Canon EOS C50 Lab Test – Rolling Shutter, Dynamic Range and Exposure Latitude

Lab Test

by Gunther MachuJanuary 26^th, 2026

Subscribe to Lab Tests

Following the announcement of the Canon EOS C50 in September last year, it took a while for a production camera to arrive at our CineD headquarters, where we could run it through our standard CineD lab test procedures. Hence, in a similar fashion to last year, we start the year of camera lab tests with a Canon C camera. Curious to see how it fares? Then read on …

For a detailed breakdown of specifications and features of this new member of the Canon C family, please head over to our article here – however, I have to give a special mention to the new 7K 12bit Canon RAW (CRAW in this article), 3:2 open gate mode which was a “first” with this camera in the EOS cinema line (the C400 is following with a firmware update).

Also, I have to explicitly draw your attention to the beautiful mini-documentary that my colleague Johnnie shot in Japan with the C50 here.

Last but not least, I have to mention again my colleague Florian Milz, who helped to shoot and analyze this test – thank you, Florian!

Rolling Shutter of the Canon EOS C50

As usual, we use our 300Hz strobe light to reveal the pattern of black and white bars – a typical artefact of rolling shutter sensors, which allows precise measuring of the rolling shutter.

For 7k 3:2 open gate full frame CRAW at 25fps we get a rolling shutter of 18.1ms (less is better):

This is good, but not exceptional. Image credit: CineD

Except for the 4K120 mode, which clocks in at 7.1ms of rolling shutter, all other 17:9 modes on the EOS C50 (CRAW or compressed) are at 14.3ms of rolling shutter in full frame and S35 modes.

Dynamic Range of the Canon EOS C50

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

Let’s have a look at the waveform plot of shooting our Xyla21 chart using 7K CRAW 25fps, developed to CLog2 at ISO800 (the lower native ISO) in DaVinci Resolve in a 7K timeline:

This looks really noisy – quite typical of Canon RAW, which uses very little, if any, noise processing in camera. IMATEST reveals the following:

We are getting 9.98 stops at a signal-to-noise ratio (SNR) of 2, and 11.2 stops at SNR = 1. This is half a stop less than the Canon EOS C400 in 6K CRAW at ISO800 (lab test here), 0.8 stops less than the Canon R5 Mark II in 8K CRAW (lab test here), and 1.3 stops less than the 6K CRAW Canon R1 (lab test here).

However, as shown in the bottom-right “Noise spectrum” in the diagram above, the EOS C50 maintains a very high amplitude at high resolutions (frequency), hence yielding a very detailed image!

At the second native ISO of ISO6400 CRAW, we get 8.7 / 10.1 stops at SNR = 2 / 1. Hence, we lose a bit more than a stop.

Quite strangely, moving to S35 / APS-C 5K CRAW at ISO800, we are getting 11 / 12.3 stops at SNR = 2 / 1. Basically, on a pixel-to-pixel level, the dynamic range in RAW should stay the same regardless of whether we crop into the sensor, and that’s what most cameras show in our tests. Not so here; hence, some kind of processing is happening here that yields better results. This fact can also be seen in the lower-left graph below, where the noise levels in the shadows are shown as “Noise (% of max. pixel)”. On the left side of the diagram, we see that the noise levels (of the RGB channels) range from 1.5 to 2. On the same diagram for 7K full frame, we see higher noise levels of around 3 (as shown in the IMATEST result above).

Again, at the second native ISO of 6400 in S35 5K CRAW, we are now getting 8.9 / 10.5 stops at SNR = 2 / 1. Again, we lose more than 1.5 stops.

Now, using the internal compressed 4K XF-AVC 10-bit codec in full-frame mode at ISO800, we are getting 12.5 / 13.9 stops at SNR = 2 / 1.

At the second native ISO of 6400, we get 10.8 / 12.6 stops at SNR = 2 / 1. About 1.5 stops less.

We are seeing the typical Canon philosophy at work here. 12-bit Canon RAW is mostly unprocessed, yielding the highest resolution but a lot of noise, whereas the internal compressed 10-bit codecs are quite noise-reduced / processed out of the box, yielding high values in the IMATEST analytics but limited in resolution (typically, noise reduction reduces higher-resolution details). Absolutely fine if you just want to use the internal compressed codecs out of the box without much post-processing.

Please head over to our camera database for results across all other shooting modes for this camera.

Exposure Latitude of the Canon EOS C50 using 12bit CRAW at ISO800

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

Typically, we use the highest-resolution, bit-rate, and bit-depth codec for this test, which is 7K 12-bit Canon RAW, developed to CLog2 in the Camera Raw tab in DaVinci Resolve 20.3 at ISO800.

Our studio base exposure is (arbitrarily) set to an (ungraded) luma value of 60% on the forehead of our subject on the waveform monitor. In this case, my colleague Nino:

From here, we can overexpose by 4 stops and still keep Nino’s forehead just below the clipping point of the red channel:

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. Above or below that, ISO and the typical lift, gamma, and gain controls were used. Also, a 3-node tree was used: the first node converts the CLog 2 files to DaVinci wide gamut (via a color space transform), the middle node applies all adjustments, and the final node uses another CST to reach Rec709. Noise reduction was always made on the first node.

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

Already at 2 stops under and pushed back (at 6 stops of exposure latitude), we see quite some chroma and luma noise creeping into the image. Nothing yet to worry about, but this came surprisingly early.

At 3 stops underexposed, pushed back, we cannot ignore the noise any longer:

We are at 7 stops of exposure latitude. The noise is still finely dispersed and can be cleaned up using noise reduction in post.

Now let’s move to 4 stops under, pushed back to base:

Noise is now all over the place in the moving image (the still above looks way better), and some sort of fixed patterns are showing up (faint horizontal and vertical lines). It’s easier to see this using noise reduction:

We are at 8 stops of exposure latitude, which was already state-of-the-art for consumer full-frame cameras a few years back. Can we push it to 9 stops of exposure latitude? Here you go:

At 9 stops of exposure latitude (or 5 stops underexposed and pushed back to base), noise is corrupting the image to an extent that it cannot be fully salvaged using noise reduction in post:

Faint fixed pattern noise (horizontal and vertical lines) is visible (much more pronounced in the moving image), and the image turns greenish. Also, the shadow side of Nino’s face cannot be recovered without overdoing it with noise reduction. All details are gone (look at Nino’s hair, for example). Game over.

For your reference, here is the 6 stops underexposed image (10 stops of latitude), pushed back to base:

Well, what should I say? It is 2026, and since 2019, we have seen consumer full-frame cameras that could do 8 stops of exposure latitude even with internal compressed 10-bit codecs. For example, the Panasonic Lumix S1 that was released in February 2019 … that is 7 years ago …, or the Lumix S5 that we tested in 2020.

In 2022 the APS-C camera FUJIFILM X-H2S (lab test here) already scored close to 9 stops, in 2024 the first Sony global shutter full frame camera A9 III scored 9 stops of exposure latitude (lab test here), and yes, even the Panasonic LUMIX GH7, with a sensor roughly four times smaller than the Canon EOS C50, measured 8 stops of exposure latitude in RAW mode in 2024.

Last year, the Panasonic LUMIX S1II (lab test here) scored a solid 10 stops of exposure latitude in 12-bit RAW mode, yielding effectively the same result as the ARRI Alexa Mini LF!

For your reference, the benchmark so far is the ARRI Alexa 35 camera with 12 stops of exposure latitude.

Summary

The Canon EOS C50 shows an average to lower-end performance in our lab test. The rolling shutter values are okay at 14.3ms for most DCI 17:9 modes, dynamic range puts it behind its siblings like the Canon EOS C80 or EOS C400, and even the R5 Mark II, and exposure latitude just confirms this further.

Compared to other cameras in this price bracket, I find it hard to see any standout features (in terms of the lab test criteria). It’s definitely a very capable gimbal camera or a less expensive B-camera if you already have a Canon EOS C80 or EOS C400 and want to seamlessly intercut between them.

Have you used the Canon EOS C50? What is your experience? Please let us know your thoughts in the comments section below.