Coding needed for 35mm with UV/IR filter?

[prav66]

Advertisement (gone after registration)

I've been hearing conflicting reports on whether Lens coding is needed for the 35mm focal length lens. Sean Reid's comments have suggested it is helpful, while others have noted they didn't find it neccessary.

 

I am thinking of getting a Zeiss 35/2.0 which can't be permanently coded unless I send it to a machinist so was wondering if I could get away without doing so.

 

To complicate matters further, I believe the new Leica filters cause less cyan cast compared to the standard UV/IR B+W filters so perhaps what little cast that was noted with the earlier filters doesn't exist any more with the Leica filters?

 

Just wondering before I go out & order the lens and coresponding Leica filter...

[Advertisement]

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

[MP3]

Leica said "Yes, need IR filter and v1.10 from 35 to wider".

[lars_bergquist]

It is a matter of judgment. The cyan corners effect is there even with a 35, though it is smaller than with a 28 or certainly with a 21. The question is, how disturbing do you find it to be? My own tests, on an evenly lighted white surface, indicate that I will want to code my 35 and 28 mm lenses when I finally get all the filters I need. Other people may legitimately decide otherwise. I do think though that photogs who depend on selling their pictures will code their 35s.

 

Even by Leica factory standards, lenses from 50 mm and up (or out) will not need coding, either for brightness or colour vignetting. (Don't forget that there is old-fashioned optical and mechanical vignetting too, and here the reasoning is essentially the same as above.)

 

The old man from the Age of Roll Film

[rwfreund]

Yes,

On my sample (35 chron with B+W 486), I see an improvement in the corners with lens detection on.

-bob

[scott kirkpatrick]

It is a matter of judgment. The cyan corners effect is there even with a 35, though it is smaller than with a 28 or certainly with a 21. The question is, how disturbing do you find it to be? My own tests, on an evenly lighted white surface, indicate that I will want to code my 35 and 28 mm lenses when I finally get all the filters I need. Other people may legitimately decide otherwise. I do think though that photogs who depend on selling their pictures will code their 35s.

 

Even by Leica factory standards, lenses from 50 mm and up (or out) will not need coding, either for brightness or colour vignetting. (Don't forget that there is old-fashioned optical and mechanical vignetting too, and here the reasoning is essentially the same as above.)

 

The old man from the Age of Roll Film

 

Remember that this is a gradual effect, starting with a small overall shift towards a cooler image at the center, and increasing as a function of radius as you move towards the sides and corners. So 2/3 of the way to the corners of a white wall shot with a 21mm-eff lens (angle of view about 90 degrees) you see roughly the red vignetting (or cyan drift) that would occur in a 35 mm-eff lens.

 

There are other reasons for coding a lens. Accurate EXIF information will help to distinguish Summiluxes from Summicrons, and remind you of the focal length used for a shot months after it was taken. Coding permits corrections for vignetting in the luminance channel and overall color shifts when UV filters are added. The longer lenses (the Noctilux is the extreme example) vignette at the edges when wide open, because you can't see the whole aperture from the far corners of the image. Leica can correct for that, if the lens is coded and the internal aperture estimation algorithm works well enough. I am hoping to see some user experience on this with 1.10 and such lenses. The users may conclude, or Leica may have already decided, that such vignetting is part of the critical "fingerprint" of their widest aperture lenses, and that they must not affect it. We'll see, I hope.

 

scott

[Hans Roggen]

With my 35 cron asph i got cyan corners with the IR filter in place. Now with the 1.102 firmware and a couple of sharpie marks on the lens they have disappeared.

 

Hans

[LJL]

Advertisement (gone after registration)

With my 35 cron asph i got cyan corners with the IR filter in place. Now with the 1.102 firmware and a couple of sharpie marks on the lens they have disappeared.

 

Hans

 

Had the same experience with a CV 35/1.2 Nokton. I coded it as a Leica 35/1.4 ASPH, and now it works very nicely with the filter AND it reads as a 35 in the EXIF. The adjustments being made in-camera for the filters make it worthwhile to use coding on lenses from 35mm and wider, as Leica is recommending.

 

LJ

[35mmSummicron]

if you like shooting wide open with your 35mm lens, then i would suggest the coding.

 

i have found that the lens detection on (no filters) alone removes about a full stop worth of vignetting in the far corners. so i can imagine you will also get similar results once you mount the filters on (i'm still waiting for mine)

 

hope this helps.

 

andy

[scho]

I've been hearing conflicting reports on whether Lens coding is needed for the 35mm focal length lens. Sean Reid's comments have suggested it is helpful, while others have noted they didn't find it neccessary.

 

I am thinking of getting a Zeiss 35/2.0 which can't be permanently coded unless I send it to a machinist so was wondering if I could get away without doing so.

 

To complicate matters further, I believe the new Leica filters cause less cyan cast compared to the standard UV/IR B+W filters so perhaps what little cast that was noted with the earlier filters doesn't exist any more with the Leica filters?

 

Just wondering before I go out & order the lens and coresponding Leica filter...

I coded my Zeiss 35/2 biogon as a 35/2 summicron using a sharpie pen and use it with a B+W 486 UV/IR cut filter. I didn't see any cyan drift problems at all using 1.092, but I haven't tried it yet with 1.102. Several images on this page were taken with the Zeiss 35 and 486 filter (look at exif).

http://www.pbase.com/scho/leica_m8&page=3