Sacrifice, Sacrilege, or Sweet Spot? Using M-lenses on a Panasonic camera

I recently did a deep dive in my quest to determine the pros, cons, and in-betweens of using M-lenses on a Panasonic camera. In this case, a Lumix S5IIx body. I also compared how the lenses performed on several other cameras. Some results were as expected, others were not. I also encountered a few surprises.

Earlier this year, I wrote a detailed comment on a Macfilos article published in June 2024, about the Lumix S5II. It was on the topic of sensor-stack thickness. In particular, how that thickness affected whether you could use M-lenses on a Panasonic, or other non-M-body. The comment elicited a discussion with Macfilos editors and Jonathan Slack, a beta-tester of Leica cameras, and regular Macfilos contributor.

After some back and forth on how M-lenses performed on different camera models, Jonathan asked me a very good question. In essence, he asked why not just get an SL3, which Leica say is optimised for M lenses?

In part, my answer comes down to personal preference (size, weight, handling) and also cost. 

But it would also depend on how well other options performed. Some of you are in the same boat: wondering which cameras work best with M-lenses. Given the relationship between Leica and Lumix, you might be especially interested in whether you can confidently use M-lenses on a Panasonic camera.

So here goes.

Zen, birds and practicality

In the same way that birds are imprinted with their parents at birth, I may have been imprinted with rangefinders.

At age 13, I got my first 35mm camera, a hand-me-down Kodak Retina IIIc rangefinder. While working in a camera store to put myself through university, I managed to buy a Leitz Minolta CL, another rangefinder. The kit included 40mm and 90mm Minolta M-Rokkor lenses (Leica M-mount). That camera went with me everywhere.

I started my photography business after graduating, still in the film age. For that, I went to SLRs, followed by autofocus SLRs, then on to DSLRs. As a full-time photographer, and having started from scratch, it generally hasn’t been practical for me to own Leica gear.

From a business standpoint, it has made more sense to own capable but less expensive systems. I started with Canon and eventually moved to Sony and FujiFilm. Along the way I got some Leica gear. But it never really worked out as an everyday system for my photo business.

With all the excitement of the digital era, I didn’t look back.

I suppressed it for a while, but eventually found myself missing the zen experience I remembered from my days using all manual cameras and lenses.

Returning to my roots

Enter M-lens adapters: I can get beautiful compact manual focus lenses and put them on my digital Electronic Viewfinder (EVF) cameras. At least in theory. Having made my living for many years selling large fine art prints, I am very particular about the quality of the files coming out of the camera.

When I make two-meter prints for patrons (not that unusual), everything must be flawless or nearly so. There are exceptions for some types of images but those are, well, exceptions.

Owning many Sony cameras over the years, I realized there are more lens adapters for Sony than just about any other full frame camera. Sony cameras can be soulless, but are technically competent. So, I figured, why not try putting some nice M-mount lenses on one.

While researching the subject, I learned that the thick Sony sensor stack (sensor + sensor filtration) is incompatible with some rangefinder lenses, especially wide angles. Thicker stacks can cause smearing and exacerbate weaknesses, especially at the edges and corners of an image. Thinner stacks tend to perform better.

I knew that Leica designed the SL line, such as the SL2/S and SL3/S models, to work well with M lenses. Presumably, SL bodies have a thin sensor stack for this purpose. But an SL3 wasn’t in my budget. And after using an SL2 and SL2-S I wasn’t so keen on the size and weight either.

Hunch

I had a hunch that with the Leica/Lumix relationship, the S5II(x) might have a thin enough stack to be suitable for rangefinder lenses. Or at least it would be thinner than a Sony a7-series body. So, what results would you get using M-lenses on a Panasonic camera?

I got in touch with Ilija Melentijevic, founder of Kolari, knowing he is an expert on sensor stacks. Amid all the great information he provided, he confirmed my hunch. A Panasonic Lumix S5II/S5IIx sensor has a thinner stack than a Sony a7R series camera. Whether the S5II(x) sensor stack was designed with Leica input or not, I don’t know. But my new-found knowledge about the thickness, or should I say, thinness, set me on a path of exploration.

The proof of the pudding is in the eating

Wanting to try all this out for myself, I bought both the S5IIx and a Sony a7RIII, which I had owned in the past, along with adapters for M-mount lenses. Within a week, I returned the Sony body and kept the S5IIx for its noticeably superior performance with my M-mount lenses. Although I have an extensive FujiFilm system, I didn’t have any great desire to use the M lenses with my X-T5 bodies. The one exception might have been an infrared-modified model. 

I am grateful to Camera West (Walnut Creek, California), and in particular, store manager Norman, for lending me the Leica gear I needed to do a thorough, hands-on investigation. This allowed me to evaluate M-lenses on a Panasonic camera — my Lumix S5IIx — and compare that to both an SL2-S and an M11.

The SL2-S is a very good Leica counterpart to the S5IIx (no SL3-S bodies were available.) In case you’re wondering, my purchase of the ‘x’ version of the S5II has no significance, apart from my preference for its aesthetics. For still photography, there is no difference between the S5IIx and the S5II.

My main goal was seeing how the S5II(x) performed against the SL2-S with various M-mount lenses. But it was also useful having the M11 as a baseline/standard for reference when I needed it.

The M-lens cadre

Here are the lenses I used:

• Leica 50mm f/2 Summicron-M
• Leica 28mm f/2 Summicron-M
• Voigtländer 21mm f/1.8 Ultron VM
• Voigtländer 75mm f/1.9 Ultron VM

I didn’t expect the 75mm to be much of a challenge for the cameras. But it had done poorly on the edges with the thick-sensor-stack Sony until about f/5.6, so I included it.

Methods

I first tested some M-to L-mount adapters, to find the best one. I wanted to use the same adapter for both the S5IIx and SL2-S to eliminate that as a variable. Eventually, I settled on the Light Lens Lab adapter.

I conducted tests on all the lenses at both infinity, with a detailed horizon, and in the studio at closer distances, using ISO 12233 resolution-test targets. My wall of targets is arranged so that lenses are at fairly realistic working distances when testing. This may not be entirely scientific, but I’m not counting lines of resolution on the charts, and it’s quite useful for comparisons among lenses. As a side benefit, I have collected a useful reference for dozens of lenses over many years.

In this context, my judgment of sharpness is really perceived sharpness, as there are many factors that contribute to how I, or any human, “sees” sharpness in an image.

This isn’t meant to be a typical review, but I wanted a consistent way to compare the lenses. More meaningful, to me anyway, are the shots I did of landscapes on the coast. These encompass various distances, tones, and details in a single frame.

Expectations and a surprise

28mm

I chose a Leica 28mm f/2.0 Summicron-M, which turned out to be the most challenging of the bunch. It has a prominently protruding rear element, close to the sensor, moving closer still at infinity.

Although the 28mm ‘Cron’ (Leica users’ parlance for Summicron) does well enough with the S5IIx at f/5.6 and looks excellent at f/8.0, the SL2-S provides better images, especially at wider apertures and at corners and edges.

I should note that even on the M11, the type of camera for which this lens was designed, the 28mm f/2 Summicron exhibits soft corners, from wide open through f/4.0. On any of the bodies discussed here, I would stop the 28mm down to f/8.0 for a large print. With a main subject only in the centre, I would be comfortable with it wide open.

21mm

After testing the 28mm f/2.0 Summicron, I assumed an even wider-angle lens, such as the Voigtländer 21mm f/1.8 VM, would have noticeably worse performance on the S5 IIx. In fact, it did pretty well. The infinity test with such a wide-angle lens makes it more difficult to scrutinise fine detail because that detail is so small. Occasionally, I had to zoom in to 200% to evaluate differences.

Bear in mind that if you are going to make a print, even a large one, viewing at 200% doesn’t tell you much that is useful. When I make large prints, I will view at 50% on my monitor to get an approximate idea of how a print will look when examined closely. But even 50% is greater detail than someone sees at normal viewing distances. After decades of exhibiting my work, I could always tell when people looking at my artwork were photographers. Their viewing distance was based on the length of their nose!

The 21mm lens doesn’t show as much difference between the SL2-S and S5IIx as the 28mm Cron did. At the widest apertures, the SL2-S has slightly better sharpness in the centre, but the S5IIx appears slightly sharper mid-frame. I think this is more about the contrast and lack of/or less antialiasing (AA) filtration of the S5II sensor.

By f4.0, the optimum aperture for this lens, there isn’t much difference, even out to the far corners. If I were to use the S5IIx file, which has a little more colour fringing, I would use the “Remove Chromatic Aberration” checkbox in Lightroom, or the equivalent in whatever raw processor I use.

50mm

This is the one that surprised me. The Leica 50mm f/2.0 Summicron-M certainly lives up to its long-established reputation for excellence. The surprise was that it performed better on the S5IIx than the SL2-S. This was the case in all parts of the frame, though was most noticeable in the centre, wide open at f/2.0.

By f/4.0 the two cameras are close to equivalent in the centre. But the S5IIx is still slightly better at the edges and corners, from f/2.0 through f/8.0.

I didn’t believe the results at first, so I reshot the SL2-S on a different day and rechecked everything. I also compared both the multi-shot, which produces a 96MP image, and the standard 24MP versions¹.

This particular example of using M-lenses on a Panasonic camera (S5IIx with 50 Cron) demonstrates that not only can I achieve comparable results to pairing this lens with the Leica SL2-S, but in this case, even a bit better.

75mm

I also tested my Voigtländer 75mm f/1.9 Ultron VM, a really compact and fast portrait-length M-mount lens. At most apertures, the SL2-S exhibits a slight advantage, but I sometimes needed a 200% ‘crop’ to see it.

There isn’t as much difference in the corners as with some of the other lenses. Typically, a longer focal length shouldn’t exhibit as much “smearing” of the edges and corners as a wide-angle lens. So I didn’t think this lens would be much of a challenge. I will say that when I tested it on a Sony a7R III, it was quite soft (“smearing”) at the edges.

Interestingly, at the closer test distance for this lens (3m/10ft in the studio) the S5IIx yields noticeably better sharpness in the centre wide open. Once the lens is stopped down, their performance is fairly close, with perhaps the SL2-S having a slight edge overall.

Bottom line

Yes, there are some sacrifices when using M-lenses on a Panasonic Lumix S5II/S5IIx body. However, they are not as many, or as sizeable, as I would have expected before this evaluation. Leica does seem to have optimised the SL2/S and SL3/S bodies to work well with their rangefinder lenses. The SL2-S generally did an outstanding job with the M-mount lenses I used. But, in a few cases, I feel the M11, not unexpectedly, got the most out of them.

The whole point of this exercise was to see if I could use M-lenses while taking advantage of the benefits of a non-rangefinder, mainstream camera, possessing an easy-to-view EVF, image stabilisation, and other accoutrement of a modern digital camera body. Sans autofocus of course.

Is it sacrilege, putting those beautiful and expensive German made (or Canadian or Portuguese made) Leica M-lenses on a Japanese-made Panasonic camera? That depends on your sentiments.

Like many of us, I put Japanese made Voigtländer lenses on my Leica camera, so why not the other way around? I will admit that you do indeed lose some Leica je ne sais quoi in doing so. For that, there is really no substitute.

Consider that the cousin to the Lumix S5II/S5 IIx, the Leica SL3-S, is currently about three times the price. If you prefer the size, weight, and handling of the S5 II/S5 IIx, then the Lumix S5 II/S5IIx might just be a sweet spot.

Dataset used for this project

4 Camera bodies and 7 lenses (3 bodies and 4 lenses included in article)

197 Studio resolution test images for comparisons shot on 3 different days

254 Infinity scene tests shot on 2 different days to account for lighting differences

461 Images of nature and people on several occasions, for enjoyment and comparisons. It’s gotta be fun, or I wouldn’t do it

Lots of hours comparing images of each lens at various apertures and magnifications. I didn’t keep track, but was bleary-eyed at the end.

---

1. Both of these cameras allow you to use built-in pixel shift to achieve a 96MP resolution. Although the handheld mode for this works pretty well, I locked everything down on a tripod. It proved to be a real and useable improvement in resolution, allowing me to examine fine detail. ↩︎

---

Read more about the Lumix S5II(x)	Read more about the Leica SL2-S
Joel Wolfson Website	Joel Wolfson Blog

---

A cup of coffee works wonders in supporting Macfilos

Did you know that Macfilos is run by a dedicated team of volunteers? We rely on donations to help pay our running costs. And even the cost of a cup of coffee will do wonders for our energy levels.

---