Leica M10 vs Leica SL vs Canon 5D Mark IV - Comparing Color Rendering
In a previous article, I had commented that images from the Leica SL appeared flatter than images from the Leica M10. I had come to this conclusion, after comparing the two cameras. This was an unexpected finding, because one would likely assume all Leica cameras to be similar, and that the images they rendered shouldn't be terribly different. After all, both the M10 and the SL have 24 megapixel sensors. In that way, shouldn't the images be more or less similar?
Well, that wasn't exactly the case. The SL did not render images like the M10. It was an unexpected observation, to say the least. Invariably, this led me to more questions. Is the sensor on the Leica M10 very different from the Leica SL? And if so, why is this the case? These were the question which formed the basis of my investigation.
As luck, or rather flexibility in planning would have it, we were able to stay in our current location for a second day, for the sake of this investigation. It was after all the ideal place to compare image rendering between the SL to the M10. How could it not be? The interior was hot red - I mean really hot red. Besides, it's not as if I got my money's worth with it on my previous post. So needless to say, I'm going to go full remedy on it with this post.
For this comparison, we had decided on just the one Leica 35mm f/1.4 Summilux ASPH FLE. We figured we were going to be indoors, anyways. As such, swapping lenses between the Leica SL and the Leica M10 wouldn't be much of an imposition. And to make this investigation an even more colorful comparison, Anna decided to get a vibrant blue Chinese dress from Shanghai Tang - in keeping with the contemporary Chinoiserie of the interior. Only problem is, she forgot to bring the matching flats.
And for a second straight day, Anna also forgot to bring the mounting plate for the tripod.
To make up for this shortfall, albeit reasoned in hindsight, I brought along a third camera for this investigation - namely the Canon 5D Mark IV + Canon 35mm f/1.4L USM II. I thought it would be interesting to see if the image rendered on the SL was closer to the 5D4 than the M10. Previously, I had also commented that the SL rendered more like a Canon than a Leica. That seemed true with the Canon 1Dx Mark II, so it would be interesting to see if this is the case with the more common Canon 5D Mark IV.
Set 1 - 3
On first glance, the rendering does appear to support the argument that the Leica SL is more similar to the Canon 5D Mark IV than the Leica M10. In most of the image sets, the color temperature of the SL was cooler than M10, and similar to the 5D4.
But what about the tonal range of the three cameras? Does the M10 have more of a 3D pop? Is the rendering of the SL similar to the 5D4?
Unlike my previous comparisons, the testing conditions have changed in this one. The addition of a colorful interior without direct outdoor lighting seemed to have impacted the presence of that thin veil of gray I characterized with the rendering of the SL. Furthermore, the color rendering of the 5D4 appears to be noticeably more pleasant than the 1Dx Mark II. So it's not exactly comparing apples to apples then.
However, I am still of the opinion that the colors are slightly more muted on the SL and 5D4 than the M10. Why this happens, I can only speculate. Perhaps Leica M mount lenses are more sensitive to color, and can therefore render colors more optimally on M mount sensors. Perhaps the algorithm of the M10's RAW compression adds an artificial sweetener that gives the color an extra edge over the uncompressed RAW of the SL and the 5D4. Or maybe it's just a difference in sensor.
Whatever the reason is, the M10 does appear to have a slight advantage in tonal range in giving its rendering a 3D pop.
Set 4 - 5
In layman's term, what does this all mean?
Basically, the rendering of the Leica M10 feels more intimate, while the Leica SL and 5D4 feel more professional. It makes sense when you think about it. The M10 is a very deliberate camera, while the SL and 5D4 are more indiscriminate. As a result, the warmer tones of the M10 makes its rendering feel more personally invested, while the cooler tones of the SL and 5D4 make its rendering feel more neutral for general use.
I admit it's unreasonable to assert that manufacturers could imbue a camera with personality. But from a technical stand point, it really isn't outside the realm of possibility either. Consider what is necessary to create warmer tones. All that is required is some tweaking with the white balance. But with the M10, I don't think that is the case. In examining the sample images, I am of the opinion that it has something to do with how these cameras deal with reflected light?
What do I mean by that?
I'm talking about the light bouncing off all the surfaces surrounding the frame. If the surface is hot red like my testing situation, the warmth of the red tones would reflect off the surface, and find it's way to the image sensor. An interesting hypothesis? But when put to the test, there seems to be some truth to it - it would seem. On the first two sets, the rendering from the M10 is noticeably warmer, given that the surrounding area is red. But I suppose all that could just be a coincidence, right?
But then you examine the third set, and voilà! All three sample images are noticeably cool? How could that be? The M10 is suppose to render warmer! Well, not necessarily. Given the absence of my beloved hot red surface, the reflected light wasn't warm, which resulted in a cooler rendering, made cooler by the sunlight coming in. That is why the rendering from the M10 had the same temperature bias as the SL and 5D4.
Of course, this is all speculation.
Set 6 - 7
Or is it? It got me thinking. If I were the manufacturer, and if I wanted to imbue the M10 with a more intimate personality, wouldn't it make sense for the image sensor to be more receptive to collecting warm color tones?
Putting this assertion to the test, I decided to compare a sixth image set, where extreme backlighting was used to intentionally obscure image capture. I reasoned that better recovery of warm tones in post would demonstrate the image sensor's bias, since the foreground underexposure and the interference from the bending light would limit what the sensor could capture.
Based on the recovered images, the rendering from the M10 was noticeably warmer than the SL and the 5D4. The M10 retained the most red tones, after all. Clearly, the M10's sensor exhibited a preference to warm tones, followed by the SL, and then by the 5D4 at the rear.
Still, I suppose it could have nothing to do with the sensor's receptiveness to warm tones. Maybe it has something to do with the RAW files of the M10. Maybe at the pixel level, the warm tones of the rendered images are just more sensitive to editing. To put that to the test, I decided to conduct a seventh image set, comparing optimization in post, using the same edited values - for the sake of rendering the same results between the M10, SL, and 5D4.
In the final edits, the warm tone from the M10 clearly exhibited greater sensitivity to editing. This you can see from increased redness in skin tone on the face, arms, and hands. As for the SL, it also seems a little warmer than the 5D4. Although, it should be noted that the captured color temperature of the SL was noticeably cooler than both the M10 and 5D4. As a result, it had to be tweaked, for the sake of keeping the skin tone consistent between the three rendering.
I admit, not the most scientific of comparison. But like I said, I'm just speculating. With that said, the Leica M10 does appear to be a warm and intimate camera, whereas the Canon 5D4 is cool and indiscriminate in application. Surprisingly, the Leica SL is not as similar to the 5D4 as I had originally assumed. The SL might appear cooler at times. But at heart, with it's sensor and RAW files, it does appear warmer than the Canon. I guess in that way, the SL is much closer to the M10 than I previously concluded. But why shouldn't it be? It is a Leica, after all.
Having gone through this comparison with the M10, SL, and 5D4, I can only assume that differences in rendering characteristic is probably true for most digital cameras. Then, it got me thinking again. What if I expanded this comparison? What if I compared the rendering from eight different cameras? Wouldn't that be fun?
The additional five camera I selected for this last comparison are as follows:
1. Leica M240 + Leica 35mm f/1.4 Summilux ASPH FLE
2. Leica M9 + Leica 35mm f/1.4 Summilux ASPH FLE
3. Nikon D700 + 35mm f/1.4G
4. Sony A7R Mark II + Metabone Adapter + Leica 35mm f/1.4 Summilux ASPH FLE
5. Fujifilm S5 Pro + Nikon 24mm f/1.4G (full sensor äquivalent is close to 35mm)
For this, I'll let you make your own assessment. But if you ask me, I really like the rendering from the Fujifilm S5 Pro... as for the Sony, less so. But in its defense, I didn't use a Sony lens.
So what is the point of comparing the image rendering of all these cameras? It is something I haven't thought about since the days of film. But, now in the digital age, we take for granted the medium of image capture. Given the limitation of digital cameras, we no longer have the same options that we used to have with film. After all, we only have one digital sensor per camera.
But then again, given the flexibility of editing software, is it really necessary to chose between digital sensors the way we used to chose between film? The thing is, the differences between digital sensors can easily be made the same in post, without significant effort.
I'll give you one more set to prove my point. Can you guess which camera rendered which image? I'll give you a hint. You probably won't be able to tell by the sensor, but you may be able to tell by the lens. If you want a bigger challenge, the third image of this set is the title image. But for that one, you won't be able to rely on the lens, which in itself is another clue. Good luck.
Exposure has been optimized in Lightroom. Images have been cropped and leveled unless otherwise stated. All other values have been edited at varying degrees - none for the first couple of sets, but increasingly more as the comparison progressed.
PS: I love the hot red interior.