6-bit coding and Auto Lens Detection

[jaapv]

Advertisement (gone after registration)

Simple. Two switches activated by the mechanism will do the trick. 

[lct]

1 minute ago, jaapv said:

Simple. Two switches activated by the mechanism will do the trick. 

This does not explain why coding a 50mm lens as 35mm makes the camera recognize it as a 35 in spite of framelines being 50/75. If lens detection depended on the selected frameline, the same lens should be recognized as 50 or 75 instead of 35.

[misteracng]

I believe the frameline position is a secondary input which is not used that much right now, the 6 bit code is priority. I don’t believe there are that many lenses having the same 6 bit code yet for this to matter like the MATE but the frameline selection can allow more then one lens to use the same 6 bit code, as long as they do not use the same frameline set. It’s a pretty ingenious solution to lens detection without electronics.

[mujk]

2 hours ago, misteracng said:

I believe the frameline position is a secondary input which is not used that much right now, the 6 bit code is priority. I don’t believe there are that many lenses having the same 6 bit code yet for this to matter like the MATE but the frameline selection can allow more then one lens to use the same 6 bit code, as long as they do not use the same frameline set. It’s a pretty ingenious solution to lens detection without electronics.

Yes, this seems to be the case. But, except for the MATE, what exactly has Leica had in mind? At least based on a few quick tests I made yesterday with the M9 and M10, the implementation does not seem to be very consistent.

With some coded lenses I happen to own (21 SEM, Summarit 75) the frame selector does not seem to have any effect at all, while with a few others changing the lever position changes lens type to "uncoded". The exception is the Cron 35 asph v1, where it also changed to 2.0/50mm in the 28/90 position, but on the M10 only. I don't have an M11 at hand to test with.

[jaapv]

7 hours ago, lct said:

This does not explain why coding a 50mm lens as 35mm makes the camera recognize it as a 35 in spite of framelines being 50/75. If lens detection depended on the selected frameline, the same lens should be recognized as 50 or 75 instead of 35.

You asked about how the detection could work physically. In other words could be read from a mechanical system and used as electronic data.
 

Now you are talking about how the camera handles the detected value which appears to be a bit sloppy on the M 11 judging by this thread. 

[lct]

7 hours ago, misteracng said:

I believe the frameline position is a secondary input which is not used that much right now, the 6 bit code is priority. I don’t believe there are that many lenses having the same 6 bit code yet for this to matter like the MATE but the frameline selection can allow more then one lens to use the same 6 bit code, as long as they do not use the same frameline set. It’s a pretty ingenious solution to lens detection without electronics.

Not sure to follow you:
(a) If 2 lenses (35 & 50 for example) use the same 6 bit code (35) with the same frameline set (35/135), there is no conflict between code and framelines, so both lenses will be detected as 35mm as expected.
(b) If said 35 & 50 lenses use the same 6 bit code (35) with another frameline set (50/75), such conflict arises and the code prevails (35) rending the framelines useless for lens detection.
(c) If said 35 & 50 lenses use the same 6 bit code (35) with 2 different frameline sets (35/135, 50/75), another conflict arises and the code still prevails (35) rending again the framelines useless for lens detection.
(d) Hence the question to know if framelines play whatever role in lens detection.
Please feel free to correct me if needed.

[jgeenen]

Advertisement (gone after registration)

I did a little experiment - just out of curiosity. If I attach the APO-Summicron 2/35 or the APO-Summicron 2/75 or an APO Lanthar 50mm coded as APO-Summicron to my M11. Both are factory coded and the 6-bit code is correctly recognized. If then press the frame lever, the displayed lens switches the detection type to "manual", and selects 2,8/50 which is (my) first manual override setting. If I do the same with my 1.4/35 Summlilux asph or or a Elmarit 90mm coded Skopar, nothing changes.

So it seems that Leica has implemented some detection mechanism, but that is not consistently used throughout the range of lenses. I would guess (without being able to test the behavior), that the Tri-Elmar 28-35-50 might the only lens that makes use of this mechanism.

It however seems obvious that Leica is not using this lever information to differentiate two lenses with the same 6-bit-code (because I haven't found any except where the optical formula is close).

[UliWer]

Please remember that it is a "Bit-Code". The frameline selector adds a bit with two possible values: either "0" or "1" (like the painted code on the lens: white is "0", black is "1". So if a lenses code has a "1" as "lever-bit", the lever's position does matter. If it has an "0" as "lever-bit" the lever's position is insignifcant.  

Edited Thursday at 12:56 PM by UliWer

[lct]

6 minutes ago, UliWer said:

Please remember that it is a "Bit-Code". The frameline selector adds a bit with two possible values: either "0" or "1" (like the painted code on the lens: white is "0", black is "1". So if a lenses code has a "1" as "lever-bit", the lever's position does matter. If it has an "0" as "lever-bit" the lever's position is insignifcant.  

I would challenge this for the pleasure of the discussion, but also to understand what mysterious role the frameline selection could play on lens detection. Would you have a practical example to show that such role, if any, could be in any way relevant?

[UliWer]

Well, i don‘t think that lens detection is really relevant…😉, though if you want an example to see whether the position of the lever does make a difference you may look at your posting in #11:

Quote

On 9/15/2025 at 12:33 AM, lct said:

Just tried with 3 Summilux lenses coded by Leica.
• With a Summilux 35/1.4 FLE v1, however i set the frame selector, exif data say "Summilux-M 1:1.4/35 ASPH", so priority to lens code in this example.
• Different experience with a Summilux 50/1.4 FLE v1. With 50/75 and 35/135 framelines, exif say "Summilux-M 1:1.4/50 ASPH". But with 28/90 framelines, exif say "R-Adapter M".
• Similar result with a Summilux 50/1.4 v3. With 50/75 and 35/135 framelines, exif say "Summilux-M 1:1.4/50". But with 28/90 framelines, exif data say again "R-Adapter M". Unsure where the error comes from but framelines have apparently an effect on Auto Lens Detection.

 

As I tried to explain in #14 the 35mm Summilux seems to have the value „0“ for the frameline selector, so the position of the lever isn‘t significant.

For the two 50mm Summilux it seems the lenses need a value „1“, which is provided  in two positions of the lever but not in the third one.   

Edited Thursday at 01:49 PM by UliWer

[jgeenen]

vor 2 Stunden schrieb UliWer:

Please remember that it is a "Bit-Code". The frameline selector adds a bit with two possible values: either "0" or "1" (like the painted code on the lens: white is "0", black is "1". So if a lenses code has a "1" as "lever-bit", the lever's position does matter. If it has an "0" as "lever-bit" the lever's position is insignifcant.  

In my example, all three lenses have different combinations of 6-bit (obviously) but none of them share one (black) bit. Do you know in which bit is that lever bit? 

Edited Thursday at 03:04 PM by jgeenen

[UliWer]

5 hours ago, jgeenen said:

Do you know in which bit is that lever bit? 

No, I don't know where this extra bit "lives".

The camera's electronics react on the 6 visual bits provided by the coding and the position of the lever. So it is a 6+1-bit-code.

The "lever bit" is just an "add-on" as the lever's position is defined by the lens you use. You may change the coding, but not the function of the lever. Any change of the lever's position is just a manipulation you might use to look how your motive may look with a different focal length, but it doesn't change you focal length nor any other quality of the lens. 

 With 6 bit you may detect theoretically 64 different lenses. With the "add on" by the "lever bit" the number of different detections is enhanced. You may have identical 6-bit codes for a 35, a 50 and a 90mm lens. With the different position of the lever you can detect these lenses as different ones.

I don't know whether they actually use this possibility. As the number of lenses for the M which enable lens detection is not infinite and many new versions use the same code as previous ones, I think they could well live with the 6-bit-code alone. Though the lever position is used for detecting some lenses as you can see if you manipulate the lever: the camera will either detect a different lens or will "think" the lens was uncoded. If there is no such reaction the value of the "lever bit" is "0".  

[lct]

2 minutes ago, UliWer said:

The camera's electronics react on the 6 visual bits provided by the coding and the position of the lever.

Do you have an evidence of that?

[UliWer]

I can only repeat my offer that you look at your own posting in #11 and read again the description you gave for the 50mm Summilux.

[lct]

I remember this experiment thank you but i does hardly prove that framelines are taken into account for lens detection so far.

Edited Thursday at 10:07 PM by lct

[jaapv]

57 minutes ago, lct said:

Do you have an evidence of that?

Just try…

[lct]

2 minutes ago, jaapv said:

Just try…

I did but  cannot say that i am convinced that framelines are taken into account for auto lens detection.

[UliWer]

 

Quote

40 minutes ago, lct said:

but i'm only a retired lawyer...

 

Same as me. Though it is not so long ago to remember that every "evidence" needs to be interpreted and there are entities which won't be convinced by them.

First of all my wording "lever" is short for "activated framelines". It's of course not the lever which delivers the information, but the framelines activated by the lens. The first version of the M (Typ 240) had no lever for frameline selection, but of course the same mechanism. "Lever" is just a short and - perhaps - more obvious description.

The 1:3.8/24mm Elmar (coded by Leica) is detected as such with the 35/135mm framelines the lens activates. When I change the framelines to 50/75mm it says "uncoded", with the framelines for 28/90mm it says "2/35". 

The 1:2.5/35 Summarit (coded by Leica) is detected as such with the proper framelines for 35/135mm. If I change the framelines with the lever the detection says "uncoded" with the frames for 50/75mm and "2.8/50" with the frames for 28/90mm.

The 1:2/75mm Apo-Summicron (coded by Leica) is detected as such by the proper framelines for 50/75mm and by the framelines for 35/135mm. With the framelines for 28/90mm it says "uncoded". 

The  1.2/50mm Apo-Summicron is detected as "2/50" with the proper framelines for 50/75mm and for 35/135mm. With the framlines for 28/90mm it says "uncoded".

The 1:1.4/50mm Summilux asph.  as well as the spherical former version of the 50mm Summilux (both coded by Leica) detected as "1.4/50"with their proper framelines 50/75mm and with the frames for 35/135mm. With the frames for 28/90mm it says "uncoded".

The 3.8/18 and 3.4/21mm Super-Elmars, the 2/28mm Summicron asph, 1,4/35mm Summilux asph. FLE, the 1:2/90mm Summicron Apo-Asph and the 1:4/90mm Macro Elmar (all coded by Leica) are detected as such with the lever in all three possible positions. 

No other coded lenses available.

 

 

 

[jaapv]

M9, Summicron 35 asph. Frameline lever untouched, pushed halfway, pushed all the way Only the correct frameline reads the lens. QED. 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[lct]

26 minutes ago, jaapv said:

M9, Summicron 35 asph. Frameline lever untouched, pushed halfway, pushed all the way Only the correct frameline reads the lens. QED.

True with this lens but not with all coded lenses by far. With the Summilux 35/1.4 FLE, for instance, the 3 framelines read the lens.