Canon EOS R1 Lab Test – Rolling Shutter, Dynamic Range and Exposure Latitude

Finally, we found the time to put the new Canon EOS R1 through its paces in our standardized lab test. This top-of-the-line flagship camera surprised us in quite a few ways – curious to hear more? Then read on …

We reported about the announcement here, but just to repeat: the EOS R1’s long feature list includes a full frame 24MP sensor with 6K 60P RAW video recording, 4K 120P, C-log2 and C-log3 color space, XF-AVC-S and XF-HEVC-S, Cinema EOS file structure, simultaneous stills shooting while recording video, and even a tally lamp. It also boasts an overhauled DualPixel intelligent AF system, an 8.5-stop in-body image stabilizer, and a new level of AI in-camera integration (currently only for stills).

We were quite curious to see how it would compare to the EOS C400 in terms of lab test results, so let’s get going. Again, many thanks to my dear colleague Florian Milz, who helped me shoot and analyze this test!

Rolling shutter of the Canon EOS R1

As usual, we are using our 300Hz strobe light to get the sequence of black and white bars, typical for the read-out nature of rolling shutter sensors. Let’s start with Canon CRAW 6K at 25 fps:

Wow, we are getting a solid 8.3ms (less is better) – a very good result for 6K readout! The same 8.3ms are also obtained for 4K fine full frame readout. This is also slightly better than the EOS C400, which exhibited 9.5ms in the same mode. If we switch to 4K 120fps, the readout speed drops to an astonishing 4.1ms.

Dynamic Range of the Canon EOS R1 at ISO800

If you are not familiar with how we test dynamic range, please head over here first.

Let’s have a look at the waveform plot of our Xyla21 chart using 6K CRAW 25fps, developed to CLog2 at ISO800 in DaVinci Resolve in a 6K timeline:

We are getting 13 stops above the noise floor. Also, a 14th and a faint 15th stop can be seen inside the noise floor. IMATEST calculates the following:

IMATEST calculates 11.3 stops at a signal-to-noise ratio (SNR) of 2, and 12.7 stops at SNR = 1. CRAW is not as noisy as on the EOS C400, interestingly (which we tested at 10.5 / 11.6 stops at SNR = 2 / 1 for 6K CRAW), hence the R1 scores almost a stop higher. Also, very minimal, if any noise reduction is applied, hence high amplitudes can be seen even at high frequencies (resolution), see the lower right-hand side graph above (Noise spectrum). This is typical for CRAW – if you want the highest resolution, you should use CRAW.

Now let’s have a look at the internal compressed codec, 4K fine (subsampled internally from the full 6K sensor), and we get the following IMATEST result:

Now we are getting 13.3 / 14.2 stops at SNR = 2 / 1. Very impressive, and a tad better than the EOS C400. However, the amplitudes now drop at higher resolutions (see the lower right-hand “Noise spectrum” graph above), indicating significant internal noise reduction. Again, a typical Canon result. Internal compressed 10-bit codecs like XF-AVC provide as perfect an image as can be out of the box, while CRAW is as unprocessed as can be, hence requires a significant level of post-production.

Exposure latitude of the Canon EOS R1 at ISO800

As mentioned earlier, latitude is the camera’s ability to retain detail and colors when over- or underexposed and pushed back to a base exposure. This test is very revealing, as it pushes every camera to its absolute limits – not just in the highlights but mostly in the shadows.

Besides charts and IMATEST, which provide absolute (objective) mathematically obtained values, this latitude test is our “real world” experiment on how usable the images are using a carefully crafted studio scene.

All latitude shots were done at ISO800 with 6K 12-bit Canon RAW developed to CLog2 in the Camera Raw tab in DaVinci Resolve. Using a color space transform (CST) on the first node to DaVinci Wide Gamut, then an adjustment node, and at the end another CST from DaVinci Wide Gamut to REC709, the files were transformed into the REC709 space.

Our studio base exposure is (arbitrarily) chosen as having an (ungraded) luma value of 60% on the forehead of our subject on the waveform monitor. In this case, my dear colleague Johnnie:

Now, let’s jump to 4 stops of overexposure:

As can be seen in the 4-stop overexposed shot, the red channel on Johnnie’s face is at the cusp of clipping but remains intact.

In the Camera RAW tab in DaVinci Resolve, it’s very easy to push the files back to base exposure. Just use the exposure slider and adjust. Unfortunately, though, this only works from +3 to -3, hence the ISO values were adjusted accordingly. Noise reduction was always made on the first node.

Now, let’s start to underexpose by closing down the iris of the lens in one-stop increments (until T8), after that, the shutter value was halved.

3 stop underexposure, brought back to base exposure, we see some noise coming in, but it is not too distracting.

Only at 4 stops under and pushed back, we see significant noise creeping into the image – we are already at 8 stops of exposure latitude:

The noise is finely dispersed, and no larger blotches of chroma noise can be seen. However, some horizontal stripes can be noticed hovering across the image. Using noise reduction, we get the following:

In the noise-reduced moving image, those horizontal lines hovering around the image can be noticed, but it is still OK.

Now, let’s move to 9 stops of exposure latitude, hence 5 stops underexposure, brought back to base:

Noise is now starting to corrupt the image, especially on the shadow side of Johnnie’s face. Also, very pronounced horizontal lines can be seen, hovering around the image, which I find very distracting. Let’s have a look at the noise-reduced image:

This is now at the edge of being usable. In terms of noise reduction, chroma NR helps, but I cannot use more luma NR as the image quickly turns plasticky. Also, a pinkish cast is appearing in the shadow areas, which is difficult to remove. Hence, we have reached the limit. But we are at 9 stops of exposure latitude, which puts this camera slightly ahead of the EOS C400, interestingly. Already from the IMATEST results in CRAW, the EOS R1 fared better than the EOS C400.

Just for reference, here is the 6 stops underexposed image, brought back to base: (This cannot be saved by noise reduction in post)

Summary

Interestingly, the Canon EOS R1 clearly charges ahead of its siblings, the EOS R3 and EOS C400 (which I would both put into the same bracket image-wise). Rolling shutter is better, CRAW is less noisy and has more dynamic range, and the exposure latitude reveals about one stop more room for pushing the image around, hence 9 stops of latitude.

This puts the Canon EOS R1 on par with the Sony A9 III (lab test here), and more recently the Panasonic LUMIX S1II (lab test here) which until now was the leader of the pack of full frame consumer cameras (pending the 5.8K ProRes RAW latitude test with DR Boost “ON” which we need to finish). Just for reference, the ARRI Alexa Mini LF exhibited 10 stops of exposure latitude (lab test here), and the ARRI Alexa 35 showed 12 stops (lab test here).

Have you shot with the Canon EOS R1? What was your experience? Please share your thoughts in the comments section below