M11 Motion blur is driving me crazy(!)

[wijsbroek]

Advertisement (gone after registration)

On 5/14/2023 at 3:46 PM, lct said:

The 35/1.4 FLE keeps decently sharp on close-ups at full aperture thanks to its floating elements. FF and crop at about 1.7m below (M11 handheld @ f/1.4). 

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 did take some excellent sharp photos too! 🙂

[Advertisement]

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

[wijsbroek]

19 hours ago, hdmesa said:

Hey, Harry. Call me crazy, but in looking at the images on your website, I have a hard time believing you could take a shot this bad. The point of focus looks like it should be completely outside the frame below the image.

Can you shoot another test and instead focus at the center of the frame and make sure you don't recompose after focusing? Please make no adjustments to the DNG before exporting it. Also please do not strip out the EXIF metadata when you export the image so we can see that it was shot with an M11.

Hey James. I'll take it as a compliment 😉

I will be testing again soon and will post a DNG here if I find a situation like this. I did photograph some very sharp (non-shaken) photos at f/1.4. I posted 2 of them in this thread. I have a feeling it's getting tricky if the subject which is in focus is like 10 feet away or more. Thanks for your help.

[lct]

44 minutes ago, wijsbroek said:

I have a feeling it's getting tricky if the subject which is in focus is like 10 feet away or more.

This has little to do with motion blur or camera shake then right? Besides inaccurate focusing, you may be experimenting that the lens is softer at f/1.4 than above somewhat expectedly. Crops at about 5m below.

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!

 

 

[wijsbroek]

1 hour ago, lct said:

This has little to do with motion blur or camera shake then right? Besides inaccurate focusing, you may be experimenting that the lens is softer at f/1.4 than above somewhat expectedly. Crops at about 5m below.

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!

 

 

Thanks for your examples. I understand that lenses get sharper while stopping down (not too far). Maybe I am too demanding, but I've never experienced this (my) kind of unsharpness. I once saw this interview with the chief lens designer (?) of Leica where he stated that all lenses were designed to shoot wide open.

[Kwesi]

1 minute ago, wijsbroek said:

Thanks for your examples. I understand that lenses get sharper while stopping down (not too far). Maybe I am too demanding, but I've never experienced this (my) kind of unsharpness. I once saw this interview with the chief lens designer (?) of Leica where he stated that all lenses were designed to shoot wide open.

Nt sure if this has been suggested elsewhere in this thread but it night be worth popping into a Leica store  - if there's one close enough to you - and comparing your lens's performance to theirs.

[lct]

1 minute ago, wijsbroek said:

Thanks for your examples. I understand that lenses get sharper while stopping down (not too far). Maybe I am too demanding, but I've never experienced this (my) kind of unsharpness. I once saw this interview with the chief lens designer (?) of Leica where he stated that all lenses were designed to shoot wide open.

I would not draw any conclusion from your pic above and would redo a test on a 3D subject matter as suggested above.

[Tailwagger]

Advertisement (gone after registration)

On 5/13/2023 at 7:46 AM, wijsbroek said:

Thanks, I used the visoflex 2. I think it was in focus. Seems like it in the picture too (when not pixel peeping 😃).

Ah.. but how did you use it?

There are several methods with the Viso from straight up with no aids to focus peaking and finally zoomed view at 5x or 10x (with or without recomposition).  The results can, depending on methodology, distance, aperture, etc vary wildly.  The greatest accuracy (and time to shot) is typically achieved utilizing the cursor to find the focus point and zooming 10x.  Focus peaking is notoriously unreliable and without the zoom function one can easily be fooled into thinking things are fine when they're slightly off... particularly when wide open, close range.

[J S H]

There is an obvious answer to the question about this particular shot. The effective speed (time to read the whole sensor) of the electronic shutter in the M11 is around 1/10s, if I recall correctly. The actual issue is primarily camera shake due to the slow effective speed of the electronic shutter.  Also, maybe a little mis-focus on top of that. 

[SonneCH]

Because f1.4 cannot give deeper depth of field. Try F22 to focus from far to near.

[jaapv]

14 hours ago, wijsbroek said:

Thanks for your examples. I understand that lenses get sharper while stopping down (not too far). Maybe I am too demanding, but I've never experienced this (my) kind of unsharpness. I once saw this interview with the chief lens designer (?) of Leica where he stated that all lenses were designed to shoot wide open.

If you want ultimate sharpness at this high resolution on a ff sensor you need to use the newest APO lenses by Leica. The resolution shows up the limits of performance of your older lens - and is irrelevant for use at normal sizes.

[wijsbroek]

12 hours ago, Tailwagger said:

Ah.. but how did you use it?

There are several methods with the Viso from straight up with no aids to focus peaking and finally zoomed view at 5x or 10x (with or without recomposition).  The results can, depending on methodology, distance, aperture, etc vary wildly.  The greatest accuracy (and time to shot) is typically achieved utilizing the cursor to find the focus point and zooming 10x.  Focus peaking is notoriously unreliable and without the zoom function one can easily be fooled into thinking things are fine when they're slightly off... particularly when wide open, close range.

I used 10x focus peeking + cursor. 

[wijsbroek]

11 hours ago, J S H said:

There is an obvious answer to the question about this particular shot. The effective speed (time to read the whole sensor) of the electronic shutter in the M11 is around 1/10s, if I recall correctly. The actual issue is primarily camera shake due to the slow effective speed of the electronic shutter.  Also, maybe a little mis-focus on top of that. 

Thanks for your answer. Are you saying that despite of the shutter speed indicated (faster than 1/4000s), the sensor’s real speed is 1/10s?

[LocalHero1953]

36 minutes ago, wijsbroek said:

Thanks for your answer. Are you saying that despite of the shutter speed indicated (faster than 1/4000s), the sensor’s real speed is 1/10s?

It means that it takes 1/10 sec to read the pixels from top to bottom of the sensor. An individual pixel only has 1/4000 sec exposure, but one at the bottom will be read 1/10 sec later than one at the top. 

This is one aspect of digital photography that has got worse in some cameras as resolution has increased. IIRC from my own trials the CL and SL electronic shutters had a read-out of about 1/30 sec for their 24mp sensors. For most shots this gave me few problems of distorted movement.

[wijsbroek]

15 minutes ago, LocalHero1953 said:

It means that it takes 1/10 sec to read the pixels from top to bottom of the sensor. An individual pixel only has 1/4000 sec exposure, but one at the bottom will be read 1/10 sec later than one at the top. 

This is one aspect of digital photography that has got worse in some cameras as resolution has increased. IIRC from my own trials the CL and SL electronic shutters had a read-out of about 1/30 sec for their 24mp sensors. For most shots this gave me few problems of distorted movement.

Okay, thank you. Did not know that! 

[Photoworks]

I think you just need some practice with this lens.

you should remember that the Summilux lens has some glow at 1.4 and as you close down it becomes as sharp as the Summicron's quite quickly.

Just because you have the 1.4 it does not mean you have to shoot it at 1.4. I make a decision of the F-stop on the effect I want to get. That is why I suggest using the lens quite a bit to lens the qualities of it.

Like suggested before is that the lens doesn't have a linear focus across the field, this lens is always softer in the corner at 1.4. ( the new lens has gained some corner sharpness)
You will find the max sharpness in the middle. some of this is intentional to give the look of 3D, making your subject stand out.
 

If you wanted the arrow more in focus, try f2 or 2.8.

I often shoot with the Noctilux and use 0.95 for soft seen, if I get lots of backlight I will go to f1 or 1.4 to avoid CA

[CDodkin]

I have also been using the previous Summilux 35/1.4 on my M11, and have not been 100% satisfied with the results.

I have come to the conclusion that the lens is not up to the resolution of the M11 in critical use. It's a great lens, just not a great lens on the M11.

It's not motion blur you're seeing, but the sensor out-resolving the lens.

I somewhat anticipated this might be the case - as Leica have been refreshing their key lenses of late, likely knowing they'd run into this issue with the M11 sensor.

My solution was to try the Voigtlander 35mm f/2 APO Lanthar, which I've been test shooting with for the last week.

I believe I'll be keeping the Voigtlander - It's a significant upgrade in image sharpness. And for the modest price compared to the Summilux, a bargain. I note Hugh Brownstone has also converted to using the Voigtlander 35mm F/2 APO - and he's all about the resolution of his images for cropping. https://www.instagram.com/p/CcaYRZbOV76/

You can buy via Amazon, and return if it doesn't work for you - maybe this would be something you could try.

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 16, 2023 by CDodkin

[charlesphoto99]

18 minutes ago, CDodkin said:

I have also been using the previous Summilux 35/1.4 on my M11, and have not been 100% satisfied with the results.

I have come to the conclusion that the lens is not up to the resolution of the M11 in critical use. It's a great lens, just not a great lens on the M11.

It's not motion blur you're seeing, but the sensor out-resolving the lens.

I somewhat anticipated this might be the case - as Leica have been refreshing their key lenses of late, likely knowing they'd run into this issue with the M11 sensor.

My solution was to try the Voigtlander 35mm f/2 APO Lanthar, which I've been test shooting with for the last week.

I believe I'll be keeping the Voigtlander - It's a significant upgrade in image sharpness. And for the modest price compared to the Summilux, a bargain. I note Hugh Brownstone has also converted to using the Voigtlander 35mm F/2 APO - and he's all about the resolution of his images for cropping. https://www.instagram.com/p/CcaYRZbOV76/

You can buy via Amazon, and return if it doesn't work for you - maybe this would be something you could try.

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!

The first version of the 35mm Summilux Asph had front focus issues from about f2 to f5.6. I know, I had one. And ftr, any lens can resolve any sensor (as well as the bottom of a coke bottle). Its the lens that might be the issue - you just might not like the look or it has focus etc issues. Keep in mind Leica just released an expensive 35 1.4 that is beloved FOR its flaws. Nothing to do with the ability to resolve. IMO that is not what Leica are considering when developing new lenses - they’re going for exotic (Thambar) and new abilities like close focus. 

Edited May 16, 2023 by charlesphoto99

[CDodkin]

Post removed - Dupe post

Edited May 16, 2023 by CDodkin
Duplicate post

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[CDodkin]

10 minutes ago, charlesphoto99 said:

The first version of the 35mm Summilux Asph had front focus issues from about f2 to f5.6. I know, I had one. And ftr, any lens can resolve any sensor (as well as the bottom of a coke bottle). Its the lens that might be the issue - you just might not like the look or it has focus etc issues. Keep in mind Leica just released an expensive 35 1.4 that is beloved FOR its flaws. Nothing to do with the ability to resolve. IMO that is not what Leica are considering when developing new lenses - they’re going for exotic (Thambar) and new abilities like close focus. 

Unfortunately you cannot fully exploit the resolution potential of high-resolution (M11) sensors with the 'older' Leica lenses. The latest ($10k) Leica APO can, as can the Voigtlander APO.

This is something that people purchasing the M11 need to be aware of, and factor into their long-term lens purchasing strategy.

[jaapv]

This is an incorrect Internet myth:

https://www.lensrentals.com/blog/2019/10/more-ultra-high-resolution-mtf-experiments/

(scroll down). 

On the contrary, the higher resolving the sensor, the more lens performance is recorded. Thus any residual aberrations of an imperfect lens will show up at the extreme magnification of 100% But overall performance will improve.