View Full Version : Problem with M9 Sensor Placement
As much as I have enjoyed my time with the M9, which has been short so far, one major area of frustration is focussing. Not one of my 5 lenses is truly right on. They either focus too close or too far. I sent my 35 ASPH to DAG as it was way off and it is ok now but not truly on. My 135, 90 Summicron focus in back of the subject and the 50 Summicron a bit in front. The 90 Elmarit-M is about right.
So I called DAG and Don said that so man Leica lenses really aren't dialed in that well (paraphrase) and he has had to adjust many for the 8 and 9. So it may not be a problem of the 9 but it wasn't a problem with my M2 or M3. The exacting nature of the sensor seems to pose a bit more trouble. I guess I will have to send everything in but I never had to do this with any other camera though I know those who have had to with some of the Canons.
Pickett Wilson
04-03-2010, 03:45
How do the prints look from these cameras versus pixel peeping at 100 percent? When comparing results from an M2 and an M9, did you really look at photos from your M2 at sizes as large as 20 inches wide with your nose at the photo? I suspect you would have been unhappy with the results as well.
I guess it's not the camera so much as that people look at pictures differently. The difference with the digital camera is that people start pixel-peeping and looking at 100% crops. Then they are disappointed if they find imprecisions. Had they looked at their M2/M3 negatives with a microscope, they would have found the same imprecisions, except that they didn't do it. It wasn't necessary anyway, because people normally don't make billboard-sized prints, which would be the equivalent of a 100% crop.
That's the reason why you didn't have to do it with any other camera.
I don't know how you determine that your lenses are "on" or "off", but if it's by looking at 100% crops, you'll never have peace of mind; pixel-peeping drives up the total cost of ownership by a fair margin. If you give your camera a little bump and the rangefinder goes out of alignment even a little bit, those 100% crops of open-aperture shots will scream "Service! Service!", a long time before you would have even noticed the same misalignment with normal enlargements from film negatives.
Pickett Wilson
04-03-2010, 03:52
rxmd, GMTA ;)
I was always frustrated by my negatives in the old day looking at them on the enlarger baseboard with a grain focuser. But the prints looked fine.
The fact that this can and does happen with any camera of this type as well as autofocus SLRs should come as no surprise, right?
Of course you want to get the most out of your M9 and lenses. Focus errors will most certainly be evident in prints, obviously more so with fast, wide open lenses.
That said, I think you will find that the exact point of focus will vary somewhat between your various lenses. You need to decide if they all fall within the acceptable DOF. I had to have one lens adjusted by Don for my M8 - a Hexanon. However I also had to tweak the RF myself on the camera to get good results from all lenses. With everything adjusted the exact focus of my Summicron 90 is at a different point within the DOF than my Hex, or my Elmar, etc.
Brian Sweeney
04-03-2010, 05:02
Most of the lenses that I shimmed myself for my film cameras were fine with the M8. I use a 15x magnifier with them. A couple were off, including the 90/2 Summicron and 10.5cm F2.5 Nikkor. One J-3 was off, but more due to focus-shift.
SO: Bottom line, I do not find a difference in a properly adjusted film camera and the M8. The ability to "pixel-peep" goes way beyond normal DOF tolerances with lenses.
This unccoated Sonnar was shimmed for my Canon P, used wide-open on the M8 without change.
http://ziforums.com/picture.php?albumid=230&pictureid=2388
Nikkor-P 8.5cm F2, straight off of Ebay from reliable dealer, wide-open on the M8.
http://ziforums.com/picture.php?albumid=230&pictureid=2392
newsgrunt
04-03-2010, 05:04
I honestly believe if it was sensor related then most of the shots I've taken with dslrs would suffer similar fate using older mf as well as manually focussed af lenses. But don't. I've had no problems with the M9 I have now and all are older non coded glass. Even the Nokton 1.1 is fine.
Could this be sensor assembly related ?
Brian Sweeney
04-03-2010, 05:08
If the sensor was not properly placed, all lenses would either front focus or back-focus. As the error is "plus and minus", it's the lenses and not the camera.
Brian Sweeney
04-03-2010, 05:13
The Nikkor 10.5cm F2.5 was slightly off, one layer of copper tape of the RF cam took care of it. The perfect focus is between wide-open and F4, chosen for the focus-shift.
http://ziforums.com/picture.php?albumid=219&pictureid=2231
http://ziforums.com/picture.php?albumid=219&pictureid=2230
I feel very lucky in that none of my lenses, which are mostly prior-generation Leica, and Voitlander, has a focus problem with my M8. Is the M9 more critical in that respect?
I would stop agonizing if I were you. Although all posters above are right, film was much les critical than a sensor. For one thing, as it has a thickness, there is a good chance that either the bottom or the top would provide the sharpness. And due to the random character of grain, film is less precise than a sensor too. So it is not abnormal that you will have to have your older lenses adjusted to the new situation. Lenses delivered nowadays are built to much more narrow tolerances than before, so those will be fine unless something went wrong with the calibration. That does not happen frequently, but is does happen.
If you feel you need more exact focussing, just sent the lenses to Don Goldberg and have them adjusted.
As for the Summilux 35, that one has focus shift and will only be correct at one aperture.
Yes, I already spoke with Don. He has noted many problems with lenses being off. I shot a guy the other day and the Summicron 90 I just got focussed 3 feet behind him. Sure it isn't the camera but the lenses and that was what I was trying to say. The sensors bring this out all the more.
I should add that I started shooting professionaly back in the early 1970's using everything from 8X10 down. I do know how to focus and what to look for. All but one lens either back for front focus of the 5 I have used and as mentioned, Don fixed the 35 ASPH and it was very off. I took three shots of the guy near a wall refocussing each time and each time the wall was in focus but not the subject.
Ron (Netherlands)
04-03-2010, 10:30
Never had any problems with my M8.2 nor with the R-D1s (2x summicron 35 version 1 and V, summicron 50 latest, summicron 90/2 version 2, Biogon 25, Orion 28, Canon 135 ltm). Have one Industar LD 61 that is off on all my cameras, but another one is dead on. These Leitz lenses were sharp on all my M camera's apart from the Hexar RF, alsmost all lenses on that camera were off; I sold the camarra, since I didn't want to mingle with the lenses. If one uses lenses on different camera's, it is always the camera that has to be adjusted, not the lenses!
Assuming you want it to focus accurately at a given distance, and given aperture (e.g. wide open), here is a 1 hour test to indicate if the focus errors may be camera, VR/RF, or user focusing error:
1. Set the aperture to the aperture you want to focus most accurately on.
2. Set the camera at measured distances you want to focus accurately at on a tripod.
3. Put out some markers at your target, with a few behind by a couple cm and a couple in front by a couple of cm.
this is the most important part:
4. Don't trust your RF/VF, but do your best to focus and take an exposure, notating it with a mark on the lens.
5. Rotate the focus CCW in 1 degree increments 5 times and take 5 images.
6. Go back to the spot in #4's index focus mark.
7. Rotate the lens in 1 degree increments CW and take 5 shots.
8. Pixel peep the images at 100% at target.
9. If your indexed mark is not your sharpest image, then you need to consider that user focusing error, or your VF/RF, or eye rx, etc. may need to be looked at first. Then get back to your lens.
Pickett Wilson
04-03-2010, 12:50
Three feet behind him is clearly a problem with the lens. Can't blame the M9 for that!
newsgrunt
04-03-2010, 13:21
Just did some qwik tests with the M9 and my 21, 35, 50 and 90, wide open and ALL were sharp. No front or back focus. All worked fine on my M6 and MP so my experience tells me it's a camera tolerance issue. Can't understand how lenses that in all likelihood worked perfectly with film cameras can now be at fault. I wonder what the serial numbers of the problem cameras are and if perhaps there might be a group that slipped through qc.
mervynyan
04-03-2010, 13:26
If I were you, I wouldn't want to spend 7k without almost a perfect camera. If it is still under warranty, I'd send to Leica NJ for adjustment, I am sure there is a solution for it.
Yes, I already spoke with Don. He has noted many problems with lenses being off. ... All but one lens either back for front focus of the 5 I have used and as mentioned, Don fixed the 35 ASPH and it was very off. I took three shots of the guy near a wall refocussing each time and each time the wall was in focus but not the subject.DAG has adjusted a few of my lenses for use on the M8 as well; some needed it while others didn't.
To clarify, your 35 ASPH was very off AFTER Don "fixed" it? Since the lens does have some focus shift, what lens opening did he adjust to be ON, and is it indeed on at that f/stop?
My 35 Lux ASPH arrived at his shop yesterday for a general look-see as it's not as sharp as expected at any aperture, oddly though maybe a bit sharper on the right than on the left. Certainly different individual lenses may exhibit more or less sharpness or focus shift... I asked Don to put less priority on accurate focus wide open than at f/2-f/8, as wide open is "special use" for me and I can compensate manually for errors there.
I earlier asked the same for the 50 C-Sonnar, and he did a great job on that. I hope for the same quality work on the 35 ASPH so your comment is alarming!
Sorry, I did not mean to convey that he didn't fix it. I meant that it was way off and I sent it too him and he stated that yes, the 35 was off and he fixed it. It is still off a little so I imagine I am going to have the send the camera and lens or lenses.
On another note, I got my Leica bellows today for my Visoflex I. I put on my 135 Elmar hoping everything would match up.
Wide open with the viso and bellows the 135 is exactly on. To the hair wide open. So the viso, with it ground glass that mimics the distance to the film/sensor is right and the camera records an exact sharp image.
Don mentioned that a number of Leica lenses come from the factory out of a true tolerance because they got used to film that allowed for this. Now with the sensor Leica is experiencing problems that require the lenses to be calibrated and brought into "true" specifications.
I may just have had a bad draw of luck but most of all mine are off. The 90 Summicron does not focus correctly at infinity as s few other don't. It doesn't go past infinity but not to it.
I just did some more comparison and focusing with the bellows. The 135 Elmar f4, even wide open, is as good or better than my 180L, 100L and would be a contender with my 60mm Macro (Canon), which I find (the 60) as one of the sharpest lenses I have used. The focus is also totally exact.
Sorry, I did not mean to convey that he didn't fix it. I meant that it was way off and I sent it too him and he stated that yes, the 35 was off and he fixed it. It is still off a little so I imagine I am going to have the send the camera and lens or lenses.Thanks for the clarification; but just due to the nature of this lens (having focus shift) it's going to be a little off except at the aperture Don optimized it for, right? Is it ON at any aperture?
Brian Sweeney
04-03-2010, 18:42
Speaking for myself, it is amazing how much focus shift can be noticed with a digital camera when pixel peeping. Most of my F1.5 lenses are set for wide-open, and F4 shows a big shift. I've set a couple F1.5 J-3's to work best at F2. F1.5 is good-enough, F4 "just makes it". With the film, barely a difference can be noticed.
Yeah, the good ole film days. lol
I love the smell of fixer. I worked in a lab for some time and then my own lab, oh well. What smell from digital? :^)
I don't know about the focus shift on the 35 yet. I have to do some experimenting. All I want to do is get everything dialed in. Everyone assumes that people just pixel peep. When doing a job I don't have time or the desire to do that. I need to shoot and concentrate on just that. So, I need to know what my equipment is doing and that is what I am attempting now. I don't like surprises that piss me off and all the more, a client because I screwed up by not knowing my own equipment.
I spoke with Leica NJ today. As the tech department stated, the sensor on the 8 and 9 is a 1000 times more precise in placement than film and therefore the shifts in focus that people are seeing. I spoke with Don of DAG again today also and he reaffirmed that he has had many 8's and now 9's that needed to have the lenses "fine tuned".
Mark at Leica did state that they felt that the lenses in question were adjusted to specs but since the advent of the sensor previous concepts are changing some (paraphrasing).
Brian Sweeney
04-06-2010, 16:02
> What smell from digital? :^)
I had a Digital camera that "smoked" a mainboard. Pulled the batteries out of it real fast, at least it did not catch on fire. It was much worse than smelling Fixer. What you would get if pouring Fixer into a computer.
Nikon Bob
04-06-2010, 16:48
I spoke with Leica NJ today. As the tech department stated, the sensor on the 8 and 9 is a 1000 times more precise in placement than film and therefore the shifts in focus that people are seeing. I spoke with Don of DAG again today also and he reaffirmed that he has had many 8's and now 9's that needed to have the lenses "fine tuned".
Mark at Leica did state that they felt that the lenses in question were adjusted to specs but since the advent of the sensor previous concepts are changing some (paraphrasing).
I have followed this thread with interest but can't comment directly about the M9 not owning one. At the start I was wondering if the problem the OP had was the reason that a lot of DSLRs have an AF Fine Tune option in their menu for the very reason Leica NJ stated. I would tend to believe that the tolerances are now tighter with digital than film as to what is in spec or not.
Bob
b.espahbod
04-06-2010, 21:13
Well Focusing is an Art when your using Leica or any other RF camera.
> What smell from digital? :^)
The smell of your money going out the door?
Speaking for myself, it is amazing how much focus shift can be noticed with a digital camera when pixel peeping. Most of my F1.5 lenses are set for wide-open, and F4 shows a big shift. I've set a couple F1.5 J-3's to work best at F2. F1.5 is good-enough, F4 "just makes it". With the film, barely a difference can be noticed.
I desperately tried to align a war time LTM Sonnar 1,5 to my M9's RF - and thus discovered the Sonnar's focus shift. Brian, I'm really glad you're writing this - so it's not me, but my Sonnar! I thought I was too stupid for the job, or the lens was faulted. I knew there would be some focus shift but I didn't have an idea how much! Would it be possible that there is some curvature of field as well?
BTW, there has always been a discussion whether the original Voigtländer Nokton or the Sonnar was the "better" high speed 50 of its time (everybody seems to agree that the old Summarit was the worst of the early 1,5 lenses). Now, as I can compare the LTM versions of all three on the M9, it becomes obvious that focus shift very much influences image quality with the Sonnar, and the Nokton (as well as the Summarit) shows much less of it. Even if the use of film somewhat obstructs the effects of focus shift, the Nokton looks much like a winner to me. Stopped down a bit, even the Summarit is not so bad. As modern Sonnar-type lenses such as the new Zeiss Sonnar C 1,5/50 still show over-average focus shift, this seems to be a general property of the Sonnar design.
Peter
Leica called back today.
If a lens is just a few years old they would consider adjusting it at no charge but if more then a few years old they would charge at least 90 dollars or so and the turn around time now is around 4 weeks. So one month. Then I was told that if I have someone else do it I would void the warranty of the camera!
Now I explained that if a repairman, like Don of DAG, of whom they are aware of, of course, adjusted the lens to the camera and not the camera to the lens, which would have to be done since I get both front and rear focus, how would this void the warranty? Well, adjusting the lens to film specs won't get them to the degree of accuracy needed for a sensor. I stated, I was sure he could adjust the lenses to the camera by adjusting them and then checking them on the camera, which would have nothing to do with the warranty of the camera.
Beemermark
04-07-2010, 17:35
I never understood all this "back focus" stuff until I sold a 135/2.0 AF lens to a guy with a D3. He explain he would take the lens if it focused correctly. So I met him, he put it on the digital camera (I'm an old film guy) and proceeded to enlarge the picture on the screen about 10,000x. I was really impressed. If I want to blow a picture up to bill board size I use my 5x7 view camera.
Sure and if you want something in focus and it back focusses 2 feet behind the subject you get the lens or the camera fixed or state that it was whatever is in focus that was the subject to begin with. :^)
Brian has made my J8 usable by adjusting it for spot on at F2, heck, it's even great at 2.8 and ok at F4! and I have a J3 coming that was adjusted by him at some time in the past that I'm looking forward to using in low light.
Even non Sonnar lens have focus shift, my old zoom Nikkors for example, but the shift could be seen when manually stopped down through the VF and adjusted for without wasting film.
I don't think film users want to take 10 shots to focus bracket, so they are generally in denial about focus shift. For digital users who can pixel peep 100% crops with a couple of button presses, they notice it readily.
Here's one thought - with 4/3rds, and using the LCD or EVFs (not the optical VF) to focus through the lens, these older lenses won't necessarily need any adjustment as they do on RFs, you'll see the focus at 7x, 10x or whatever magnification through the lens.
I desperately tried to align a war time LTM Sonnar 1,5 to my M9's RF - and thus discovered the Sonnar's focus shift. Brian, I'm really glad you're writing this - so it's not me, but my Sonnar! I thought I was too stupid for the job, or the lens was faulted. I knew there would be some focus shift but I didn't have an idea how much! Would it be possible that there is some curvature of field as well?
BTW, there has always been a discussion whether the original Voigtländer Nokton or the Sonnar was the "better" high speed 50 of its time (everybody seems to agree that the old Summarit was the worst of the early 1,5 lenses). Now, as I can compare the LTM versions of all three on the M9, it becomes obvious that focus shift very much influences image quality with the Sonnar, and the Nokton (as well as the Summarit) shows much less of it. Even if the use of film somewhat obstructs the effects of focus shift, the Nokton looks much like a winner to me. Stopped down a bit, even the Summarit is not so bad. As modern Sonnar-type lenses such as the new Zeiss Sonnar C 1,5/50 still show over-average focus shift, this seems to be a general property of the Sonnar design.
Peter
Lord Fluff
04-08-2010, 00:56
Well Focusing is an Art when your using Leica or any other RF camera.
In this instance, an 'art' akin to having a paintbrush apply paint either slightly to the left, or at other times slightly to the right, of where you actually wanted it.
In this instance, an 'art' akin to having a paintbrush apply paint either slightly to the left, or at other times slightly to the right, of where you actually wanted it.
...with the paintbrush strapped to a 5-foot pole, in the case of a high-speed lens, or an M9 apparently. Certainly possible, doable and learnable, but "easy" isn't the first word I'd have thought of. (M user here.)
haha, well in the big picture, perhaps the random focus could be a tool for certain kinds of art photography.
In the mid to late '80s, adding "swing" to robotized drum machines was a cool feature.
The difference is, you could set the amount of swing you were willing to tolerate, or turn it off completely...
Art to the focus. No thank you. Sure you can get creative with it, if you want, but what I need to do is concentrate on the image and not on whether the focus is set just a tad back on the focus ring after it has come together on the split image finder.
Art to the focus. No thank you. Sure you can get creative with it, if you want, but what I need to do is concentrate on the image and not on whether the focus is set just a tad back on the focus ring after it has come together on the split image finder.
IMHO, the whole point is that many used or new RF lenses still on the market today have been designed without regard for the specific needs of digital imaging with ff digital RF cameras (flat field, no focus shift, high resolution wide open to the far corners of the frame). I now understand why Leica have made an almost complete modernisation of the RF lens offerings. The old ones were good enough for almost everything, back to the lens designs of the 50's of last century; nowadays focus shift and curvature of field can be seen by every amateur who pixel peeps at 100%; these being the same fine art enthusiasts who - before digital - hardly ever considered an enlargment of 35mm negs exceeding a factor of 8x acceptable.
In a way, it was a very courageous decision of Leica to maintain downwards compatibility with all those old lenses (that now show that what was before considered a specific character - call it "glow" -, sometimes was nothing but simple unsharpness caused by focus shift); maybe they should have changed the lens mount so that only newly designed lenses would fit, so no one could complain.
Just my 2 cents...
Peter
...In a way, it was a very courageous decision of Leica to maintain downwards compatibility with all those old lenses (that now show that what was before considered a specific character - call it "glow" -, sometimes was nothing but simple unsharpness caused by focus shift); maybe they should have changed the lens mount so that only newly designed lenses would fit, so no one could complain.Good point, and Leica did just that with the new S2 SLR. In contrast, the Pentax 645D has legacy lenses to deal with, though not so very far back.
When I spoke to Leica NJ, they made it clear that the older lenses can be adjusted to the M9. They also stated that it was the latitude of compensation that film affords over the sensor that is the problem. And I have spoken with a Leica trained repairman that stated he has seen the calibration of Leica lenses getting worse not better. While the lenses are finely made the last step seems to have slipped a bit but at least it can be corrected.
Canon pissed off a lot of people when they changed their mount but they were a large enough company to absorb the dissatisfaction. I think Leica would have had a more difficult time so it was and is to their advantage to keep the same mount, thus making their smaller market share, still a share and not turning away many possible upgraders to the new bodies and eventually, new lenses.
Nikon Bob
04-09-2010, 19:14
When I spoke to Leica NJ, they made it clear that the older lenses can be adjusted to the M9. They also stated that it was the latitude of compensation that film affords over the sensor that is the problem. And I have spoken with a Leica trained repairman that stated he has seen the calibration of Leica lenses getting worse not better. While the lenses are finely made the last step seems to have slipped a bit but at least it can be corrected.
Canon pissed off a lot of people when they changed their mount but they were a large enough company to absorb the dissatisfaction. I think Leica would have had a more difficult time so it was and is to their advantage to keep the same mount, thus making their smaller market share, still a share and not turning away many possible upgraders to the new bodies and eventually, new lenses.
Both the camera and the lenses are built to certain tolerances. I am thinking that the Lens Rentals article is pointing to what is happening http://www.lensrentals.com/news/2010.03.06/this-lens-is-soft-and-other-facts . It seems to affect both AF camera lenses and MF RF camera lenses from reading the article and what you have said here. Canon has not been free of AF problems even after switching lens mounts and neither has Nikon which has not switched lens mounts. This problem may be the reason Nikon now offers AF Fine Tune in the camera menu. Anyway, it just sounds to me like the two problems are more similar than different.
Bob
When I spoke to Leica NJ, they made it clear that the older lenses can be adjusted to the M9.
No matter how exactly you adjust a lens to the RF, you'll never get rid of focus shift or curvature of field. The best you may expect is an adjustment providing for a sharp image in the middle of the frame at one given stop.
Peter
I realize that but what I don't want is focussing 2 feet in back of the subject, unless I am going for some sort of focus abstract. Free style focussing. :^)
Well I would have to retract some of what I said about focussing. I will agree that to an extent, focussing with the Leica is a bit of art. I understand better the intent of the statement now. I still have some focus issues and need to send two lenses to DAG but in general, there is a bit of feel for what you are doing over what I have done in the past with SLR's though they are used mostly in auto focus. I have used the M3 and M2 but as we progress sometimes we see more of what we are/am doing, or I would say, "I".
vBulletin® v3.6.8, Copyright ©2000-2013, Jelsoft Enterprises Ltd.