Friday 9 April 2010

Lenses Sensors and Cameras


Its almost obligatory that after getting a "new" or in this case S/H lens for the Leica M9 it gets tried on m4/3. So I put both recently acquired lenses onto my GH1.

Panasonic GH1 Zeiss 21mm f2.8 Biogon Summarit-M 90mm f2.5

Both lenses produced decent pictures, which is to be expected.
These from the Zeiss 21mm F2.8 Biogon.

Panasonic GH1 Zeiss 21mm f2.8 Biogon

Panasonic GH1 Zeiss 21mm f2.8 Biogon

These from the Leica Summarit-M 90mm F2.5.

Panasonic GH1 Leica Summarit-M 90mm f2.5

Panasonic GH1 Leica Summarit-M 90mm f2.5

As you can see they performed well. The exercise was mainly for my own curiosity, rather than with future use in mind. Neither offers me any particular advantage over what I have already. The 20mm f1.7 and Nikon 85mm f1.8 that I use with the GH1 produce equally good results and have the advantage of faster apertures. The minor difference in focal length is neither here nor there. There's no denying that the lenses look good fitted to the camera but apart from that I doubt that the combinations will get much use.

A year ago I would have been more enthusiastic. At that time m4/3 consisted of one camera, the G1, plus two lenses 14-45 and 45-200 zooms. By using adapted lenses on the camera all sorts of possibilities opened up. However with the release of the 7-14 zoom, 14-140 zoom, 20mm F1.7 and 45mm F2.8 macro lenses Panasonic provided more versatility and quality, to a certain extent eliminating the need to use adapted lenses. If you want wide apertures/lens speed then adapted lenses make sense, but the idea of using them for the sake of it has never appealed to me.

There was a thread in another forum where someone was asking for advice about using a Voigtlander 15mm lens on m4/3. I bought and used this lens for some time. The advice I offered was basically don't bother with it on m4/3 unless the size and slight reduction in weight is vital. In terms of performance I always thought the 14-45 kit zoom worked very well and probably better than the 15mm. It's obviously a faster lens. Add to this the problems of "smearing" and "maze artefacts" that the Voigtlander produces and there doesn't seem much reason to use it. A reviewer in a well-respected UK magazine made the observation that the Voigtlander 15mm would produce better results on m4/3 than the kit lens. Certainly my experiences with both lenses have led me to a different conclusion.

To a certain extent the actual performance of the lenses doesn't matter. To many the enjoyment of the exercise is in the use of vintage lenses on a modern camera & I'm not immune to that either. People use c-mount lenses and lensbabies on cameras & while it has no appeal to me, if its something that people enjoy doing then I have no problem with it.

Michael Reichmann in an article on Luminous Landscape tested a few Leica lenses on m4/3 and came up with the conclusion that while they worked well there was no significant advantage over using "native" lenses and his opinion was that there is little point in buying these lenses for use on m4/3. I would very much agree with that. If you have the lenses already then by all means use them, you won't be disappointed, but on the current m4/3 sensor you'll not get better performance than if you use many of the excellent Nikon F & Canon EF lenses available for a fraction of the cost. Certainly thats been mine and others experience.

To me there is a far more important reason that determines the performance of my lenses and the quality of the files they produce. That is the Anti-Aliasing filter. Back in 2003 I was using a Kodak pro 14n. This had a 14MP sensor with no filter. While the sensor was prone to all sorts of fringing, moire and noise problems it was incredibly sharp. In terms of raw resolution and image sharpness only the Leica M9 of the cameras that I have used beats it. That includes the D3X. It did however have lots of problems & "cleaning up" the files took ages. I eventually gave up on it and its successor, as did Kodak! However once I'd seen what the potential of a non-AA-filtered file was I was always dissatisfied with anything else. Kodak & Leica came up with the M8 sensor which was a vast improvement over the 14n attempt (which was Interestingly not made by Kodak) but which still had a few deficiencies.

The M9 sensor is close to the finished article. There's still (admittedly small) areas of moire on some of the files but for the most part they are a thing of beauty. I was once looking at a flower picture taken with the M9 on my iMac screen and I involuntarily reached towards the screen to touch it. It seemed so real. When I process the M9 files I apply no sharpening either to the raw conversion or the finished file. There is no need to. Even with the D3X 25MP files I still needed a bit of sharpening to make them look "finished". That is why I sold substantial amounts of equipment to buy the M9 and lenses.

Now that I have the Zeiss 21mm, I have a wide-angle lens that realises the potential of the M9. Since this is a lens that I will use a lot, its difficult to now see a situation where I'm out shooting landscape and travel images when I wouldn't use the Leica.

If you want to look at a Leica M9 raw file Ken Rockwell has some available for download on in his M9 review at http://kenrockwell.com/leica/m9.htm












Finally heres an idea:-

A FF Rangefinder from Nikon based on the S2 using the D3X sensor (with no AA filter) Adapters for M-Mount, Contax G and F-Mount lenses.

Will it happen - No of course not. If it did they would charge a fortune for it anyway.

D