Proof of concept regarding self coding of lenses

[mboerma]

Hi
Can anyone tell me if the leds on the camera should go on when taking a picture with the lens detection on? I tried a couple of codes and ways to code but I can't seem to get it working.

Thanks,
Maarten

[mboerma]

In addition to the above post:

I do have a tip to be able to remove the ink of a permanent marker. I use a whiteboard marker and go over the dots I made with the permanent marker and than use a tissue to remove it al together.

Maarten

[sean_reid]

Originally Posted by Philippe D.

I think all that stuff is much too thick.
It will not last long, quite shorter life then the bare marker spots, whitch are 1/250 mm think only.
That why it as been made in a hollow place by Leica. They just cogitate hard before doing it.

Any kind of marking directly on the adapter (paint, marker, etc.) will wear off eventually. I say this from very extensive experience <G>. There just isn't meant to be any clearance between the mount and the adapter surface. John Millich has helped various members here by milling adapters (sent to him) with the correct indentations for the 6-bit codes. He doesn't sell these, just does this as a favor to forum members as his time allows. All of my Leitz adapter rings have been milled by John so that all my 35 mm and wider LTM lenses are durably coded. I paint in the white and black using enamel nail polish.

To my knowledge, patent issues may prevent anyone (other than Leica) from selling coded adapters rings but various individuals are finding machinists who can do the work.

Here's an example:

Cheers,

Sean

[sean_reid]

Here's a CV 35/1.7 with a Leitz 13.5 cm adapter that was milled by John and then painted with enamel nail polish.

Cheers,

Sean

[valtof]

Originally Posted by mike prevette

I Just coded my 28mm Voigtlander and my Zeiss 21mm. The interesting thig with the Zeiss is that since it doesn't bring up the proper framelines on the M8 I can turn the lens detection on or off with a simple press of the frame lines selection switch. The coding is indeed tied to what frame lines the M8 is displaying. This is a great little bonus. I love bieng able to tell the camera when to do the correction and when not to.

_mike

Hi, Mike and all of you.
1/ I've just succeeded in coding my Elmarit f:2.8 / 90 without any problem at the first trial.

2/ I've also succeeded with my Elmarit f:2.8 / 21 but it was harder, because of the screw problem, as you probably know.
For those who are interested, here is the way I did : I filled the screw hole with a little piece of white adhesive paper and then I put the only necessary Sharpie mark at the good position (000001). So far, It didn't work and I suspected the white paper being unshiny to be the problem. So I covered all the zone with some autoadhesive transparent Covertec protection, and then It worked. Of course the lense becomes a little bit harder to mount, due to the additional thickness of the Covertec film... but it works well.

3/ All my other lenses or not Leica so far, that is the CV12, the CV 1.9/28 and the f:1.4 / 50 Canon. They all need an LTM adapter that has the problem of the missing metallic part at the edge.
As I read you succeeded with you CV28, could you explain me/us how you did ?

In your opinion, what would be the appropriate coding for the CV12, the Tri-Elmar-M 16-18-21mm ?

Thanks in advance
Cheers

[sean_reid]

The best way to code the CV 28 is to use a Leitz 9 cm adapter which will key the 28/90 frame lines. It does not have a cutout. I wouldn't code the CV 12 at all and would use it with no filter and Jamie's profiles.

Cheers,

[hsdee]

Would you care to elaborate why you wouldn't code the CV12? I saw some of your pictures using the CV12 and they were fantastic. I've posted a couple in another thread that I took with my CV12 shot RAW with IR filter and they had a green tint. I used Edmund's LoSat2 profile and was able to remove most of the green tint except for the corners that were distinctly greenish. I'm using 1.92 firmware. Is this why you recommend not using an IR filter? Will 1.10 address this problem? Does the WATE suffer the same greenish tint with the IR filter on or does the coding of the lens deal with that issue?

Thanks

Herb

[jaapv]

Sorry for the grotty photograph. From another thread: this works:

[valtof]

Originally Posted by sean_reid

The best way to code the CV 28 is to use a Leitz 9 cm adapter which will key the 28/90 frame lines. It does not have a cutout. I wouldn't code the CV 12 at all and would use it with no filter and Jamie's profiles.

Cheers,

Thanks for the quick answer. I will try and find this adapter for the CV28.
As for the CV12, I agree with you that buying its adapter and the 77mm IR/cut filter (at leat a 200$ budget) is something I decided not to do, espacially because this very expensive filter would not have any protection and the resulting overall solution would become much too obstrusive. And of course, I'm using Jamie's profiles in C1.

Moreover, my goal is more :
1/ to get the info in the metadata to be sure the picture was taken with the CV12 and not the 21 (even if the CV12 is obviously much wider, sometime it happend that I mix them up) ;
2/ When you mount the CV12 and keep the lens detection activated, it happens that the M8 detect a 35mm instead of nothing, due to the cutout of the adapter that let a gap towards the black body of the lens, I presume, given that the 1.4/35 and the 2/35 have both 4 black digits on six...

So do you think that coding the CV12 like being a Tri_Elmar 16-18-21 would lead to strange corrections, not using an IRcut filter and when the 1.10 firmware will make these corrections possible of course, which is not the case right now if I understand well ?

Cheers

[guy_mancuso]

Besides the whole filter issue the orginal intent was for vignetting in the corners and the coding would help correct that, so yes i would code the 12mm for the WATE lens at least you can the 16mm vignetting applied. Than if you want to use a IR filter than that same coding will be helpful with the new firmware that corrects the cyan cast. You may not get all the way in the corners but you will be close. I would certainly try and code any CV lens for the leica counterpart

[grdglass]

For Sean Reid:

Sean, have you tried to code your old Canon 28mm 2.8 RF lens? I assume a 9cm adapter is best. Which code is best?

Are the adapters that read "9cm" equivalent to those that read "M2 90 M3?" You have mentioned 28/90 adapters -- do you mean 9cm or 90mm adapters?

Thanks,
Helene

[sean_reid]

Originally Posted by hsdee

Would you care to elaborate why you wouldn't code the CV12? I saw some of your pictures using the CV12 and they were fantastic. I've posted a couple in another thread that I took with my CV12 shot RAW with IR filter and they had a green tint. I used Edmund's LoSat2 profile and was able to remove most of the green tint except for the corners that were distinctly greenish. I'm using 1.92 firmware. Is this why you recommend not using an IR filter? Will 1.10 address this problem? Does the WATE suffer the same greenish tint with the IR filter on or does the coding of the lens deal with that issue?

Thanks

Herb

Whom are you asking these questions to?

Cheers,

Sean

[sean_reid]

Originally Posted by grdglass

For Sean Reid:

Sean, have you tried to code your old Canon 28mm 2.8 RF lens? I assume a 9cm adapter is best. Which code is best?

Are the adapters that read "9cm" equivalent to those that read "M2 90 M3?" You have mentioned 28/90 adapters -- do you mean 9cm or 90mm adapters?

Thanks,
Helene

Hi Helene,

For the Canon 28/2.8, I use the same adapter as for the CV 28/3.5. It's a Leitz 9 cm that has been milled for the coding indents and then coded as a Leica 28 Summicron Aspherical. Vignetting correction is actually slightly stronger for the 28 Summicron Asph. than for the 28 Elmarit Asph. (as of 1.09).

The older Leitz adapters (with no cutouts) are usually marked in centimeters. There may also be some that are marked in mm or as 28/90, etc. Either way, the goal is to get one with no cutout.

Cheers,

Sean

[sean_reid]

Originally Posted by valtof

Thanks for the quick answer. I will try and find this adapter for the CV28.
As for the CV12, I agree with you that buying its adapter and the 77mm IR/cut filter (at leat a 200$ budget) is something I decided not to do, espacially because this very expensive filter would not have any protection and the resulting overall solution would become much too obstrusive. And of course, I'm using Jamie's profiles in C1.

Moreover, my goal is more :
1/ to get the info in the metadata to be sure the picture was taken with the CV12 and not the 21 (even if the CV12 is obviously much wider, sometime it happend that I mix them up) ;
2/ When you mount the CV12 and keep the lens detection activated, it happens that the M8 detect a 35mm instead of nothing, due to the cutout of the adapter that let a gap towards the black body of the lens, I presume, given that the 1.4/35 and the 2/35 have both 4 black digits on six...

So do you think that coding the CV12 like being a Tri_Elmar 16-18-21 would lead to strange corrections, not using an IRcut filter and when the 1.10 firmware will make these corrections possible of course, which is not the case right now if I understand well ?

Cheers

If one uses a 486 filter with the 12, the cyan drift will be quite pronounced. Using corrections in 1.10 that are designed for a 16 mm focal length are unlikely to be a good match. They'll decrease the degree of cyan drift, but not correct it, so you'll end up with a file that is still problematic. Coding a lens for a non-matching lens can end up being worse than not coding it at all. I recommend either:

A) Use the filter on an uncoded 12 and then develop an action in Photoshop (or other software) to provide a proper correction for the resulting cyan drift.

B) Use no filter on the lens and correct for magenta casts, as needed, with Jaime's profiles.

I have a 12 here and have been experimenting.

As for the EXIF data, code the 21 and the 12 will be the one that shows no focal length.

Cheers,

Sean

[sean_reid]

Originally Posted by guy_mancuso

Besides the whole filter issue the orginal intent was for vignetting in the corners and the coding would help correct that, so yes i would code the 12mm for the WATE lens at least you can the 16mm vignetting applied. Than if you want to use a IR filter than that same coding will be helpful with the new firmware that corrects the cyan cast. You may not get all the way in the corners but you will be close. I would certainly try and code any CV lens for the leica counterpart

Hi Guy,

Have you actually tried a filter on the 12 yet? I think you'll find its not close to the 16 at all.

Cheers,

Sean

[guy_mancuso]

Not yet trying to buy one. Yes i agree it will be off but it should get fairly close and at least much better than having the filter on and getting the cast on a big majority of the image. i would not use a IR filter on this lens until you have 1.10 in your hands unless you fix in PS. but i would still code it for the vignetting end of it, it certainly should not hurt and also would code it for the next firmware with the cyan correction which most folks know I am beta testing it and it is working, that i can say but the 12mm i have yet to try but i would assume i could fairly get close to cleaning the cyan cast up but not all of it. My thoughts are if you can code a lens to a Leica counterpart than it is a good idea and if not all we can hope for is the option in the menu item to pick your focal length which has been requested by Sean early on and myself and from forum members. That one is Leica's call if they will give us that in the future.

[hsdee]

Originally Posted by sean_reid

Whom are you asking these questions to?

Cheers,

Sean

Hi Sean,

I was addressing the questions to you.
Thanks.

Herb

[sean_reid]

Originally Posted by hsdee

Hi Sean,

I was addressing the questions to you.
Thanks.

Herb

Hi Herb,

I don't recall posting pictures with the 12 but maybe I did and forgot. What were they? Sorry for my old and foggy mind. The reason I wouldn't recommend coding a 12 right now is that it's a very different lens from a 16. Corrections that Leica will have in 1.10 for the WATE at 16 won't be appropriate for a 12. Cyan drift gets quite strong with a lens that wide. If Leica introduces its own coded 12 that would be a different story because there would then be a potentially good reference lens for the CV12

Cheers,

Sean

[mike prevette]

Sean is spot on like always. The Cyan correction for the 12 would be massive. Also My 2 cents regarding people codeing the 15 to be the WATE is that it is also going to result in some abstractions. The optical design of the WATE is considerable different than the CV15, the exit pupils are quite a bit different. Hence the sensor vignetting will also be very different. The cyan correction will remain the same because the filter is in front of the lens, but the luminance vignetting correction will be wrong. Those of you who have done it might not notice any ill effects, in which case your lucky.

The light inverts and exits the lens of the WATE at a much greater distance from the focal plain than the CV15 hense a shallower angle of attack for the light rays hitting the senor, hence less sensor vignetting from the WATE.

_mike

[guy_mancuso]

Mike i have seen results from both lenses one including my own and with the leica filters they look very good. I'm using a 49mm over the WATE which i was warned may cause trouble and it is dead on.