DavidStone Posted May 31, 2012 Share #1 Posted May 31, 2012 Advertisement (gone after registration) Because I occasionally travel with just hand-baggage, and the M8 body can be a fairly substantial object under these circumstances, I recently acquired the small and lightweight NEX-5N and an E to M adapter, which makes an altogether smaller and lighter outfit. I got the Voigtlander adapter, which seems generally considered to be of good quality. Carefully focused using the built-in 'focus peaking' facility, the quality of the results is excellent, if perhaps a bit short of the clarity that the unfiltered Leica sensor produces. But I gradually realised that what was in focus on the screen, and on the sensor, was nowhere near what the focusing scale on the lens was indicating. I've tried this on several lenses now, and the results are consistent. When the lens is set on infinity, the focus is in fact beyond infinity, so for true infinity focus the lens has to be set for a slightly closer distance. And this holds true right through the scale. The subject distance is consistently greater than that shown on the scale. So the adapter must be slightly thinner than it ought to be. As long as I'm focusing using the screen, everything works as it should and the results are what I'd expect. But I do sometimes pre-focus using the scale on the lens, and with this setup I can't do this. It's a pain having to send something back and ask for a replacement, and then carefully check that for errors, so first i'd just like to know if anyone here has had a similar experience of an inaccurate adapter? David Link to post Share on other sites More sharing options...
Advertisement Posted May 31, 2012 Posted May 31, 2012 Hi DavidStone, Take a look here Problem with E-M adapter. I'm sure you'll find what you were looking for!
gyoung Posted May 31, 2012 Share #2 Posted May 31, 2012 Because I occasionally travel with just hand-baggage, and the M8 body can be a fairly substantial object under these circumstances, I recently acquired the small and lightweight NEX-5N and an E to M adapter, which makes an altogether smaller and lighter outfit. I got the Voigtlander adapter, which seems generally considered to be of good quality. Carefully focused using the built-in 'focus peaking' facility, the quality of the results is excellent, if perhaps a bit short of the clarity that the unfiltered Leica sensor produces. But I gradually realised that what was in focus on the screen, and on the sensor, was nowhere near what the focusing scale on the lens was indicating. I've tried this on several lenses now, and the results are consistent. When the lens is set on infinity, the focus is in fact beyond infinity, so for true infinity focus the lens has to be set for a slightly closer distance. And this holds true right through the scale. The subject distance is consistently greater than that shown on the scale. So the adapter must be slightly thinner than it ought to be. As long as I'm focusing using the screen, everything works as it should and the results are what I'd expect. But I do sometimes pre-focus using the scale on the lens, and with this setup I can't do this. It's a pain having to send something back and ask for a replacement, and then carefully check that for errors, so first i'd just like to know if anyone here has had a similar experience of an inaccurate adapter? David Yes, one of my m4/3 adaptors (from Hong Kong) is like that, the other SRB one is fine, although the first example didn't reach infinity, so was replaced. Easy to live with a too short adaptor if its cheap, too long is useless! The Voigtlander however will not have been cheap so should (IMHO) be right. Gerry Link to post Share on other sites More sharing options...
Blindjock Posted May 31, 2012 Share #3 Posted May 31, 2012 Depends how the adapter is made but if the mounting surface is removable you can shim out the adapter until it is exactly the right size. Tin foil works quite well and is circa 0.03mm thick depending on brand. It might need a few layers though. Link to post Share on other sites More sharing options...
DavidStone Posted May 31, 2012 Author Share #4 Posted May 31, 2012 Depends how the adapter is made but if the mounting surface is removable you can shim out the adapter until it is exactly the right size. Tin foil works quite well and is circa 0.03mm thick depending on brand. It might need a few layers though. I might have done this with a cheap adapter, but, as Gerry pointed out the Voigtlander is not cheap, so I suppose it had better go back. What a pain. David Link to post Share on other sites More sharing options...
giordano Posted May 31, 2012 Share #5 Posted May 31, 2012 Are you certain that the problem is with the adapter and not with the body? With powered focusing mechanisms in the lenses and focus sensing on the sensor itself, there's no need to maintain the tight tolerance on the flange-to-sensor distance that SLR and RF cameras require. Link to post Share on other sites More sharing options...
DavidStone Posted May 31, 2012 Author Share #6 Posted May 31, 2012 Are you certain that the problem is with the adapter and not with the body? With powered focusing mechanisms in the lenses and focus sensing on the sensor itself, there's no need to maintain the tight tolerance on the flange-to-sensor distance that SLR and RF cameras require. Good point. Almost all my knowledge and experience comes from rangefinder cameras, so I'm accustomed to absolute accuracy in a camera body. This could indeed be the common factor, and someone who used just the standard kit lens would never know. Maybe I should stop worrying and accept that a NEX is not an M-series Leica. It is at least useable, and the results are more than OK. David Link to post Share on other sites More sharing options...
lars_bergquist Posted June 1, 2012 Share #7 Posted June 1, 2012 Advertisement (gone after registration) Remember that focus peaking does not work from a rangefinder separate from the lens. It works by automatically analysing the actual image projected onto the sensor by the lens – just as the matte screen of a SLR camera lets you evaluate – per eyeball – the actual image in the focal plane. If that image is in focus, then it is in focus ... by definition, in the case of a camera focusing off the sensor. Now if your SLR finder tells you that the image on the screen is correctly focused, but the picture actually taken is not, then the inevitable conclusion is that the film/sensor plane and the matte screen are at different focus distances. In other words, in your case there's nothing wrong with the lens, including the adapter, but the fault is with the camera. Whatever the actual extension of the lens, however it is focused, the camera is analysing the optical image that is falling on the sensor, and will be captured by it. And this analysis is wrong. Talk to Sony, not to Cosina. The old man from the Kodachrome Age Link to post Share on other sites More sharing options...
ericborgstrom Posted June 1, 2012 Share #8 Posted June 1, 2012 I use the Nex-5N with the Voigtländer VM adapter and the excellent Sony EVF. With this outfit and using focus peeking I can reliably focus the APO-Telyt-M 3,4/135 something that is an iffy undertaking for me with the M9. I hope the Leica M10 will intelligently integrate this focusing method as an option. Btw I find it less difficult to focus the Elmar 4/135 then the APO-Telyt on the M9, perhaps due of the less steep focusing helicoid on the Elmar. Eric Link to post Share on other sites More sharing options...
DavidStone Posted June 1, 2012 Author Share #9 Posted June 1, 2012 Remember that focus peaking does not work from a rangefinder separate from the lens. It works by automatically analysing the actual image projected onto the sensor by the lens – just as the matte screen of a SLR camera lets you evaluate – per eyeball – the actual image in the focal plane. If that image is in focus, then it is in focus ... by definition, in the case of a camera focusing off the sensor. Now if your SLR finder tells you that the image on the screen is correctly focused, but the picture actually taken is not, then the inevitable conclusion is that the film/sensor plane and the matte screen are at different focus distances. In other words, in your case there's nothing wrong with the lens, including the adapter, but the fault is with the camera. Whatever the actual extension of the lens, however it is focused, the camera is analysing the optical image that is falling on the sensor, and will be captured by it. And this analysis is wrong. Talk to Sony, not to Cosina. The old man from the Kodachrome Age Crossed line here Lars. I don't have a focus problem, and the only SLR camera I posess is a Pentax S1a bought new in 1965 with the proceeds of my first professional job. I made two assumptions - that focus peaking would give me a sharp result, since it shows what the sensor is seeing - this turned out to be true. And that the NEX-5N flange to sensor dimension had to be accurate. This second assumption, as Giordano pointed out, is, these days, not necessarily true. So I have a slightly shorter flange to sensor dimension than, in a perfect world, I should have, and I have to compensate by "racking out" the lens slightly. Which means that the distance scales on the lens cannot be used. But I can still get sharp photos, using the electronic equivalent of direct focusing, so I think I'll live with it. But I have to agree that the fault is almost certainly with Sony. David Link to post Share on other sites More sharing options...
gyoung Posted June 1, 2012 Share #10 Posted June 1, 2012 Perfectly possible the camera is out, to no effect with its own af lenses, but if so its not to spec, such things are usually specified tightly, first suspect Imho is the adapter, if that is right then the nex body needs to go back to be brought to spec Gerry Link to post Share on other sites More sharing options...
Blindjock Posted June 1, 2012 Share #11 Posted June 1, 2012 The camera is possibly in spec. It is just that the design tolerance does not need to be that accurate as the AF lens can compensate. Makes the manufacturing cheaper for Sony. I'd tweak the adapter to suit the camera or live with the issue. Link to post Share on other sites More sharing options...
DavidStone Posted June 1, 2012 Author Share #12 Posted June 1, 2012 The camera is possibly in spec. It is just that the design tolerance does not need to be that accurate as the AF lens can compensate. Makes the manufacturing cheaper for Sony. I'd tweak the adapter to suit the camera or live with the issue. I think I'll do exactly that. Live with it for the time being, and shim the adapter if I feel it's worth the trouble. David Link to post Share on other sites More sharing options...
gyoung Posted June 2, 2012 Share #13 Posted June 2, 2012 A lot of the fun of using manual focus lenses, and their usefulness for me, is using the dof scales to ensure that what I want 'sharp' actually is, at the very least for that you need accurate focus scales, poor registration ruins that. Actual dof can be worked out externally if the lens scales don't suit the format, I have an android app for it. Gerry Link to post Share on other sites More sharing options...
Manolo Laguillo Posted June 3, 2012 Share #14 Posted June 3, 2012 How big is the disagreement between the real shooting distance and the one marked on the lens barrel, David? It must be big enough in order to you being able to detect the difference... I would make a test with another Nex 5N body: if the problem is still there, the ring is wrong. If it disappears, it's the camera that is wrong. Does the lens fit snuggly on the adapter, or does it wobble? My first Leica M to Nex adapter was wrong because too thick, therefore infinity focussing was impossible. The fitting was not perfect, the lens "danced" a little. Link to post Share on other sites More sharing options...
IkarusJohn Posted June 3, 2012 Share #15 Posted June 3, 2012 I have the Novoflex LEM, and haven't noticed a problem. I will try to check for you today. Link to post Share on other sites More sharing options...
Joe S Posted June 3, 2012 Share #16 Posted June 3, 2012 I have the Novoflex and it is right on with my lenses distance scale on my NEX 7. It measures .383" thick if you have a means to compare. I originally bought a cheap $15 ebay adapter and it was as you describe, significantly undersize. They advertise that they will focus to infinity so the cheap ones always seem to be undersize as a way of insuring that. Many have confirmed this. The Voigtlander should do better without reworking for the price. From my experience I would be skeptical that the Sony camera is at fault. As a side note, I used the cheap ebay version to fabricate a "visoflex eliminator" to allow the use of my 280mm Telyt and bellows setup on the Sony without a visoflex. Link to post Share on other sites More sharing options...
IkarusJohn Posted June 4, 2012 Share #17 Posted June 4, 2012 Checked my Novoflex with my NEX-5n and Summicron 28. It's spot on at infinity. Link to post Share on other sites More sharing options...
DavidStone Posted June 4, 2012 Author Share #18 Posted June 4, 2012 How big is the disagreement between the real shooting distance and the one marked on the lens barrel, David?It must be big enough in order to you being able to detect the difference... I would make a test with another Nex 5N body: if the problem is still there, the ring is wrong. If it disappears, it's the camera that is wrong. The adapter seems well made and solid, but I have no means of checking its accuracy. The same applies to the body. Substituting other bodies and/or adapters would tell me more, but, short of buying these, I have no access to either, so this is not an option. I have used SLR and TLR and large-format cameras by focusing directly on a screen, so this is not a problem. What is lacking is the ability to pre-focus by scale, but I'm prepared to live with this for now. And sending either the camera body or adapter back for replacement without clear evidence of a fault would take me back to square one. David Link to post Share on other sites More sharing options...
eon Posted June 10, 2012 Share #19 Posted June 10, 2012 My Elmarit 28/2,8 ASPH focuses well beyond infinity with my Novoflex adapter on my NEX-5N. Link to post Share on other sites More sharing options...
jank Posted June 10, 2012 Share #20 Posted June 10, 2012 My Elmarit 28/2,8 ASPH focuses well beyond infinity with my Novoflex adapter on my NEX-5N. I think that Sony has loose tolerances on flange distance to sensor. They could do it , since focusing is derived from the image projected on sensor.Their lenses are surely made such that they can focus paas infinity in all cases ( stacked tolerances). Adapter mfg. then make adaptors thickness such (a bit short) that in all ( almost) cases the lens can focus pass infinity. One has to match the adaptor to the camera body by shims.Then all lenses would have matching scales provided they were perfect to begin with. Jan Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.