Q 28mm lens design

[adan]

Advertisement (gone after registration)

Nothing in life is free, and you lose something if you stretch (as in digital distortion correction) X-many pixels' worth of original data to cover X-times-Y-many final pixels of area.

 

However, assuming Chrismuc's estimate of this lens having about 10% distortion at the corners, digital stretching of the data 10% to straighten things out reduces the corner resolution about 1.1² times, or 1.21. Which equates to about 19 megapixels. So the corners are lower res (as though shot on a 19-megapixel sensor), but not by a huge amount.

 

I.E. it will still have more corner resolution (assuming both lenses are really good) as a near-perfect, well-corrected 28mm on an M9/MM (18 Mpixels).

 

If this were an interchangeable lens that could be used on a film M, where digital correction isn't possible (wet darkroom) or would result in weird stretched grain (if scanned and corrected manually) - it would be horrible. But it isn't.

 

Personally, I don't respect digital post-correction. It smacks of the car makers who can't build parts that fit, but instead hammer on them during assembly to force them to fit. Cheap - yes. In every sense of the word.

[alee]

Personally, I don't respect digital post-correction.

 

 

Digital post correction is really no different than the zillions of different ways you could develop film to create various effects.  Every photo is processed in some way.

 

Based on the samples, unless your photographic art is brick walls, grids and objects in far corners of the frame, the Leica Q lens, in its corrected form, looks to be spectacular.  I've seen some great photography with it so far, and a lot of sub-par photography as well.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[40mm f/2]

I just found the technical data for the Q and found more info about the lens but have not seen much discussion about the lens design here. It's design is quite different from the M lenses with the most rear lens quite close to the sensor and focuses by shifting only 2 lens groups and the rest of the lens is apparently stationary (except for a OIS element). I was reading that the lens was optimized for good performance across the entire field but looking at the MTF diagram at f 1.7 it has a high center performance but that drops strongly to the corners. One thing is remarkable that sagittal and tangential looks very similar.

At f 5.6 it is quite good across the field. I am also curious about the macro performance.

I wonder why that lens was designed that way (except for the fact that 2 small lens elements are easier for fast autofocus than moving a whole lens). One comment I was reading that 35mm focal length was considered but would have made a too bulky lens. If I look at 35mm Summicrons I not quite understand what leads to the lens design choices for the Q

Edited October 15, 2015 by 40mm f/2

[bernie.lcf]

The M lenses are designed under the assumption that they need to work on film as well. 

 

With all the discussion about the Leica Q's digital correction, people seem to forget that somebody (=Leica) had to figure out how to perfectly correct the lens in software and then put the resulting complex formula into the DNG file. The results are significantly better than creating a simple profile for Cornerfix or a similar tool. In terms of research and design it would be fair to state that it probably wasn't any easier for Leica. The final product just uses a simpler optical construction which saves money, is easier to build properly etc. because part of the optical design uses algorithms rather than polished glass. 

Edited October 16, 2015 by bernie.lcf

[tgm]

I just found the technical data for the Q and found more info about the lens but have not seen much discussion about the lens design here. It's design is quite different from the M lenses with the most rear lens quite close to the sensor and focuses by shifting only 2 lens groups and the rest of the lens is apparently stationary (except for a OIS element). I was reading that the lens was optimized for good performance across the entire field but looking at the MTF diagram at f 1.7 it has a high center performance but that drops strongly to the corners. One thing is remarkable that sagittal and tangential looks very similar.

At f 5.6 it is quite good across the field. I am also curious about the macro performance.

I wonder why that lens was designed that way (except for the fact that 2 small lens elements are easier for fast autofocus than moving a whole lens). One comment I was reading that 35mm focal length was considered but would have made a too bulky lens. If I look at 35mm Summicrons I not quite understand what leads to the lens design choices for the Q

Were did you found MFT curves for the summilux 1.7?Could you please provide a link.

 

many thanks

[40mm f/2]

Were did you found MFT curves for the summilux 1.7?Could you please provide a link.

 

many thanks

http://us.leica-camera.com/Photography/Leica-Q/LEICA-Q/Downloads

look for technical data