Mysterious Summilux 75/1.4 focusing problem

[dennersten]

Advertisement (gone after registration)

I have bought a good sample 75/1.4, but it has a seemingly random focusing problem. I have a M11 with Visoflex it works 100% if you use that. But I want to use it with the rangefinder also of course and use it on my M9 and M6. 

Yesterday I took a walkabout in our train station I shot only simple objects to focus on to minimize user error. I shot only at 1.4 

Picture 1-5 was good, 6 to 15 focus was way off, 16-17 good, 18-20 focus way off, 21-30 Good  etc 

It seems to be totally random, and when the focus is off, it's way off. See the attached example. I was focusing on the screen above the head of the phone speaker. 

 

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!

[pgk]

I bought one some years ago which simply would not focus at all! Regardless of the focus setting nothing was in focus whatsoever. The dealer shipped it to Leica before letting me have it and it worked perfectly afterwards so there was a mechanical issue somewhere. If this continues I would ask Leica to look at it, or a highly competent repairer.

Edited May 30, 2023 by pgk

[Advertisement]

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

[wizard]

Just a thought, because the focus is way off indeed: Could it be that you lined up the wrong pair of "lines" in your rangefinder? To explain a little more, if you are focussing on an object that has a left line and a right line delimiting its physical body, it has occasionally happened to me that I was aiming at the left line of that body, but did in fact turn the focussing helicoid of the lens so far until the right line of that body matched the left line I was aiming at, resulting in the focus being completely off.

[wizard]

vor 30 Minuten schrieb pgk:

... a mechanical issue somewhere.

If, as the OP says, it works 100% with the Visoflex, then I don't think it is a mechanical problem, at least not with the lens.

[pgk]

Just now, wizard said:

If, as the OP says, it works 100% with the Visoflex, then I don't think it is a mechanical problem, at least not with the lens.

This depends on whether the lens is focusing correctly in terms of focus scale and rangefinder, as well as the Visoflex (ie when focussed at 1.5m on the visoflex is this the setting on the lens and does the visoflex agree with the rangefinder too?). Mine wouldn't focus at all and if anywhere appeared to be focusing somewhere too close for the camera (ie forward of the closest focus on the lens or which the rangefinder would work at). Something had obviously come out of alignment within the lens and if this is happening it may or may not allow for visoflex focusing and might not operate the rangefinder nor be aligned with the distance scale. My best guess was that linkage between the focus sleeve and helicoid had shifted somehow but I have no knowledge of the mechanics of the lens so this is a guess.

[LocalHero1953]

This lens is notorious for r/f focusing and calibration issues., and there are other threads on the forum about this. I sent mine to Skyllaney in the UK for 6-bit coding. I also asked for the calibration to be checked. I was sent these comments:

"Just a note, these lenses are notorious for being calibration compromised.

What this means is, their effective focal lengths often do not match the helicoid exactly (on purpose), which results in usually the lens focusing accurately at either infinity or close distances, but not both; and this is how Leica set them up to try and balance out their focus shift. If using film, often this is not much an issue as film is often forgiving.  With digital, the issue is seriously magnified and can be an annoyance to some. The good thing is the later digital M’s have live-view/ EVF attachments that can permit the critical focus.

With that said, at best, usually the 75mm Summilux can be optimized for a certain aperture and distance range (say f/1.4 at close to mid distances).  As you close the aperture down to f/2.8 or so, the focus shift could then be utilized to get the far distances/ infinity subjects accurately in focus.

Its not really possible to get the 75mm Summilux say focus accurately from 0.8m to Infinity at apertures f/1.4 thru f/16."

 

and with respect to my own lens:

 

"The lens seems to be optimised for f/1.4 wide open at the close-mid distance.  The lens has focus shift, so by f/4, the focus shift has already caused the lens to go out of focus.

From my quick tests, this seems to be how the lens is setup: f/1.4 - Close to Mid distances, hard stop of the focus ring at infinity will cause infinity to be sharp also (though the RF patch does not fully align, which is to be expected with the focus shift formulas when they optimise for close focus distances).

The real issue is the mid to long range distances, you either need to shoot wide open, or stop down to about f/8, because the focus shift from f/2.8 thru f/5.6 will cause the lens to not be in the same focal plane as the RF patch.  In this way, the lens acts like a Sonnar formula.

I know the 75mm Summilux is notorious for getting calibrated into an acceptable state, it can only really achieve a single aperture and distance zone optimisation due to how the formula behaves. If you plan on using this lens wide open for close-mid portrait distances, this is nearly where the lens currently is calibrated to in its present form."

 

This seemed satisfactory to me as far as I am qualified to judge. I can live with it as it is.

You are in the best position to judge if these comments apply to your lens performance, and whether you wish to get it adjusted to your usage.

[dennersten]

Advertisement (gone after registration)

57 minutes ago, wizard said:

Just a thought, because the focus is way off indeed: Could it be that you lined up the wrong pair of "lines" in your rangefinder? To explain a little more, if you are focussing on an object that has a left line and a right line delimiting its physical body, it has occasionally happened to me that I was aiming at the left line of that body, but did in fact turn the focussing helicoid of the lens so far until the right line of that body matched the left line I was aiming at, resulting in the focus being completely off.

No that's why i shot text and similar to rule out user error 

[dennersten]

Thanks @LocalHero1953 i think you cracked it. Looking at all my testpictures again it seems the lens is calibrated at very short length. Under 4-5 meters seems to be good.  Rest comes out as crappy. 

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!

[pgk]

49 minutes ago, LocalHero1953 said:

..... and with respect to my own lens:

Mine worked absolutely fine throughout the focus range after its visit to Leica. I shot both close up and landscapes with it and used it just like any other lens. Perhaps I was lucky? Ultimately though its bulk and (dare I say it) the 'bokeh' was not to my taste, so I sold it. The Summarit now serves me well and I prefer the lens to even the fabulously precise Summicron.

Mine was out of focus not due to focus shift or callibration but simply because there was an internal mechanical problem. Looking at dennersten's first photo I would say that this has nothing to do with shift or callibration either. Its either misfocus or an intermittent mechanical problem because, as stated in the OP focus is 'way off'.

[Al Brown]

It’s one of the most horrible lenses to focus accurately. There’s the close-focus-to-focusing-distance vs. infinity-to-focusing-distance technique that says if you start focusing from one of those positions you have a much bigger chance to nail the exact focus. Try how those work for your sample.

Edited May 30, 2023 by Al Brown

[hmzimelka]

It's a good point to raise. With rangefinder lenses, the minuscule mechanical play within the various helicoids result in a different focus point based on which direction one focused from. 
 

As mentioned above, focusing (with rangefinder) from minimum focus towards the subject will result in a back focus when compared to focusing lens from infinity position to the subject. 
 

It's pretty strong with my Voigtlander 75/1.5 Nokton but present in many lenses. Even in the APO Summicron-m 50mm Asph I shortly owned. 

[evikne]

A Spyder LensCal is a very useful tool for revealing focus shift and getting to know your lenses so that you can take the right precautions if necessary. The large grid pattern is perfect for focusing with a rangefinder. Here is from a test I did with my 75 Summilux after the fourth (!) submission to Leica. Then the focus was finally perfect. I told them that I was mainly going to use the lens wide open at "portrait distance", 2-3 meters.

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!

Edited May 30, 2023 by evikne

[pgk]

The photo in the OP is NOT as a result of focus shift!

[LocalHero1953]

4 minutes ago, pgk said:

The photo in the OP is NOT as a result of focus shift!

This has to be true, since they are all at f/1.4!

Edited May 30, 2023 by LocalHero1953

[lct]

Focus shift happens at f/4-f/5.6 on mine so no problem at f/1.4.

[dennersten]

I got a call from the repair shop. It was one lens group which was a little loose. Problem solved. Thanks for all your insights. Will be fun testing it when it comes back. 

 

 

Edited June 4, 2023 by dennersten

[Kiwimac]

I look forward to the FLE version of this lens (although not the price increase that comes with it!!).

[69xchange]

On 5/30/2023 at 2:46 AM, evikne said:

A Spyder LensCal is a very useful tool for revealing focus shift and getting to know your lenses so that you can take the right precautions if necessary. The large grid pattern is perfect for focusing with a rangefinder. Here is from a test I did with my 75 Summilux after the fourth (!) submission to Leica. Then the focus was finally perfect. I told them that I was mainly going to use the lens wide open at "portrait distance", 2-3 meters.

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!

I sent my 75mm to DAG and it focus perfectly on my M9 @ wide open from close distance to near infinity. However when at infinity, it front focuses by quite a bit. Basically, I can’t shoot at infinity @ 1.4 - 2.8 until I stop it down to f4 then it’s sharp and in focus at infinity. Did you have this issue as well? 
 

I heard about the focus issues with these lenses and you can only pick one or the other.. is this a result of having perfect focus from close distance to semi- infinity @ wide open?

[evikne]

13 minutes ago, 69xchange said:

I sent my 75mm to DAG and it focus perfectly on my M9 @ wide open from close distance to near infinity. However when at infinity, it front focuses by quite a bit. Basically, I can’t shoot at infinity @ 1.4 - 2.8 until I stop it down to f4 then it’s sharp and in focus at infinity. Did you have this issue as well? 

I don't know. I don't see any reason to shoot wide open at infinity, so I have never tried.

For me, the widest apertures are mainly for use at shorter distances, to separate the subject from the background.

Edited March 23 by evikne

[Advertisement]

Hallo,

Look here: Leica M

[69xchange]

9 minutes ago, evikne said:

I don't know. I don't see any reason to shoot wide open at infinity, so I have never tried.

For me, the widest apertures are mainly for use at shorter distances, to separate the subject from the background.

I see, that is true. I am debating if I should send the lens back to DAG for inspection to make sure everything is working properly. Just wanted to know if this is “normal” for the lens before sending it back