Hi,
I always find it is good to take all tests with a grain of salt. Lot's have been written about the new Leica M, here are some observations:
Erwin Puts who is famous for his Leica tests has published some observations which are erroneous.
1)
http://www.imx.nl/photo/blog-2/lens-test-or-system-test.htmlHere he states that Lloyd Chambers has overestimated focus shift on by a factor of 2500. Mr Chambers reported a focus shift of 20 mm at 1.8m while Erwin Puts has measured 0.008 mm on the optical bench. What Mr Puts misses is that Lloyd is discussing focus shift in the subject plane while he has measured the focus shift in the image plane.
Let's look at the lens equation that says: 1/f = 1/a + 1/b with a being distance from lens to subject a b being distance to from lens to film. So for a 50 mm lens focused at 1.8 m we would get b=1/(1/f - 1/a)) -> 51.42857 mm, with 2 cm focus shift we would get 51.41243, that is the focusing error would be 0.016 mm.
This is twice the amount measured that was measured on the optical bench by Mr. Puts. Pretty decent estimate. I found no way to contact Erwin Puts to point out the mistake.
2) In the article below the writing is not very clear, but the edge contrast curves show a clear indication of sharpening:
http://www.imx.nl/photo/leica/leica-m-part-2-comparison.htmlThe reason that the program (Imatest?) gives MTF above 100% is neither the program nor the sharpness of the lens but the sharpening in raw conversion. That said, the is the lens is excellent, in all probability, but it is hard to deduce from the article as it is not clear how much of the measured MTF is attributable to sharpening, which sharpening parameters were used and it is not always clear what each curve shows.
3) Tim Ashley has tested the M (240) with a Summilux 35/1.4 FLE comparing it to Sony RX1, his finding was that the Summilux was tricky with a wavy field curvature. Tim's article is here:
http://tashley1.zenfolio.com/blog/2013/4/leica-m-240-with-35mm-f1-4-fle---some-observations.
Leica:
http://tashley1.zenfolio.com/img/s8/v76/p1528541534.jpgZeiss:
http://tashley1.zenfolio.com/img/s8/v74/p1528544446.jpgThe above images are worth a look, the red labels slightly of axis in the Leica image are clearly out of focus while the text under "winners" is sharp.
4) Getting back to Mark Dubovoy's test, i have some observations:
- He uses camera white balance. Having a nice WB from a camera is certainly helpful, but in many situations a correct WB can be meassured using a grey card.
- An image with colder balance is generally perceived sharper.
- Making large size prints and using observers unaware of print identity is a good technique to compare images, but it is essential to keep processing out of the equation. Including a color checker and use for grey balance and exposure/contrast adjustment is a good way to compare images.
- Mark indicated that the Nikon exposes less to the right by default, that means that DR is not fully utilised. For complete utilisation of DR ETTR is essential. Most testers found that the M is somewhat more noisy in the shadows, but Mark's observations go into another direction.
Best regards
Erik