewhere that it's a judgment call. And R5II seems to be similar to the R1 and Sony cameras in question at least at ISO 100.

Search

Search

R1 Reviews

SvO2

Quarkcharmed

EOS R5

Quarkcharmed

EOS R5

frankchn

neuroanatomist

Quarkcharmed

EOS R5

SvO2

Similar threads