Leica Digital M

[Simbad]

The current edition of British Journal of Photography (BJP) carries a review of Zeiss's new M fit 15mm Distagon lens and btw give it a pretty good rating. They also mention that Leica DM will have a mechanical optical device fitted to the camera lens mounting ring to automatically 'read' the latest Leica M lenses produced from July. Lenses will be fitted with a chip producing binary codes that allows the camera to identify focal length, maximum aperture and possibly known aberrations to allow auto correction within camera firmware. Some of this data will show up in EXIF files. Older Leica lenses may be able to have the chip retro-fitted at a cost of 95 Euros each. The review also hints at a statement from Leica with regard to the launch of the digital M in late August 06.......so not long now!

BJP hint that this innovation from Leica is a smart marketing move to reduce Zeiss's attack on the M fit market. It could also be a very good reason to say with your RD-1 if you have a lot of old Leica, Zeiss or Canon lenses!

The BJP review used both an RD-1 and Leica M6 to test the lens with digital sensors and Kodak BW400 film. Results were excellent with film and "variable" with digital, they say.

 

[Jim Watts]

It's not absolutely clear from the review but the "variable" results on the R-D1 seem to refer to flare performance with the R-D1 sensor: "On digital, when the lens becomes a useful 23mm equivalent, the results obtained were variable at the maximum aperture and more so when subjects were deliberately selected for strong back or side lighting."

Elsewhere in the review the lens performance on the R-D1 is praised for contrast and delineation of fine detail as well as lack of colour fringing, so it dosn't appear to be failing here. An example is given (taken on the R-D1) of a crop that would equal a 105 inch long enlargement.

Vignetting performance on film in a Leica M6 was thought to be excellent and the supplied 1.5 (centre?) ND filter a matter of choice rather than need. Although comparisons were made to the VC 15mm Heliar with regard to the vignetting the review is not too specific about vignetting on the R-D1; "Using an aperture of f5.6 for all exposures including film, frames depict an exact geometry with quite even illumination - it is difficult to detect faults in a full frame print 12 inches wide."

Another interesting comparison with the VC 15mm is that the Zeiss lens gives a noticable slightly wider field of view.

For film the lens is thought to be at the leading edge. For digital "Just how good a performer this will turn out to be only only depends on the camera manufactures choice of imaging sensor and the software used to manage the data from it" Hence I think the long preamble in the review about Leitz lens chipping.

The reviewer Jonathan Eastland is a long time user of the R-D1 so it would be nice to know more about the "variable" results here. Perhaps the review was edited down? One would though assume from much of the review that performance on the R-D1 is much better than the VC 15mm, as it should be given the price difference.

URL for BJP review, but you need to be a subscriber to read the full review:

http://www.bjp-online.com/public/showPage.html?page=bjp_contents

 

[Mark Norton]

We've known about the new coding for a few weeks now and 8 of my lenses are back in Solms for upgrading.

The coding is not a chip, it's simply a pattern of black and white dots engraved in the lens mount surface which are read optically by the camera. We expect there will be lens-unique image processing in the camera specifically to address issues with wide-angles.

I'm interested that the results with the Zeiss Distagon are "variable" with digital. This is an expensive lens and offers (with the R-D1), 22.5mm equivalent angle of view. Looks like this test is illustrating the problem of the combination of lens and sensor design.

 

[jaapv]

We've known about the new coding for a few weeks now and 8 of my lenses are back in Solms for upgrading.

How long will it take, Mark? I still have to send mine.

I'm interested that the results with the Zeiss Distagon are "variable" with digital. This is an expensive lens and offers (with the R-D1), 22.5mm equivalent angle of view. Looks like this test is illustrating the problem of the combination of lens and sensor design.

Your remark about the synergy of lens and sensor is interesting and seems to be a pretty solid reason to wait for the dedicated Leica-Leica combination.


Simbad, You'll find far more information and even more speculation on the dedicated "digital M "forum. Btw Jonathan Eastland may be a long-time RD-1 user, but he is one of the all-time Leica gurus and has been publishing about Leica for decades.

 

[Jim Watts]

Your remark about the synergy of lens and sensor is interesting and seems to be a pretty solid reason to wait for the dedicated Leica-Leica combination.
QUOTE]

It will be interesting to see what more than EXIF data Leica include with the lens software combination.

If DxO Labs were at some point in the future to include digital rangefinders and many of their available lenses in their excellent Optics Pro software package "chipping" although helpful (for quick recognition of the lens, etc.) would not be essential to a resonably automatic correction of lens/sensor aberations.


http://www.dxo.com/en/photo/home/default.php

http://www.northlight-images.co.uk/article_pages/dxo_35.html

 

[saxshooter]

Maybe this has been stated already but I am guessing part of the camera "knowing" what lens on the camera will result in some in-camera processing to minimize vignetting caused by wider angle lenses?

 

[jaapv]

Yes. And CA. (Chromatic aberration)

 

[Mark Norton]

How long will it take, Mark? I still have to send mine.

No word back from them yet other than that they have arrived, so DHL did a great job! If I was in the Netherlands, I'd almost take a day trip to drop them off.

 

[J. Borger]

I just wonder what the function of this magic in camera processing will be if shooting raw?
I just wonder why people are sending the lenses already in for a camera they know nothing about??
I will sent nothing in untill i know more about the camera and the raw files!
If the M8 does not deliver with the older lenses (my favourite : the 35 lux pre-asph is not even on the upgradable list) .. i'll buy another R-D1 or go film and buy an M7 !!

 

[Dougg]

I just wonder why people are sending the lenses already in for a camera they know nothing about??

As I recall Leica anticipated a heavy demand for modifying older lenses and advised acting sooner to avoid long waits. It certainly does no harm to have the lens upgraded and should improve its future salability. And then the lens is ready to fully communicate with an M8 should one land in your lap some day. 😀

 

[Mark Norton]

It may well be the in-camera processing will be replicated in a raw plug-in but they will both benefit from knowing the lens being used without it being manually selected, even if that is possible. I could imagine that the "special" processing would only be available in the raw plugin if the lens type were known, both to thwart the use of third party lenses and also to avoid errors.

Most of us here will shoot raw but for in-camera processing, it's certainly to Leica's advantage that JPEG images are fully usable straight out of the camera. I certainly do not want to spend time processing every raw image and if the JPEGs have the benefit of the processing already applied, that's more than fine.

I've sent my lenses in to beat the queue and because I'm intrigued. I don't need to wait to be convinced of the benefits. And Dougg mentions "should an M8 land in my lap". I'd use the word "When"...

 

[David Kieltyka]

Any in-camera processing can be duplicated in RAW conversion software. Adobe's Camera RAW already includes light falloff (vignetting) and chromatic aberration correction. You can create lens-specific templates for this too, which is what I've done for the 15mm Heliar on the R-D1. And of course Epson's RAW software offers the same features. Leica's scheme potentially saves you some time by including lens info in the EXIF data, which is nice.

Now if Leica intends to alter the RAW data before saving it as RAW based on the lens in use...I'd personally be opposed to such a thing. It would be fine for JPEGs but I'd prefer the RAW data to be as non-manipulated as possible.

-Dave-

 

[lightwriter]

Anyone know what the cost is to get lenses "updated"? I have not this information posted.

 

[ZeissFan]

I thought I saw 85 euros per lens -- that's about USD90 or so, maybe a bit more. Figure $100 or more per lens after you include shippping.

Seems costly but certainly cheaper than buying one lens.

 

[Ergo]

"The coding is not a chip, it's simply a pattern of black and white dots engraved in the lens mount surface "

i wonder if they will do ltm>m adapters

 

[J. Borger]

Thanks for the clarification Doug.
I never ever shot a Jpeg in my digital life, which goes back to the days of the canon D30.
So i just hope Leica Raws are real raws and there are correction tools in the software too ..... this would make the dotting obsolete for me.

 

[RML]

"The coding is not a chip, it's simply a pattern of black and white dots engraved in the lens mount surface "

i wonder if they will do ltm>m adapters

Once someone figures out the sequences of B&W dots (and that will be pretty soon, I reckon), someone will maqke cheap stick-on sequences. Those could probably be used on ltm-m adapters as well, as long as you stick with 1 focal length on that adapter and don't go switching about. Whether it'll be very useful, however, will depend on whether Leica will make sequenses for ltm leica/leitz lenses. That, I guess, probably won't happen.

 

[Dougg]

The cheap dot pattern stickers would have to have a nasty acid in the adhesive to eat into the metal surface enough to make itself flush with the surrounding flange...

 

[Didier]

Eur 95

Eur 95

Anyone know what the cost is to get lenses "updated"?

EUR 95 as announced on the Leica website, see http://www.leica-camera.com/imperia/md/content/pdf/objektive/18.pdf
That's about $119.

Though I believe the coding will improve the M8's digital performance, I fear this pricing policy is grist for the mill of those who pretend Leica is for retired lawyers and dentists only...