Jump to content

Are M lenses really interchangable?


Recommended Posts

Advertisement (gone after registration)

Alan--

Your latest post sneaked in while I was working on my previous one, so I didn't address it above.

 

I'm not an engineer either, but I'm a perfect dilettante and know little enough about a lot of things to spout off about them. (And I'm glad to be corrected because that way I know just a little more about them, and see better how little I know.) And I worked for Leica long enough to know their thinking and commitment.

 

The current problems are definitely solvable, else we wouldn't have a digital M. Remember, a few years back Leica said that wasn't possible, but now we have it. They're still discovering things they didn't expect, but these are the minor ones.

 

Consider the focusing matter. Let me postulate that some bodies are at the "plus" end of tolerances, and that some lenses are also at the "plus" end of tolerances. On a film body, the combination would produce usable results. Because of the difference in the digital sensor, an M8 at the "plus" end coupled with a lens at its "plus" end won't. One of the reasons (sorry, Ed, I'm on my HIGH HORSE again ;) ) is that we can now inspect those images in a way we couldn't before. I think a major breakthrough occurred when Tim Ashley got Leica to recognize the problem with his 35 Summicron.

 

With the M8, Leica has tightened camera tolerances. I think they need to do the same with lenses. I may be wrong, but aren't most of the focus problems coming with the higher speed and newer design lenses? These are the ones in which all the stops have been pulled out. Flaws in QC can be seen more quickly in designs of this type because each surface has to do more, and because we can see the results immediately at 100% and larger.

 

Putting that another way, the same lens quality standards that worked for film won't work for a sensor. Maybe they need to adjust their focus mounts to tighter tolerances. "Vieles wäre zu sagen davon," to quote Hölderlin. On the 1D MkIII, Canon asks you to do that yourself, but the whole gist of _this_ thread is that the manufacturer should be doing it for us, particularly considering what we pay for his lenses (L series or Leica either one).

 

You're right that Leica may need to change some very basic things. The Leica rangefinder dates back what, 70 years? It's what we like about the camera, but it's also getting to be very old technology. A number of people have suggested that a new design is needed. I'm glad they are making the progress they are: They solved a lot of problems to present the M8, but they kept one part that they understand intimately, the rangefinder. Without it, we probably wouldn't be seeing the sales success we are with the camera. But look at Canon, the powerhouse of digital innovation. They are trying a new 'cross-type' AF sensor with the 1D MkIII, and apparently some of their cameras are showing gross errors because of it. (See Canon 1D MKIII Autofocus- Interactive Review - Pro Photo HOME.)

 

I think Steven K Lee may be about to bring Leica into the 21st century. I think the R10 will knock our socks off.

 

The M8 has knocked the socks off a lot of us as well, but this focus issue is a real problem for some people. I would be mad as the dickens if I had the problems that some people have had--Tim's two 35 Summilux ASPH's for example, with an explanation from Solms that indicated they didn't understand; followed by a 35 Summicron ASPH with exactly the same problem. But by then, Tim was either exasperated enough or burned out enough to be able finally to get through to Leica what they hadn't yet understood.

 

Sorry, I'm repeating myself. There's a problem. Leica is aware of it. It's a simple problem to handle with changes of procedure. That's my take on it. Sorry if I seem dispassionate on the matter, but I think it's on the way to being resolved, and I think this thread shows that.

 

Slight change of topic: Someone said before something to the effect that it was only Leica's lenses that show the focus problem. First, that's reasonable, given the differences in Zeiss's and Leica's designs. But second, there's another active thread in which a new user of the C-Sonnar 50/1.5 says he notes some focus deviation with it as well. It's a fact of the digital world that we can see variations today that we couldn't see with film. Part of the adjustment will come from the manufacturers, but perhaps part will need to come from us users as well.

 

Hello. My name is Howard. I was in my late 50's when I peeped my first pixel, but it's such an insidious, seductive thing ... :rolleyes:

--HC

Link to post
Share on other sites

  • Replies 119
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

My experience is very similar to Jaap's

I have two bodies, and I've used a large number of lenses, all of which I've tested fairly thoroughly (and I do care - around 60% of Nikkors I've bought have been sent back with problems).

 

Only one lens back focused badly (a 90mm f4 macro - bought new) it went back to Solms for 6 weeks, and came back slightly better but still not good enough - it's now been back in Solms for a further 6 weeks.

 

Still, my other lenses (both new and old) have been really good, so I can live with it. I do understand your frustration (like mine with my early D2x), but I don't believe that it represents a majority experience

Link to post
Share on other sites

To me it seems we're dealing with the 'half full, half empty' problem, or the 'invisible gardener' problem.

 

Some people clearly have major focusing problems with the camera, most (so far as I can tell) don't.

 

I think we've gone through the list of possible explanations, and some of us just like to complain. There's nothing wrong with that. We aren't accomplishing anything by complaining, but we certainly feel better afterward, and we have the hope that one more voice will give Leica greater impetus to solve the problem.

 

Here's something you could help me with if you're interested. I'm curious about the 1/2 second consistency in M8 skipped files, as I mentioned at http://www.l-camera-forum.com/leica-forum/leica-m8-forum/29625-m8-skips-file-numbers-anyone-else.html#post310499. With the interest and enthusiasm expressed in the present thread, I bet some of you would have some insight into why and under what circumstances the M8 skips file names. Several reasonable proposals have been raised in the thread, but I bet there are more that haven't occurred to us.

 

It seems to me an interesting programming problem if nothing else, and putting our combined knowledge and experience to work on the issue could produce some information that could help correct a minor snag in the M8's programming.

 

--HC

Howard I started this thread with a purpose. My theory was that to achieve critical sharpness..and this may be in the eye of the beholder...both the camera and the lenses may need adjustment. Leica has told me this several times. I was most interested in hearing from individuals that have actually tested their M8 s and lenses for critical shrapness. Ed and I spent all morning testing our lenses and cameras . We had a decent sample size 4 M8 bodies and about 15 lenses between us . Each of individually focused on a test target at about 55inches. We were able to agree on every result within 5mm..I might call it 10mm back ..Ed might say closer to 5mm. Since we repeated each test several times , I believe our tests were sufficient to determine if a lens was properly calibrated . For example we took my older 50 summicron and ran tests on 4 different bodies. We then took Ed s new 50 summicron and ran the same test. We followed this test with a target at 10M..obtaining the same results. I beleive this is the type of testing that is necessary to draw conclusions. Simply stating that only a few people have had problems isn t helpful and I am asking you to stop restating your opinion . If you have tested your lens and found them to be all be 100% ..good for you . Say it once and move on.
Link to post
Share on other sites

Alan--

But second, there's another active thread in which a new user of the C-Sonnar 50/1.5 says he notes some focus deviation with it as well.

--HC

 

That is a different issue. Some lens designs change focus with aperture. Zeiss has said this is the case with the 50/1.5 and that they set focus at 2.8 so at f/1.5 it front focuses and at f/5.6 it backfocuses. Some users have described a similar problem with the 35 summilux ASPH. I have no idea why some users don't have this problem with the 35/1.4 and some do. This is not normally something that can be fixed unless you have an RF cam that can change according to aperture as it is a shortcoming of the lens design not a QC error.

 

A lens that is backfocusing because its out of calibration can be fixed with a shim to get the register distance correct to a standard. That is an inexpensive fix with a qualified service tech. Calibrating the RF on a camera is also not a big deal for a qualified technician. You would expect a new camera out of the box to have a properly calibrated RF and same for a new lens. Unfortunately this is not always the case.

 

Digital devices generally work or they don't. When they work they are very precise. Mechanical devices can be anywhere along a curve from way out of spec to dead on and still function. They require adjustment regularly and generally degrade gracefully with use and abuse. Now that the M is dependent on digital I wouldn't mind seeing some of the more fussy and imprecise mechanical bits converted to digital. The RF finder is the pinnacle of 1950's opto-mechanical technology but you would think in 50+ years there would be some room for improvement.

Link to post
Share on other sites

If you have tested your lens and found them to be all be 100% ..good for you . Say it once and move on.

Roger--

I agree completely. If you'll notice, my repetitions have simply been in response to people who are themselves repeating arguments that have already been answered.

 

Still open is the question I raised above about early adopters / late adopters. Ed's answer was off the topic, answering with serial numbers, and you didn't respond at all. That makes it look as if you're not serious about the question you raised in the thread title.

 

You say you and Ed looked at four bodies. What about Carsten's eight (wasn't it) mentioned above?

 

As I said, I'm glad you opened the thread. But anecdotes aren't evidence. Quantify.

 

Your complaint that I'm repeating myself doesn't cite any repetitions and shows no ellipses. That's sloppy editing.

 

Nor, as I read it, does your latest post say anything about the results of your and Ed's test. You say you had a certain number of bodies and lenses and agreed about results within 5 mm, but you skipped reporting on your findings. You say you opened the thread with a purpose, and here you are apparently ignoring the purpose. :cool:

 

--HC

Link to post
Share on other sites

If I was starting out, I would buy just one body and two lenses - 28/2, 75/2 and would then have a reasonable chance of a consistent set. As it is, I'm overwhelmed by the level of variability I'm seeing with 3 bodies and 16 lenses. With the M8, less is more.

 

Well I've been more than satisfied with the performance of my M8 and my nine lenses, including golden oldies like the Canon 19mm f3.5 and 50mm f1.2

But now you have planted the seeds of doubt.

 

Well get out the tripod and ruler....full speed ahead!

 

Rex

backsliding pixelholic

Link to post
Share on other sites

Advertisement (gone after registration)

That is a different issue. Some lens designs change focus with aperture. Zeiss has said this is the case with the 50/1.5 and that they set focus at 2.8 so at f/1.5 it front focuses and at f/5.6 it backfocuses.

Agreed. And one forum member even asked Zeiss to adjust his sample so it was in agreement with the rangefinder wide open. I was merely responding to the other post I mentioned. We've gone over the matter with regard to Leica lenses in other threads as you know, and LFI is to discuss it in the next issue.

Some users have described a similar problem with the 35 summilux ASPH. I have no idea why some users don't have this problem with the 35/1.4 and some do. This is not normally something that can be fixed unless you have an RF cam that can change according to aperture as it is a shortcoming of the lens design not a QC error.

I don't understand the logic. Tim's problem with the 35/2 was minor backfocus combined with the lens's intended focus shift. I think that's likely the same as the problem he had with the 35/1.4's. Tim's case was fixed simply by correcting the backfocus. Seems to me that a similar modification to the Summiluxes might do the same. As you say, some samples show the problem and some don't. As I see it, that kind of variation isn't likely to result from a design flaw.

A lens that is backfocusing because its out of calibration can be fixed with a shim to get the register distance correct to a standard. That is an inexpensive fix with a qualified service tech. Calibrating the RF on a camera is also not a big deal for a qualified technician. You would expect a new camera out of the box to have a properly calibrated RF and same for a new lens. Unfortunately this is not always the case.

No argument. It shouldn't be happening today and it shouldn't have happened in the past.

Digital devices generally work or they don't. When they work they are very precise. Mechanical devices can be anywhere along a curve from way out of spec to dead on and still function. They require adjustment regularly and generally degrade gracefully with use and abuse. Now that the M is dependent on digital I wouldn't mind seeing some of the more fussy and imprecise mechanical bits converted to digital. The RF finder is the pinnacle of 1950's opto-mechanical technology but you would think in 50+ years there would be some room for improvement.

We're in agreement again. You're echoing my remarks of an hour ago on the matter.

 

Hank, we're in agreement on every point except that I don't understand your reasoning that if some Summiluxes show a problem and others don't, it's a matter of lens design and not of quality control.

 

--HC

Link to post
Share on other sites

............ We had a decent sample size 4 M8 bodies and about 15 lenses between us . Each of individually focused on a test target at about 55inches. We were able to agree on every result within 5mm..I might call it 10mm back ..Ed might say closer to 5mm. Since we repeated each test several times , I believe our tests were sufficient to determine if a lens was properly calibrated .........

 

Good God man, 5 or 10mm backfocus at 55" is your tolerance level! For us Americans thats less than 1/4" to 1/2" ! Forgive the explanation points, but I can't begin to hold a camera that steady to begin with. Now if you were talking 3 or 4 INCHES, maybe you have a point. I have pretty good vision but I still find the standard deviation for repeated shots to be a lot more than 5mm. I hope this threads rant is about more than 5mm backfocus at 55". Please.

 

Rex

who was about to set up his tripod.

Link to post
Share on other sites

Roger--

I agree completely. If you'll notice, my repetitions have simply been in response to people who are themselves repeating arguments that have already been answered.

 

Still open is the question I raised above about early adopters / late adopters. Ed's answer was off the question, answering with serial numbers, and you didn't respond.

 

You say you and Ed looked at four bodies. What about Carsten's eight (wasn't it) mentioned above?

 

As I said, I'm glad you opened the thread. But anecdotes aren't evidence.

 

--HC

 

Well my evidence is not anecdotal. I have personally tested 4 new out of the box M8s between February and July each with 5 lenses. All 4 M8s clearly back focused.

Link to post
Share on other sites

.

 

Hank, we're in agreement on every point except that I don't understand your reasoning that if some Summiluxes show a problem and others don't, it's a matter of lens design and not of quality control.

 

--HC

 

My point was that if the 35/1.4 shifts focus with aperture as does the C-sonnar it should happen with all samples of the lens. So I am not sure what sort of issue the Summilux has. Simple mis-calibration or some issue revolving around the lens design or a bit of both.

Link to post
Share on other sites

Good God man, 5 or 10mm backfocus at 55" is your tolerance level! For us Americans thats less than 1/4" to 1/2" !

 

I don't believe he was refering to a 5-10mm "tolerance" level. Just that each person could standardize his observation within 5mm of the other. For all we know one lens could have only been off by 5-10mm. Which you may conclude is negligible. But another lens could have been off by about 100mm but one thought it was 100mm and the other thought it was 105mm.

 

We'd have to see the test result measurements before we could decide what is an acceptable tolerance for ourselves.

Link to post
Share on other sites

If I was starting out, I would buy just one body and two lenses - 28/2, 75/2 and would then have a reasonable chance of a consistent set. As it is, I'm overwhelmed by the level of variability I'm seeing with 3 bodies and 16 lenses...

Mark--

No doubt. But couldn't this also be the case if you had conducted such tests on M2's, M6's etc?

 

Effects that were already present are becoming visible. I believe you originally felt your 35 Summilux didn't show any of the defects of Tim Ashley's samples, but then after testing decided it did, to a much lesser degree.

 

 

"And a thing is not seen because it is visible, but conversely, visible because it is seen"—motto of Diane Arbus, after Plato, Euthyphro, part 04

 

--HC

Link to post
Share on other sites

Mark--

No doubt. But couldn't this also be the case if you had conducted such tests on M2's, M6's etc?

 

Effects that were already present are becoming visible. I believe you originally felt your 35 Summilux didn't show any of the defects of Tim Ashley's samples, but then after testing decided it did, to a much lesser degree.

 

 

"And a thing is not seen because it is visible, but conversely, visible because it is seen"—motto of Diane Arbus, after Plato, Euthyphro, part 04

 

--HC

 

:rolleyes: I give up!

Link to post
Share on other sites

I don't believe he was refering to a 5-10mm "tolerance" level. Just that each person could standardize his observation within 5mm of the other. For all we know one lens could have only been off by 5-10mm. Which you may conclude is negligible. But another lens could have been off by about 100mm but one thought it was 100mm and the other thought it was 105mm.

 

We'd have to see the test result measurements before we could decide what is an acceptable tolerance for outselves.

 

I'm not so sure Roger was referring to 5mm of *standard deviation* for better words. I think he may be referring to total backfocus of 1/4 to 1/2" of backfocus being unacceptable.

What's missing from this how discussion is any quantifing the AMOUNT of backfocus that is deemed intolerable by the individual tester. What I discovered in my earlier testing of my RD1 was that, even in critical applications such as my Canon 50mm F1.2 LTM, a small backfocus or focus shift of a 1/2" is of no practical significence because camera/subject motion is the real determining factor in what becomes the most sharply defined subject plane. I will grant you that the DOF with a really that lens can be less than 1/4" but unless you have the camera on a tripod and the subject motionless beyond anything realestic, luck is the determining factor in getting that left eye in focus and not the tip of the nose. Backfocus has little to do with it.

 

Rex

Link to post
Share on other sites

:rolleyes: I give up!

 

+1

 

Howard just won't quit. His bodies are perfect, his lenses are perfect. So from his point of view all M8's and new Leica lenses are perfect and we, the ones that are seeing problems, are just NUTS.

 

You do know Howard you could just GO AWAY from this thread and let the people that do have a problem work this out on our own.

Link to post
Share on other sites

My point was that if the 35/1.4 shifts focus with aperture as does the C-sonnar it should happen with all samples of the lens. So I am not sure what sort of issue the Summilux has. Simple mis-calibration or some issue revolving around the lens design or a bit of both.

Thanks, Hank, for the clarification. We agree on this as well.

 

Tim's initial communication from Solms said that the 35 Summilux ASPH lens was designed with a great deal of focus shift, the 35 Summicron ASPH with less, and the 28 Elmarit ASPH with none.

 

The problems he was having seemed extremely peculiar--edges in focus but center very far out, three meters or so. At first, none of us reading the thread could figure out what the problem was. But later when Tim's 35/2 had a similar problem, Leica found it to be due to slight backfocus along with the intentional focus shift.

 

We don't know whether the same solution would have repaired the Summiluxes, but if so, the problem is minor, as you said--simply shim to correct the backfocus. The focus shift itself, of course, is part of the reason the lens performs so well.

 

--HC

Link to post
Share on other sites

Howard just won't quit. His bodies are perfect, his lenses are perfect. So from his point of view all M8's and new Leica lenses are perfect and we, the ones that are seeing problems, are just NUTS.

 

You do know Howard you could just GO AWAY from this thread and let the people that do have a problem work this out on our own.

Ed--

You're obviously not a native speaker of English, so you can't be faulted too deeply, but you clearly haven't understood a thing I've said.

 

And so far, all you've done is complain. You haven't advanced a single step toward trying to solve the problem.

 

Sounds as if you and Roger had a hidden agenda in testing your 4 cameras and 15 lenses and then refusing to report your results despite requests. :D

 

But at least I'm glad YOU'VE GOT A CAPS LOCK key. It obviously gives you pleasure. :p

 

--HC

Link to post
Share on other sites

You do know Howard you could just GO AWAY from this thread and let the people that do have a problem work this out on our own.

 

With a thread title of 'Are M lenses really interchangeable' I would have thought _all_ experiences would be valuable. Evidently not.

 

For what it's worth I have used 24/35/50 (Summicron and Noctilux)/75 Summicron lenses. The only lens I have had an issue with is the 75mm Summicron, and since the images don't overlap perfectly at infinity I believe it's a lens focussing cam issue and the lens is being checked by Leica.

Link to post
Share on other sites

Well my evidence is not anecdotal. I have personally tested 4 new out of the box M8s between February and July each with 5 lenses. All 4 M8s clearly back focused.

Graham--

Maybe British and American usage differ. :(

 

How much was the backfocus?

 

Were the bodies consistent among themselves? Were the lenses consistent across bodies?

 

--HC

Link to post
Share on other sites

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...