6-bit Coding Question

[Mango]

Hi, Two basic questions:

(1) The prime lenses have a unique code, as do the Tri-Elmars. The 6-bit coding has been stated as being necessary to adjust for the color fringing, etc. My question is, how does the M8/8.2 camera know to change the coding algorithm for the Tri-Elmar lenses (16-18-21 and 28-35-50) when the focal length has been changed, because (to my basic understanding), all that changes is the frame line in the viewfinder. Does anything else change by changing the focal length? I thought the focal length has to be exact in the computer of the camera so that the proper adjustment can be made.

(2) I don't know enough about how the 6-bit coding system works, but I keep wondering why Leica doesn't use a system whereby you just punch in the focal length and the max aperture (just as Nikon does with the manual lenses on D2xx< cameras). Why do they make it so difficult (with uncoded lenses requiring a trip to Leica facilities) when they could just have some firmware to handle ANY lens that could be mounted on the M8.

 

[jonoslack]

Hi There

Hi, Two basic questions:
(1) The prime lenses have a unique code, as do the Tri-Elmars. The 6-bit coding has been stated as being necessary to adjust for the color fringing, etc. My question is, how does the M8/8.2 camera know to change the coding algorithm for the Tri-Elmar lenses (16-18-21 and 28-35-50) when the focal length has been changed, because (to my basic understanding), all that changes is the frame line in the viewfinder. Does anything else change by changing the focal length? I thought the focal length has to be exact in the computer of the camera so that the proper adjustment can be made.

With the 28-35-50 tri elmar, when you change the focal length, the framelines change, and the camea puts the extra information in the exif information and deals with the IR / vignetting information in the right context.
On the WATE (16-18-21) you are asked which focal length you are using when you switch the camea on.

so, in each case the camera knows the focal length.


(2) I don't know enough about how the 6-bit coding system works, but I keep wondering why Leica doesn't use a system whereby you just punch in the focal length and the max aperture (just as Nikon does with the manual lenses on D2xx< cameras). Why do they make it so difficult (with uncoded lenses requiring a trip to Leica facilities) when they could just have some firmware to handle ANY lens that could be mounted on the M8.[/QUOTE]

Now there is a question! and one which has been asked many times. The cynical answer is that they make money out of coding.

 

[Bob Parsons]

..................................Now there is a question! and one which has been asked many times. The cynical answer is that they make money out of coding.

.................. and it makes it more difficult to use other manufacturers wide angles. As Jono says the question has been asked many times and I believe Leica's official response was that they wouldn't allow manual lens entry because "people could enter the wrong data." Make of that what you will!


Bob.

 

[Mango]

.................. "people could enter the wrong data." .....

I don't know whether to laugh or cry.

 

[scott kirkpatrick]

I don't know whether to laugh or cry.

I suggest laughing. Crying will not have the slightest effect.

scott

 

[GrahamWelland]

.................. and it makes it more difficult to use other manufacturers wide angles. As Jono says the question has been asked many times and I believe Leica's official response was that they wouldn't allow manual lens entry because "people could enter the wrong data." Make of that what you will!


Bob.

The irony is of course that with the WATE that's exactly what you do ... (other than it recognizes the WATE is present). I admit though that I regularly forget to enter the correct focal length when I change it. :angry: It's almost enough to have me want to resort back to a couple of fixed primes again instead.

But ultimately it's Leica making it difficult for non-coded Leica glass. A $$$ thing.

 

[Erik Five]

If youre lenses is user lenses anyway you can just use a dremmel. Works like a charm. I never coded any of my lenses besides a 50 Lux asph that was coded when I bought it. I used 35 a lot, but I also use B&W in Exposure2 a lot so I guess im biased. I actually coded a couple of Zeiss lenses the Millich way. Worked great. You can also just buy the coder kit and do it with a marker...

 

[marknorton]

The corrections applied are mostly (all?) to do with sensor vignetting - the decreasing sensitivity of the sensor towards the edge of the frame due to the increasing angle of incidence which varies by lens type. Add an IR cut filter and there's additional correction to be done because of the red vignetting - the further you are from the optical axis, the more visible red light is removed.

The lens coding system does nothing more than identify the lens to the camera - not just the focal length but the optical fornulation - which tells the camera what corrections to apply. The M8 is not over endowed with processing horsepower so it does a basic correction of levels according to the distance from the optical axis.

The original Tri-Elmar adjusts the framelines as you change the focal length to one of three click-stopped positions and the camera senses this to determine the correction required. Leica skimped on the design of the wide angle Tri-Elmar so that there is no coupling of selected focal length to the camera - you have to select it manually. This lens is said to be a true zoom (albeit of extremely modest zoom ratio 1:1.31) amd I'm looking forward to trying it out on a G1 but the M8 only knows how to correct the image at each of the three specific focal lengths.

 

[scott kirkpatrick]

More on what the M8 and M8.2 do with the lens identification -- the vignetting corrections used when lens identification is ON but no UV/IR filter is in front of the lens were written before the first M8's were shipped, and have been available from the earliest versions. They give a quite strong correction for luminance vignetting, and a weak correction for the green cast in the corners that results from the (weak) IR absorptive filter that is in the sensor cover glass. Since the red vignetting in the absence of a filter depends only on the angle made by light rays leaving the back of the lens, one routine was sufficient for the WATE at any of its three focal lengths, which are changed by moving an internal lens group, not the rear element.

With the decision to use IR filters in front of the lens, a second set of de-vignetting routines were released. These appear to be independent of the first set, and are enabled when you set lens detection "ON-UV/IR". The red vignetting with the WATE now depends on the angle at which light enters the front of the lens, so it was necessary to add the dialog in which you tell the M8 which WATE setting you want to use. In these routines the correction for luminance vignetting seems to be about half as strong. The correction applied to red vignetting is of course stronger for the short focal lengths. Shooting in b/w or with focal lengths of 50 or more mm, this gives you some options.

scott