problem with front focusing

[adli]

Advertisement (gone after registration)

I've found the best way to check focus on a M is to put a very tiny object on a single colored textile surface witch has a coarse texture, so you can easily identify the focus plane.

 

Focus on the tiny object and you will clearly see where the focus plane is on the textile.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[hydeca]

Frankly I didn’t expect service like that. I send pictures, send my test equipment. All the time very similar result. As well as scratches.

 

Sheesh that's been butchered. I'd be asking for a replacement.

[jaapv]

I'm not sure what you mean; the two points on the ruler at 45 deg. are different distances from the camera, so they should show different focus.

 

If you mean there is an error from using the edge of the patch to focus, compared to the center--I've heard that before, but not seen any rationale or practical demonstration.

 

If you mean there is an error from using an oblique target, I've heard that one too, but not seen any rationale or practical demonstration.

Not too difficult - just visualize the way the system works. An oblique target changes the RF generated focus plane into a gradient of focus from top of the RF patch to the bottom, thus into a zone of focus. See this rough sketch. You can use an oblique target - if you manage to test focus distance exactly in the optical axis of the RF system i.e. the middle of the RF patch. With amateur-type rulers that is very error-prone. Better to use a parallel target and use the whole patch.

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!

[pico]

My ignorance of the RF is amazing, considering that I've used them for decades on a number of different cameras (Leica, Super Technika, Printex, Super Ikonta, Graflex), but in each I've found that there is a sweet-spot for focus and none are exactly correct throughout the whole range.

 

To test an M9 with a lens known to have focus shift at a close distance at an aperture other than that which for which the RF was designed might be unrealistic. No?

[ho_co]

Not too difficult - just visualize the way the system works. An oblique target changes the RF generated focus plane into a gradient of focus from top of the RF patch to the bottom.... You can use an oblique target - if you manage to test focus distance exactly in the optical axis of the RF system i.e. the middle of the RF patch. With amateur-type rulers that is very error-prone. Better to use a parallel target and use the whole patch.

 

That's why an experienced M user will turn the camera vertical to focus on such a patch as the one I mentioned above, the Nikon D70 Focus Chart, which is cheap and gives "real-world" results. It's useless if used with the M in horizontal position, as Jaap says.

 

Jaap, I think the black-and-white square pattern just to the left of the LensAlign ruler is the actual focus target in Elia's picture at http://www.l-camera-forum.com/leica-forum/leica-m9-forum/173793-problem-front-focusing.html#post1654954. The rest of the LensAlign contains controls to be used to guarantee the parallelism of the target with the camera's sensor plane. I think what we're seeing is a crop.

[likalar]

.... but I expect for 7k at least that repair done right. I have it for 5 month and 3 of them this camera been at Leica repair center.

 

Please save all your shipping documents and correspondence regarding time the camera was at the factory repair facilities. Documentation will be helpful in case you request the warranty be extended by the amount of time that the camera was in their hands.

 

Larry

[Guest #12]

Advertisement (gone after registration)

Not too difficult - just visualize the way the system works. An oblique target changes the RF generated focus plane into a gradient of focus from top of the RF patch to the bottom, thus into a zone of focus. See this rough sketch. You can use an oblique target - if you manage to test focus distance exactly in the optical axis of the RF system i.e. the middle of the RF patch. With amateur-type rulers that is very error-prone. Better to use a parallel target and use the whole patch.

 

Thank you for the diagram, that's great. (Sorry to drag up an old thread, but I missed this one.)

 

O.K., there will always be some angular uncertainty, making a little triangle between two focus planes as in your figure. If you use the usual 1/1000 figure, you can do this one in your head--the error is less than a millimeter. (1m distance, 45 deg. target.) So it's not really significant.

 

What I wonder is, your picture doesn't show the focus patch as a moving mask...there might be an error, just shooting a point target, from using the edge of the patch versus using the center. ???

[Guest #12]

...

Jaap, I think the black-and-white square pattern just to the left of the LensAlign ruler is the actual focus target in Elia's picture at http://www.l-camera-forum.com/leica-forum/leica-m9-forum/173793-problem-front-focusing.html#post1654954. The rest of the LensAlign contains controls to be used to guarantee the parallelism of the target with the camera's sensor plane. I think what we're seeing is a crop.

 

I think this is a different brand ... $60, no "gun-site" alignment. There's a thread devoted to it.

[pop]

I don't follow ... As long as the camera is square to the test chart, and you focus somewhere on the centerline, you will get the same result if the camera is vertical or horizontal.

 

Due to the way the rangefinder works, you need to focus on a line or edge which runs vertical if the camera is in "landscape" orientation. If you can focus on the small lettering in the center bar of some focusing charts, your eyesight is much better than mine. Otherwise, you turn the focusing chart such that the line you are focusing on runs parallel to the shorter side of the image.

[Guest #12]

That's why an experienced M user will turn the camera vertical to focus on such a patch as the one I mentioned above, the Nikon D70 Focus Chart, which is cheap and gives "real-world" results. It's useless if used with the M in horizontal position, as Jaap says.

...

 

I don't follow ... As long as the camera is square to the test chart, and you focus somewhere on the centerline, you will get the same result if the camera is vertical or horizontal. No?

[Guest #12]

Due to the way the rangefinder works, you need to focus on a line or edge which runs vertical if the camera is in "landscape" orientation. If you can focus on the small lettering in the center bar of some focusing charts, your eyesight is much better than mine. Otherwise, you turn the focusing chart such that the line you are focusing on runs parallel to the shorter side of the image.

 

yes, I have drawn in a cross on that particular chart. I thought H.C. meant something else, though.

[ho_co]

Drawing a cross doesn't help for the reason Jaap mentioned--the top part of the cross is further away than the bottom part.

 

But doing so does mean you're aware of the problem.

 

Anyone who has seriously worked with the Leica M will automatically seek out (as much as possible) a line perpendicular to the rangefinder's focusing movement, IMHO.

[Guest #12]

...

What I wonder is, your picture doesn't show the focus patch as a moving mask...there might be an error, just shooting a point target, from using the edge of the patch versus using the center. ???

 

No error I see, because you can slide the camera.

[Guest #12]

Drawing a cross doesn't help for the reason Jaap mentioned--the top part of the cross is further away than the bottom part.

 

...

 

You focus on the center of the cross, in effect. You can focus on a point whether the camera is horiz. or vertical.

 

Again, I don't think the error from the target being oblique has any significance. There is no variance when I shoot a target at 1m.

[ho_co]

Good for you, #.

[jaapv]

Thank you for the diagram, that's great. (Sorry to drag up an old thread, but I missed this one.)

 

O.K., there will always be some angular uncertainty, making a little triangle between two focus planes as in your figure. If you use the usual 1/1000 figure, you can do this one in your head--the error is less than a millimeter. (1m distance, 45 deg. target.) So it's not really significant.

 

What I wonder is, your picture doesn't show the focus patch as a moving mask...there might be an error, just shooting a point target, from using the edge of the patch versus using the center. ???

You mean the parallax compensation - indeed that is an added complication in this setup. Again, the easy way out is to set up a vertical focussing target beside your angled ruler to facilitate precise focussing. Or set up an elaborate system like Leica has, which has cost an amazing number of Euros.

 

In general I see little sense in this "testing" by the average user. If you are not able to do the adjusting properly - and this is the territory of professional Leica technicians and very accomplished hobbyists- it is more sensible to develop the impression that your images are off, and send the camera in to either Leica or one of the reputable Leica repair facilities, where they will do proper testing and adjustment. Not only is it not expensive, mostly the turnaround is very quick. It is advisable to do so as soon as one feels the need. A properly set up rangefinder is a joy to use and is accurate within millimeters. Without magnifiers too. An other parameter to get good results is of course to have a proper diopter correction on the camera.The third being a proper focussing technique. See the M9 FAQ for that.

 

Btw, lens adjustment - hence testing - is not done at just one meter. It is also at three meters , ten meters and infinity.

[jaapv]

 

Jaap, I think the black-and-white square pattern just to the left of the LensAlign ruler is the actual focus target in Elia's picture at http://www.l-camera-forum.com/leica-forum/leica-m9-forum/173793-problem-front-focusing.html#post1654954. The rest of the LensAlign contains controls to be used to guarantee the parallelism of the target with the camera's sensor plane. I think what we're seeing is a crop.

That test you link to is a funny one, as the lack of distance perspective in the shot suggests the ruler was parallel to the sensor, instead of at the proper angle....

[ho_co]

In general I see little sense in this "testing" by the average user. If you are not able to do the adjusting properly - and this is the territory of professional Leica technicians and very accomplished hobbyists- it is more sensible to develop the impression that your images are off, and send the camera in to either Leica or one of the reputable Leica repair facilities, where they will do proper testing and adjustment.

...

Btw, lens adjustment - hence testing - is not done at just one meter. It is also at three meters , ten meters and infinity.

 

Amen to that!

 

For a dSLR with lens-specific focus adjustment capability, something like the LensAlign is a good way to adjust your own equipment.

 

When I find a lens isn't making the picture I expect, I send it to Leica.

[Advertisement]

Hallo,

Look here: Leica M

[Guest #12]

Amen to that!

 

...

 

Brother Howard, you can't Amen testing abstinence after you just Amen'd the D70 Focus Test Chart twice in this thread

[ho_co]