"The 50mm Summilux ASPH was Designed to Front Focus" <-- Leica NJ Technician. Thoughts?

[rtai]

Advertisement (gone after registration)

If independent repair people don’t have the equipment to calibrate the Leica APO FLE lenses then I’ll spend my money on something else. I re-entered the Leica digital world late because of a few defective M8 M9 and MM bodies as well as a quirky DMR so no thanks to the Leica support dependency or the lack of it.

Edited November 26, 2022 by rtai

[Jeff S]

3 hours ago, rtai said:

If independent repair people don’t have the equipment to calibrate the Leica APO FLE lenses then I’ll spend my money on something else. I re-entered the Leica digital world late because of a few defective M8 M9 and MM bodies as well as a quirky DMR so no thanks to the Leica support dependency or the lack of it.

DAG can calibrate modern Leica lenses.

Jeff

[Advertisement]

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

[Harpomatic]

On 11/22/2022 at 11:57 AM, jaapv said:

As far as Leica M memory reaches, it has always been the technique for optimal focus to go down from infinity. The other way will be slightly inaccurate due to unavoidable mechanical tolerances in the mechanism. Your lens is calibrated perfectly if focus is spot-on from infinity. Always return it to infinity for the next shot. Hunting focus will result in focus errors on a rangefinder. 
Thank you for highlighting this half-forgotten fact. 

I beg to differ here:

I use Leica M professionally and have used the M system for years now. Some of you will also know that I have a decent technical knowledge from my reviews of the 50mm fast lenses for Leica and other articles on my blog. 

Let’s think it through for a moment here: if that was true, we would never be able to fine tune our focusing should we not hit it perfectly in one movement: as soon as you go past the focus alignment in the patch you come back to correct and align it. Should this “tolerance issue” be real, we would never be able to hit precise focus that way, we would always have to revert to infinity and try again to hit focus from there.

Follow focus would also be impossible, because you would never be able to hit focus at all should the subject move away from you.

What about fast paced documentary shooting? Shooting something close, then mid-distance, then close again, a meter further away…without returning to infinity every single time!

Have you ever done manual focus drills? To practice precision and speed? You choose a far target, a medium distance target, a close target. You focus from one to the other randomly and quickly, checking your focus often to verify that you get it right (digital allows us to get the shots and check them). We would never get anything in focus!

I have done all the above and never, ever found the MFD to point of focus different in precision to infinity to focus.

I have used the M240, M-P240, I’m using the M10-R professionally and testing the M11 right now.

I have used the lens in question, the Summilux 50mm ASPH for a documentary project, and numerous other lenses in the years for thousands upon thousands of frames.

I am testing now the M10-R and M11 with the Summicron 50mm version 5 and the Nokton 50mm 1.5 II at 1.5 and 3 meterts, infinity to focus point and MFD to focus point, both wide open: always spot on, every single time. No difference. 

I think the OP has a problem with his equipment and is absolutely right to be outraged with the response from Leica and some of the responses here on the forum.

Let’s stop justifying the beloved brand and have a look at reality: the rangefinder works and very precisely, certainly precisely enough to get every shot in focus wide open on the Summilux 50mm ASPH at any distance and in any focusing direction - I am talking from direct experience. 

Let’s stop trying to say that we should accept such ludicrous “tolerances” in the most expensive camera system out there short of medium format!

We should expect the tools to work seamlessly and without fail.

 

[jaapv]

1 hour ago, Harpomatic said:

Let’s stop trying to say that we should accept such ludicrous “tolerances” in the most expensive camera system out there short of medium format!

Show me one mechanical linkage without tolerance play...  The only thing you are saying is that your tolerance for misfocus exceeds the tolerance of the rangefinder mechanism...  

[Harpomatic]

1 hour ago, jaapv said:

Show me one mechanical linkage without tolerance play...  The only thing you are saying is that your tolerance for misfocus exceeds the tolerance of the rangefinder mechanism...  

You can make fun of others as much as you like.

I am talking about spot on focus, no misfocusing and pin sharp images. On M10-R and M11 as well, which have a lot less leeway for misfocus when viewed at 100%.

Which means that the mechanical tolerance is small enough to allow for such precision in focusing.

I’ll stop discussing it though, I am a happy Leica M user with no focus errors apart from user mistakes. The OP has a very, very valid point and by no means should accept the level of inaccuracy he is experiencing.

That’s my opinion and experience.

I’ll sit back now and withdraw from the discussion.

[evikne]

Perhaps only partially relevant, but here is my experience:

Tests I make with my 50mm Summilux (pre-ASPH) on a Spyder Lenscal, uncover that the focus is best at f/2, and slightly front focuses at f/1.4. But in real life I just shoot at f/1.4 without any worries, and the focus always seems to be spot on.

[hmzimelka]

Advertisement (gone after registration)

5 hours ago, Harpomatic said:

I beg to differ here:

I use Leica M professionally and have used the M system for years now. Some of you will also know that I have a decent technical knowledge from my reviews of the 50mm fast lenses for Leica and other articles on my blog. 

Let’s think it through for a moment here: if that was true, we would never be able to fine tune our focusing should we not hit it perfectly in one movement: as soon as you go past the focus alignment in the patch you come back to correct and align it. Should this “tolerance issue” be real, we would never be able to hit precise focus that way, we would always have to revert to infinity and try again to hit focus from there.

Follow focus would also be impossible, because you would never be able to hit focus at all should the subject move away from you.

What about fast paced documentary shooting? Shooting something close, then mid-distance, then close again, a meter further away…without returning to infinity every single time!

Have you ever done manual focus drills? To practice precision and speed? You choose a far target, a medium distance target, a close target. You focus from one to the other randomly and quickly, checking your focus often to verify that you get it right (digital allows us to get the shots and check them). We would never get anything in focus!

I have done all the above and never, ever found the MFD to point of focus different in precision to infinity to focus.

I have used the M240, M-P240, I’m using the M10-R professionally and testing the M11 right now.

I have used the lens in question, the Summilux 50mm ASPH for a documentary project, and numerous other lenses in the years for thousands upon thousands of frames.

I am testing now the M10-R and M11 with the Summicron 50mm version 5 and the Nokton 50mm 1.5 II at 1.5 and 3 meterts, infinity to focus point and MFD to focus point, both wide open: always spot on, every single time. No difference. 

I think the OP has a problem with his equipment and is absolutely right to be outraged with the response from Leica and some of the responses here on the forum.

Let’s stop justifying the beloved brand and have a look at reality: the rangefinder works and very precisely, certainly precisely enough to get every shot in focus wide open on the Summilux 50mm ASPH at any distance and in any focusing direction - I am talking from direct experience. 

Let’s stop trying to say that we should accept such ludicrous “tolerances” in the most expensive camera system out there short of medium format!

We should expect the tools to work seamlessly and without fail.

 

Ironic that the only lens I have not had this issue with, is the Voigtlander Nokton 50mm f/1.5 ASPH II.

Unfortunately I've left it in South Africa with a friend for sale, so I cannot repeat test it now again just for my sanity's sake. However, my Summicron 50mm V has this issue too, a little.

Edited November 28, 2022 by hmzimelka

[dugby]

On 11/24/2022 at 2:24 AM, jaapv said:

You should have read this forum - over the years there a dozens of posts mentioning this technique... Those great shots were mostly on film BTW, which is less critical and many of them are not pixel-peeping sharp either. 

The above mentioned technique for the 50FLE of focus down from infinity, is certainly my experience with my 2005 LHSA brass 50mm ASPH lens. It frustrated me enough to buy a 50APO as my regular fifty.

However when I use my 50FLE and take it very slowly using the above technique I get some awesome f1.4 bokeh shots on a M10-R, which for this reason I can’t bring myself to let this LHSA lens travel from the remotest city in Australia to either Wetzlar or USA for a CLA, as the journey (24,000km round trip) will be at the mercy of the most unknowing care-free transportation handlers and likely to come back with different issues.

As I live with the tuning of a Linn Sondek LP12 and Naim Audio warming requirements, I’ve learnt to live with such hand-made tech idiosyncrasies.

I often wonder if early versions of the 50FLE (like mine) are susceptible to this focussing technique with digital Ms ?

 

Edited January 4, 2023 by dugby

[Malvolio]

57 minutes ago, dugby said:

The above mentioned technique for the 50FLE of focus down from infinity, is certainly my experience with my 2005 LHSA brass 50mm ASPH lens. It frustrated me enough to buy a 50APO as my regular fifty.

However when I use my 50FLE and take it very slowly using the above technique I get some awesome f1.4 bokeh shots on a M10-R, which for this reason I can’t bring myself to let this LHSA lens travel from the remotest city in Australia to either Wetzlar or USA for a CLA, as the journey (24,000km round trip) will be at the mercy of the most unknowing care-free transportation handlers and likely to come back with different issues.

As I live with the tuning of a Linn Sondek LP12 and Naim Audio warming requirements, I’ve learnt to live with such hand-made tech idiosyncrasies.

I often wonder if early versions of the 50FLE (like mine) are susceptible to this focussing technique with digital Ms ?

 

Why does Naim gear need “warming”

[dugby]

5 hours ago, Malvolio said:

Why does Naim gear need “warming”

It’s the Naimee’s equivalent to foreplay, without it you may not get euphoria 

[Jeff S]

Many audio systems/components sound better after warmup, not just Naim, some dramatically so.
 

Jeff

[colonel]

5 hours ago, Jeff S said:

Many audio systems/components sound better after warmup, not just Naim, some dramatically so.
 

Jeff

My valves need to "bake" for a few hours before shining

Unfortunately they are ex-soviet valves from a 1970s ICBM control system. Not so politically correct today ....

Anyhoo ... a lot of chat, but a Summilux 50mm f1.4 should focus tack on any Leica M at f1.4. Its the USP of this lens.

Luckily where I am in London, Leica are great, they just send stuff back to Wetzlar, or replace for new if recently sold ...

 

Edited January 4, 2023 by colonel

[Alberti]

On 11/23/2022 at 7:25 PM, jaapv said:

Precisely.

   On 11/23/2022 at 1:53 PM,  hmzimelka said: 

The lens need not be at MFD... simply the direction of focus (closer to further) that will lead to the back focus, no matter if started at MFD, or in the middle of the len's focus throw. If the focusing is in the direction towards infinity, it leads to back focus.

@jaapv        Precisely.

So the direction is the issue, that there is a lagging flaoting group. One direction, it is close, the other far. To me this means that the concept is wrong.

I just did a series of tests. At F/1.4 there is significant close focus (at 1m about a cm or two) and the actual point is lost. At F2 it is rather good. At F/2.8 it starts to drift backwards into the background with a cm or so. Now the FLE in engineering (dixit marketing) was actuall meant to prevent the focus shift. Seems at implementation time something else was done by the technical staff. 😥  get better resolution? Better stick to the Gauss design, folks! My Summicron and Canon F/1.4 are fine. Dependable. 

There is a Danish guy who claimed that the FLE group was driven from the aperture ring. That would have made sense, but I am sure (gut-feeling) that is not the case.

I was afraid upfront about this, now I am very very worried. About my BC-purchase, a beautiful piece of equipment everyone should be proud of.

[lct]

As i said above in 2022, my Summilux 50/1.4 asph didn't suffer from back focus whatever direction i may turn the focus ring. It still works the same in 2024 so i guess your lens needs some repairs it is suffers from this defect.

[pgk]

I have a 50/1.4 asph. It focuses exactly as I would expect - accurately from infinity to MFD regardless of the focus position I start from. I would expect no less. If such a lens does not do so it is in need of appropriate adjustment (which includes the 'floating group' which is intended to maintain image quality at closer distances). If it cannot be adjusted to correct focus at all distances it is faulty.

[hmzimelka]

9 hours ago, Alberti said:

 

   On 11/23/2022 at 1:53 PM,  hmzimelka said: 

The lens need not be at MFD... simply the direction of focus (closer to further) that will lead to the back focus, no matter if started at MFD, or in the middle of the len's focus throw. If the focusing is in the direction towards infinity, it leads to back focus.

@jaapv        Precisely.

So the direction is the issue, that there is a lagging flaoting group. One direction, it is close, the other far. To me this means that the concept is wrong.

I just did a series of tests. At F/1.4 there is significant close focus (at 1m about a cm or two) and the actual point is lost. At F2 it is rather good. At F/2.8 it starts to drift backwards into the background with a cm or so. Now the FLE in engineering (dixit marketing) was actuall meant to prevent the focus shift. Seems at implementation time something else was done by the technical staff. 😥  get better resolution? Better stick to the Gauss design, folks! My Summicron and Canon F/1.4 are fine. Dependable. 

There is a Danish guy who claimed that the FLE group was driven from the aperture ring. That would have made sense, but I am sure (gut-feeling) that is not the case.

I was afraid upfront about this, now I am very very worried. About my BC-purchase, a beautiful piece of equipment everyone should be proud of.

The lag comes from how much "slack" is in the helicoids. Most lenses have it, and so far I have only had one lens that doesn't change focus based on which direction one turned the focus ring, and that was the Nokton 50mm f/1.5 ASPH II which I've since sold. So for Leica mount lenses, always focus from infinity to the subject for consistent focus. 

The FLE is definitely not driven by the aperture ring. The rear group is rotated with the front group, moving back and forward, only difference is the rear group either rotates on a cam or on a helicoid that allows the rear group to move a slightly different distance relative to the front group. This allows for better optimisation for spherical aberration and other properties that allow a lens to have a good balance of near and far focus performance. Focus shift can largely be reduced or eliminated using aspherical lenses already, the FLE group can reduce it even more. 

The AFS Nikkor E 28mm f/1.4 ED has internal focusing, aspherical and ED lenses, and a floating group and still focus shifts. Look for focus shift and one will find it, however its most often so small that it doesn't (shouldn't) get in the way of things. The Summicron-M 50mm also has focus shift. 

The Summilux 35mm ASPH FLE (both new and old) focus shift a lot more than the Summilux 50mm ASPH. I've tried two copies where the focus will move backwards behind the subject after f/2 and be slightly out of focus until somewhere between f/4 and f/5.6. The other copies I tried were very well behaved and the subject never left focus at any aperture setting. 

The way a lens is assembled, collimated and adjusted will likely influence the focus shift from one lens copy to the next. Every 50mm Summilux I've tried to date has been slightly different to the next one. It's been one of the most inconsistent lenses I've tried from Leica, followed by the Summilux 35mm. My experience last month with both new copies of the new close focus Summilux lenses, 50mm and 35mm, has given me peace of mind that I don't need or want a new Leica lens. 

[Alberti]

Sounds horrible, specifically what you said about the 35mm FLE too.

I did a lot of checking last night and at 1.4 there is a cm or more front, about right at f/2, then after f/2.8 it is slightly back and fades into the generall sharpness. Really annoying is that on my M9M it looks consistent; on the M10-R I see everything.

Quote

copies where the focus will move backwards behind the subject after f/2

Still have to test my 35 FLE thourougly, but the pictures at F1.4 looked roughly good. 

Note I traded in : a 50-Summilux V2 and a 35mm Summicron V4; very and extremely sharp. Should I not cry what came over me?? 

Center crops. All at about 1,5 m. Wide open, the vertical in the 4 is what I targetted. The 5 is sharp though.

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!

good: 

moving back at 2.8 and after that it is all sharp.

To me a picture with a false plane of focus - because it only goes to the front - is utterly unacceptable:

Edited March 20 by Alberti

[hmzimelka]

5 minutes ago, Alberti said:

Sounds horrible, specifically what you said about the 35mm FLE too.

I did a lot of checking last night and at 1.4 there is a cm or more front, about right at f/2, then after f/2.8 it is slightly back and fades into the generall sharpness. Really annoying is that on my M9M it looks consistent; on the M10-R I see everything.

Still have to test my 35 FLE thourougly, but the pictures at F1.4 looked roughly good. 

Note I traded in : a 50-Summilux V2 and a 35mm Summicron V4; very and extremely sharp. Should I not cry what came over me?? 

At about 1,5 m. Wide open, the vertical in the 4 is what I targetted. The 5 is sharp though.

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!

good: 

moving back at 2.8 and after that it is all sharp.

That sounds like too much focus shift for a Summilux 50mm ASPH. 

I'll PM you a link with  DNG files that I took on a tripod with a fixed focus point on very textured paving. There you'll see how focus shift progresses through the aperture range to what I would consider normal/reasonable amounts.

 

[Alberti]

Like the the worried original poster says who hurt his eyes trying in vain to see correct focus 😉

My pictures are hand held but show it. 

I sold my excellent Jupiter-3 lenses because of focus shift (I could not get it into line in shooting habit with my excellent Summicrons 35 v1+4, 50 v5). I now also have a Canon 50mm 1.4 LTM that shows no propensity to drift either. 

• I checked my also acquired 35 FLE (I) now and it shows no drift in any way, it is sharp at ever RF point and through the aperture stops. I used life view of an object on 70 cm with clear letters and turned the dia. Same with bookcase at 3 m. It amazes me how good it is. Say 99,5%??

Edited March 20 by Alberti

[Advertisement]

Hallo,

Look here: Leica M

[Al Brown]

My two 50 summiluxes ASPH (black and chrome) have had no problem whatsoever with front focus. It is 99,7% precise, and I have used or tested at least three other samples, including the retro housed special edition. Although my small number is statistically irrelevant it proves that the lens SHOULD NOT FRONT OR BACK FOCUS. If it does, there is something wrong with your camera's cam or with the lens assembly.

I support everything that @Harpomatic said, and @jaapv and @hmzimelka are correct about go-from-infinity technique for nailing focus - one of the most important unwritten knowledge facts, especially for lenses like the always delicate summilux 75/1.4...