scott kirkpatrick Posted August 17, 2016 Share #21 Posted August 17, 2016 Advertisement (gone after registration) I've seen mostly color shifting and vignetting corrections with R and M lens profiles, not geometric correction, but then I haven't really examined the latter very closely as yet. There are certainly geometric corrections applied if I use Lightroom's lens profiles after the fact. The test to see whether Leica or Adobe is introducing geometric corrections requires going to Adobe's SDK to read the manufacturer's parameters for "Warp_Rectilinear" corrections. My shortcut is to look at the lens profile corrections which are made explicit in Capture One. The exist for the SL's 24-90 lens, but for my 15/2.8 R I haven't seen any corrections. If you could post a link to one of your 15 or 19mm R lens DNG files, I would be interested to see if they contain Leica-supplied corrections. scott Quote Link to post Share on other sites More sharing options...
Advertisement Posted August 17, 2016 Posted August 17, 2016 Hi scott kirkpatrick, Take a look here Leica R-Adapter L available. I'm sure you'll find what you were looking for!
thighslapper Posted August 17, 2016 Share #22 Posted August 17, 2016 For curiosity and guidance: Anyone out there to comment on the previous post? Thanks... after a bit of phoning around I have found one and it arrives tomorrow (and cancelled my order elsewhere). will check with the 100/2.8 apo with stacked adapters +/- elpro for you then ..... 1 Quote Link to post Share on other sites More sharing options...
helged Posted August 17, 2016 Share #23 Posted August 17, 2016 after a bit of phoning around I have found one and it arrives tomorrow (and cancelled my order elsewhere). will check with the 100/2.8 apo with stacked adapters +/- elpro for you then ..... Thanks! An image of the opening of the new adapter vs that of the (existing) stacked adaptors should clarify whether corner shading can be expected with some lenses (like 280 APO-R f4 and 100 APO-R f2.8 with ELPRO). Quote Link to post Share on other sites More sharing options...
scott kirkpatrick Posted August 17, 2016 Share #24 Posted August 17, 2016 after a bit of phoning around I have found one and it arrives tomorrow (and cancelled my order elsewhere). will check with the 100/2.8 apo with stacked adapters +/- elpro for you then ..... I'll be curious to see how it translates the APO 100 with adapters (one or both) do yours have rom connections? scott Quote Link to post Share on other sites More sharing options...
ramarren Posted August 17, 2016 Share #25 Posted August 17, 2016 The test to see whether Leica or Adobe is introducing geometric corrections requires going to Adobe's SDK to read the manufacturer's parameters for "Warp_Rectilinear" corrections. My shortcut is to look at the lens profile corrections which are made explicit in Capture One. The exist for the SL's 24-90 lens, but for my 15/2.8 R I haven't seen any corrections. If you could post a link to one of your 15 or 19mm R lens DNG files, I would be interested to see if they contain Leica-supplied corrections. Your inspection test presumes that the Leica lens profile corrections are applied by the DNG parametric correction mechanism. But we know that Leica lens profiles used in M and SL bodies can be applied directly to the raw data on its way to the JPEG engine or to the DNG file outputting code, and can include both parametric corrections as well as direct manipulation of the photosite data. So it's not a particularly good way to test whether there are corrections being made. Quote Link to post Share on other sites More sharing options...
scott kirkpatrick Posted August 17, 2016 Share #26 Posted August 17, 2016 (edited) Your inspection test presumes that the Leica lens profile corrections are applied by the DNG parametric correction mechanism. But we know that Leica lens profiles used in M and SL bodies can be applied directly to the raw data on its way to the JPEG engine or to the DNG file outputting code, and can include both parametric corrections as well as direct manipulation of the photosite data. So it's not a particularly good way to test whether there are corrections being made. There's some literature on this. It does seem that the natural point at which to make distortion corrections is after you have demosaiced the raw image and can shift the R,G, and B data captured in each pixel into the positions where they would lie in an undistorted image. You can do this in camera at some point in the JPEG pipeline, of course, but with the DNG file, it is simplest and fastest to pass the parameters along and let the work be done outside. By contrast, the vignetting and color shift corrections are best applied by LUTs that map the raw pixels into corrected values. That's how it was done in the M8 (the last case in which you could actually read the firmware code). scott Edited August 17, 2016 by scott kirkpatrick Quote Link to post Share on other sites More sharing options...
thighslapper Posted August 18, 2016 Share #27 Posted August 18, 2016 Advertisement (gone after registration) Thanks! An image of the opening of the new adapter vs that of the (existing) stacked adaptors should clarify whether corner shading can be expected with some lenses (like 280 APO-R f4 and 100 APO-R f2.8 with ELPRO). The corners visible in your images from the Apo 100 are absent with the R-L adapter ...... even with 2x adapter and ELPRO attached. The aperture in the adapter is 38x28mm and sensor shaped so theoretically there should be no issues with any R lenses ....... the issue is the round aperture in the M-T adapter. Works fine with ROM lenses and recognises the model ...... although in liveview the screen bizarrely show the lens as 99mm ........ aperture guessing seems accurate. The tripod attachment that screws on the bottom is handy and well designed. Price is extortionate for what it is .... but that's Leica for you ..... 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! 9 Quote Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/262576-leica-r-adapter-l-available/?do=findComment&comment=3098229'>More sharing options...
helged Posted August 18, 2016 Share #28 Posted August 18, 2016 Thanks thingslapper, much appreciated! I will check the availability around... Quote Link to post Share on other sites More sharing options...
ramarren Posted August 18, 2016 Share #29 Posted August 18, 2016 There's some literature on this. It does seem that the natural point at which to make distortion corrections is after you have demosaiced the raw image and can shift the R,G, and B data captured in each pixel into the positions where they would lie in an undistorted image. You can do this in camera at some point in the JPEG pipeline, of course, but with the DNG file, it is simplest and fastest to pass the parameters along and let the work be done outside. By contrast, the vignetting and color shift corrections are best applied by LUTs that map the raw pixels into corrected values. That's how it was done in the M8 (the last case in which you could actually read the firmware code). scott All true, but it doesn't change the fact that looking for DNG opcodes doesn't necessarily give you the information you're looking for. To see what the camera actually does, the bottom line is to do some testing with lens profiles off and on, and see whether the raw data itself is any different. Quote Link to post Share on other sites More sharing options...
scott kirkpatrick Posted August 21, 2016 Share #30 Posted August 21, 2016 All true, but it doesn't change the fact that looking for DNG opcodes doesn't necessarily give you the information you're looking for. To see what the camera actually does, the bottom line is to do some testing with lens profiles off and on, and see whether the raw data itself is any different. Extracting the opcodes from some of the M and SL raw file output does lead to some interesting differences. I had thought that the lens profiles used for M lenses by the SL would be derived, or even just copied, from the lens profiles in the M series. But, using the Super-Elmar 18-M as an example, I find that the SL generates DNG files with opcodes for correcting distortion and LCA. The M240 does not add these opcodes, and may be doing some distortion correction before exporting the DNG file. With R lenses, the SL exports opcodes at their default no-distortion values in the few cases that I have checked. Of course with SL lenses, the opcodes are necessary and are used. scott 2 Quote Link to post Share on other sites More sharing options...
scott kirkpatrick Posted August 21, 2016 Share #31 Posted August 21, 2016 (edited) More on DNG from the M240 and the SL. The M240 outputs DNG1.3 format, with back-level compatibility to DNG1.1, thus doesn't include the WarpRectilinear transform. The SL generates DNG1.4 format, which is only required to be back-level compatible to 1.3 format. That's why there are no distortion corrections in the observable, standardized parts of the M's raw files. There is a lot of other stuff, however, and there may be proprietary MakerNotes dealing with distortion that I don't see with the Adobe SDK. In the M, I can turn off lens profile and compare, but this will not differentiate what is done in the camera and what, if anything, is passed along secretly in the DNG, since both are turned off. In the SL, every 6-bit coded lens is profiled. scott Edited August 21, 2016 by scott kirkpatrick 2 Quote Link to post Share on other sites More sharing options...
MRJohn Posted August 23, 2016 Share #32 Posted August 23, 2016 Is anybody aware of an alternative source to retro-fit a ROM onto an R lens? I have asked Leica in Germany and they said they ran out of parts and are no longer able to do it. - It might seem a bit off topic but is related to the functionality of this adapter. Quote Link to post Share on other sites More sharing options...
ramarren Posted August 23, 2016 Share #33 Posted August 23, 2016 If anyone else has the parts, it would be Don Goldberg. But my understanding is that there are no more parts available. Couldn't hurt to ask. Quote Link to post Share on other sites More sharing options...
MRJohn Posted August 23, 2016 Share #34 Posted August 23, 2016 If anyone else has the parts, it would be Don Goldberg. But my understanding is that there are no more parts available. Couldn't hurt to ask. Thanks, is this http://www.dagcamera.com? Nothing is listed there but I can shoot an email. Quote Link to post Share on other sites More sharing options...
ramarren Posted August 23, 2016 Share #35 Posted August 23, 2016 Thanks, is this http://www.dagcamera.com? Nothing is listed there but I can shoot an email. Yes, that's his website. His email is on the contact page, he's usually prompt about responding to inquiries. 1 Quote Link to post Share on other sites More sharing options...
Leicaiste Posted August 24, 2016 Share #36 Posted August 24, 2016 Is anybody aware of an alternative source to retro-fit a ROM onto an R lens? I have asked Leica in Germany and they said they ran out of parts and are no longer able to do it. - It might seem a bit off topic but is related to the functionality of this adapter. I sent my R-19/2,8 V2 to Leica Wetzlar earlier this year in and it came back ROMed. Quote Link to post Share on other sites More sharing options...
MRJohn Posted August 24, 2016 Share #37 Posted August 24, 2016 I sent my R-19/2,8 V2 to Leica Wetzlar earlier this year in and it came back ROMed. Thanks, they said they ran out of parts this year. Either it was afterwards or they still have a special VIP stash which they sometimes go to. Quote Link to post Share on other sites More sharing options...
ramarren Posted August 24, 2016 Share #38 Posted August 24, 2016 Thanks, they said they ran out of parts this year. Either it was afterwards or they still have a special VIP stash which they sometimes go to. Well, remember that each lens has its specific ROM ... They could have some available but not all. Quote Link to post Share on other sites More sharing options...
MRJohn Posted August 24, 2016 Share #39 Posted August 24, 2016 Well, remember that each lens has its specific ROM ... They could have some available but not all. maybe, but I had not even specified a particular lens when I called. Perhaps I shoot them another email. Quote Link to post Share on other sites More sharing options...
jpattison Posted September 2, 2016 Share #40 Posted September 2, 2016 (edited) Nearly all mount adapters are designed to be a few tenths of a mm shorter than the nominal registration baseline to allow all lenses to achieve true infinity when focused TTL on the digital sensor. There are several reasons for this: Precise mount register distance was always kind of a farce anyway with film camera lenses because the image plane afforded by the surface of the film on the pressure plate has built into it a couple of tenths of a mm variability due to the different thicknesses of film base and emulsion base as you can see easily if you use a micrometer and measure the thickness of various different films. There's also the fact that a piece of film in the film gate on has a certain amount of curl to it from center to edge, depending on the specific design and implementation of the film rails and pressure plate as well as the thickness of a particular film. Digital sensors are far flatter than any film ever was and mount register can be specified to an absolute value. But camera lenses have their own level of variability to them due to mounting flange wear and focusing calibration, etc. So to assure that adapted lenses can achieve a true infinity focus on the digital sensor, mount adapters allow an additional bit of leeway. The downside is that this affects the infinity index marking and restricts the accuracy of the DoF scale, and the amount of error is most noticeable with short focal length lenses. I've not found these issues to be much of a hindrance other than for zone focusing, but even then I typically use deep DoF settings for zone focusing so it's only rarely been noticeable as an issue. I did a little test... Leica SL lens test ISO 50 Subject Focus = 3.2m Linhof Tripod M adapter T 30 Aug 2016 Lens Focus shown on lens Best Aperture 11135 21mm f2.8 ASPH (6bit) 1.9m f4 11891 50mm f1.4 ASPH (6bit) 3.2m f2.8 11874 35mm f1.4 ASPH 3.2m f2 (colour shift) 11118 50mm f2 (1965) 3.2m f4 M adapter T with R-Adapter-M 30 Aug 2016 21-35mm f3.5-4 ASPH (ROM) 21mm 1.8m f5.6 21-35mm f3.5-4 ASPH (ROM) 24mm 1.9m f5.6 21-35mm f3.5-4 ASPH (ROM) 28mm 1.9m f5.6 21-35mm f3.5-4 ASPH (ROM) 35mm 2.3m f5.6 28mm f2.8 PC Super Angulon 2.5m f5.6 (colour shift) 11216 50mm f2 R 2.9m f4 60mm f2.8 Macro R 3.0m f5.6 90mm f2 APO ASPH 3.1m f4 R-Adapter-L (focus distance test only) 1 Sept 2016 21-35mm f3.5-4 ASPH (ROM) 21mm 1.4m 21-35mm f3.5-4 ASPH (ROM) 24mm 1.6m 21-35mm f3.5-4 ASPH (ROM) 28mm 1.8m 21-35mm f3.5-4 ASPH (ROM) 35mm 2.1m 28mm f2.8 PC Super Angulon 2.5m 11216 50mm f2 R 2.7m 60mm f2.8 Macro R 3.0m 90mm f2 APO ASPH 3.2m Strange how far off the M 21ASPH is, it shows 3.2 on an M9. Happy focusing ! John Edited September 2, 2016 by jpattison 2 Quote Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.