6-Bit Encoding Needed for 50mm and Longer?

[CJH]

Advertisement (gone after registration)

While I'm waiting for a replacement M8 -- and since I'm traveling for much of the next month or so -- it makes sense to send lenses off to be encoded, certainly my 24 and 35. While I'm packaging things up, is there value to sending one or both of my 50s as well (Elmar or Nocti)? I believe I've read that all this really does is allow the lens to show up in the EXIF file. Is there a chance that down the road, the firmware might allow something more to happen with these lenses that would make it valuable to have them encoded?

 

I see little reason to encode my 90, as I don't think I'll be using it all that much.

 

Thanks for the advice to an M8 newbie!

 

Chris Herbert

[Advertisement]

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

[lars_bergquist]

Broadly speaking, no. There are neither any normal 'brightness vignetting' nor cyan vignetting to worry about from 50mm and up. The case is actually a marginal one even at 35mm. My own Summilux ASPH. is uncoded, and one of my 35mm lenses.

 

The situation may of course change if there is ever presented a full format digital M, maybe the hypothetical M9. This would mean that a lens of a certain focal length will have a larger field of view, with a more oblique angle of incidence of peripheral rays of light. But we are not there yet.

 

The old man from the Age of Film

[stunsworth]

My coded Nocti seems to vignette less than I remember it doing on film - taking into account that only the central part is used on the M8. I could be imagining this though as I haven't done any proper tests.

[c6gowin]

I decided to get all of my lenses coded, including the 50 and 90, as a matter of convenience. This way I can select lens detection to the "ON-UV/IR" setting in the menu and forget it - no need to change this setting when switching between wide angle and telephoto lenses. This setting may not make that much of a difference on the medium to long lenses, but I knew I wanted to use the UV/IR filters so I thought coding made sense.

 

Mark Gowin

[stunsworth]

That's a good point Mark, earlier this summer I was using an uncoded Voigtlander 90mm together with some coded Leica lenses. On a few of the shots the camera thought that a 21mm lens was fitted, and corrected the non-existant cyan drift accordingly.

[lars_bergquist]

I decided to get all of my lenses coded, including the 50 and 90, as a matter of convenience. This way I can select lens detection to the "ON-UV/IR" setting in the menu and forget it - no need to change this setting when switching between wide angle and telephoto lenses. This setting may not make that much of a difference on the medium to long lenses, but I knew I wanted to use the UV/IR filters so I thought coding made sense.

 

Mark Gowin

 

There is no need to code any reasonably modern Leica 90mm lens, they are all self-coding! The reason seems to be a screw head in exactly the right place. At least this is true for both my 4-element Tele-Elmarit and my pre-aspherical Summicron.

 

The old man from the Age of Spherics

[gesper]

Advertisement (gone after registration)

I use my CV75 with a filter but no coding and it works great. Sean Reid's review confirmed that this is the case.

[c6gowin]

I had problems with magenta cast in some of my 90 people photos and knew I wanted to use a UV/IR filter from then on. I got it coded thinking that it would be better that way with a filter. I didn't know it was inadvertantly coded by screw in the mount. You got me curious to go back through some old phots (before coding) to see if I can tell a difference although it could be hard to tell since I added the filter at the same time.

 

Mark Gowin

[peterv]

I had problems with magenta cast in some of my 90 people photos and knew I wanted to use a UV/IR filter from then on. I got it coded thinking that it would be better that way with a filter. I didn't know it was inadvertantly coded by screw in the mount. You got me curious to go back through some old phots (before coding) to see if I can tell a difference although it could be hard to tell since I added the filter at the same time.

 

Mark Gowin

 

Mark,

 

you could also take a look in the EXIF-file, and see if the 90 mm focal lenght is mentioned.

 

Peter

[CJH]

There is no need to code any reasonably modern Leica 90mm lens, they are all self-coding! The reason seems to be a screw head in exactly the right place. At least this is true for both my 4-element Tele-Elmarit and my pre-aspherical Summicron.

 

The old man from the Age of Spherics

 

 

Any idea whether that self-coding is also true for the 90 Elmarit-M?

 

Chris

[jonoslack]

I was rather against it to start with.

But now I agree with Mark - it's just much easier if everything is coded as you don't have to change anything in the menu settings.

with a 90 it doesn't much matter what you do, but if you then put on a wide angle with a filter without changing settings . . . . cornerfix here we come!

[pklein]

That's a good point Mark, earlier this summer I was using an uncoded Voigtlander 90mm together with some coded Leica lenses. On a few of the shots the camera thought that a 21mm lens was fitted, and corrected the non-existant cyan drift accordingly.

 

Steve: That happened to me, too, using a VC 90/3.5 Lanthar and a standard V/C adapter. I'm not home right now, so I can't check this for sure, but I'm wondering if there is a screw head lined up with the sensors.

 

If so, a little white nail polish on the screw head ought to remove any ambiguities between black and white, such as the shadow in the screw slots. I've head that on some of the older Leitz 90s this is an issue.

 

Related: I got my 35/2 Summicron IV back from DAG recently. I asked him to send the mount to John Milich for coding. There is a screw in the sensor path where part of a black dot should be, and it came back from Don with a black screw head :-) The coding works fine.

 

To answer the original question, I am using coded 28mm and 35mm lenses. Even with 35mm, if you have white things near the edges, you will see a touch of green if you don't code. From 50mm and up, it really doesn't matter. Except for "misunderstandings" such as the Steve mentions, if the camera doesn't see a valid code, it doesn't appear to apply corrections, and you can leave the lens detection set to "On plus UV/IR" all the time.

 

If you want the lens data in the EXIF, then you have to code.

 

--Peter

[CJH]

Thanks for all the quick replies. Informative and much appreciated!

 

I'm not afraid of a bit of overkill, so I've just sent off my 24, 35, and my Nocti for coding... damn scary to put a replacement value on that box, I'll tell you! That saves a 50 to shoot uncoded (but I do have an IR filter for it) until the coded lenses return. The idea of not having to change settings when using different lenses appeals to me.

 

I'll play with my 90 Elmarit and see how it does uncoded. It may or may not get sent in at some point. Of course, having a properly working M8 is the key to all else! Things bode well for that to arrive before too long, once I get back from a couple trips for business and a bit of pleasure. Yes, the pleasure would have been enhanced by the M8, but the D-Lux 3 will have to suffice.

 

Will report on my progress getting used to this wonderful camera!

 

Chris

[sean_reid]

Broadly speaking, no. There are neither any normal 'brightness vignetting' nor cyan vignetting to worry about from 50mm and up. The case is actually a marginal one even at 35mm. My own Summilux ASPH. is uncoded, and one of my 35mm lenses.

 

The situation may of course change if there is ever presented a full format digital M, maybe the hypothetical M9. This would mean that a lens of a certain focal length will have a larger field of view, with a more oblique angle of incidence of peripheral rays of light. But we are not there yet.

 

The old man from the Age of Film

 

What is cyan vignetting?

 

Cheers,

 

Sean

[Gene]

Any idea whether that self-coding is also true for the 90 Elmarit-M?

 

Chris

I have a 90mm tele-Elmarit from the early 1970s, and I noticed that the exif data with its photos includes statements to the effect that the focal length of the lens is 18mm and that the equivalent focal length with 35mm film would be 24mm.

 

None of my other old M lenses even get any focal length statements in the exif data.

 

-Gene

[J_Brittenson]

I agree with the notion that it doesn't terribly matter with Summicron 35's either. Perhaps a Summilux used wide open though.

[Hacker]

I was rather against it to start with.

But now I agree with Mark - it's just much easier if everything is coded as you don't have to change anything in the menu settings.

with a 90 it doesn't much matter what you do, but if you then put on a wide angle with a filter without changing settings . . . . cornerfix here we come!

 

I wonder what will happen if the M9 is released and there is no issue with the magenta cast anymore, and therefore, no need to use the UV/IR filters.

[Paul_S]

I might have a slightly different angle of view re. this topic:

Coding yes/no in general depends on your camera (lens detection) setting in the first place, not the lens or its focal length. Lens detection is a 'set and forget' setting. If you choose to set detection to ON, then consequently it is best to code all lenses. We want the best results, don't we?

 

If you use an uncode lens with lens detection ON, you are in for unpredictable results. If you want to take that risk, it is up to you. Results vary, due to type of lens. Where the frame selector, set by the lens' flange, is of any inflluence, I typically expect more deviations for 21/28/90 than for 50/75 lenses. Postings here confirm that behavior. I also like the 'screw in flange coding myth'. I do not deny its effect (or I will get flamed ) but it is not the screw (tele-elmarit, pre asph s'cron 90); I have/had them myself, and the s'cron90 pre asph was not detected as any lens on my M8. My uncoded Noctilux was detected as a 1.4/50 on the same camera. Did I already say unpredictable?

 

The use of UV/IR filters makes coding more a necessity IMHO, assuming you set the camera's lens detection accordingly.

[Advertisement]

Hallo,

Look here: Leica M

[tom0511]

I might have a slightly different angle of view re. this topic:

Coding yes/no in general depends on your camera (lens detection) setting in the first place, not the lens or its focal length. Lens detection is a 'set and forget' setting. If you choose to set detection to ON, then consequently it is best to code all lenses. We want the best results, don't we?

 

If you use an uncode lens with lens detection ON, you are in for unpredictable results. If you want to take that risk, it is up to you. Results vary, due to type of lens. Where the frame selector, set by the lens' flange, is of any inflluence, I typically expect more deviations for 21/28/90 than for 50/75 lenses. Postings here confirm that behavior. I also like the 'screw in flange coding myth'. I do not deny its effect (or I will get flamed ) but it is not the screw (tele-elmarit, pre asph s'cron 90); I have/had them myself, and the s'cron90 pre asph was not detected as any lens on my M8. My uncoded Noctilux was detected as a 1.4/50 on the same camera. Did I already say unpredictable?

 

The use of UV/IR filters makes coding more a necessity IMHO, assuming you set the camera's lens detection accordingly.

 

 

In theory I agree however so far I have not detcted any problems when using my uncoded 50/1.5 or 90/2.8.

 

When using the Zeiss even when the setting is on correction "on+UVIR" the exif does not show anything, so it seems to realize that a not-coded lens is used.

So why not check your own (not-coded) 50/75/90mm lenses and see if your M8 detects something wrong or detects just nothing and decide afterwards?

Cheers, Tom

[Paul_S]

So why not check your own (not-coded) 50/75/90mm lenses and see if your M8 detects something wrong or detects just nothing and decide afterwards?

Cheers, Tom

 

Tom, makes sense. But in some occasions users have reported inconsistent detection of a single uncoded lens, in particular for a 90mm. One day it is fine, the next the camera handles it as a 21mm with filter. That does not make you happy!

 

Bottomline is I think, that uncoded/detection ON works most of the time for >=50mm focal length, as long as you accept the risk (after you became aware of it).