M11 Motion blur is driving me crazy(!)

[lct]

Advertisement (gone after registration)

1 hour ago, lincoln_m said:

[...]The DOF charts for Leica lenses still use the old 30um circle of confusion used in film days (3M pixel scans) but if we use the more realistic 4um (because we are pixel peeping, then at 4m focus at F1.4 we find the DOF is now 4m +/- 7cm. But as mentioned earlier the object is actually 3.6m away which is infant of the 3.93m start of the DOF region. Even with an 8um circle of confusion the DOF is still only 4m +/- 15cm, object is still out of focus.[...]

The OP said he used the EVF too so i wonder if those DoF calculations are relevant to his problem. BTW with your suggested 0.004mm CoC value, most RF focusings would be inaccurate on current M cameras if i understand well how critical RF base lengths are calculated. May i ask where you found this 0.004mm CoC value or how you calculated it? Just curious.

[Advertisement]

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

[lincoln_m]

The pixel pitch for 60M pixel sensor is about 4um so I used that. But maybe 8um (2 pixels diameter circle) is more accurate? 24M pixels have 6um pitch. If we are zoomed in to 100% , one monitor pixel equals one sensor pixel then we are expecting to see sharpness at that 100% zoom. Something that was only sharp at 30um (6-7 pixels) wouldn’t look sharp when zoomed in at 100% on a 60M pixel image.

Even if the EVF was used I think I remember the OP saying it was then reframed to take the shot (camera tilted up) so the focus distance was now too long. It would be ok if the camera wasn’t reframed and the object was still in the centre rangefinder patch. But yes Evf is more accurate than RF.

[CDodkin]

11 hours ago, lincoln_m said:

I've done some experiments with my 35mm summilux ASPH FLE and my M10. I get the same out of focus issue when I focus on an object on the ground at ~4m away then reframe the shot with the object in the bottom of the frame. If the original focus distance 4m is the hypotenuse and my eye is approx 2m above the ground (yeah I wish) then the object is actually 3.6m from my feet. The lens is still set to 4m when the photo is taken but the object is now too far forward and out of the depth of field.

 

 

OP stated he used the EVF, so it's not a focus/recomposing issue

[lct]

I guess it could be a focus/recomposing issue if the OP did reframe to take the shot, instead of moving the focus point as the M11 allows to. The out-of-focus issue would have been visible though if the OP did use the focus magnification feature of the camera. As for the CoC value, even 0.008mm seems unrealistic when calculating the critical RF base length with the formula b = e*f^2/k*z where b is the critical RF base length, e the visual acuity, f the focal length, k the aperture and z the CoC value. A good colleague here used to suggest 0.015mm if i remember well but i can't seem to recall his name and even then, the critical base length would be too high according to the said formula IMHO.

Edited May 29, 2023 by lct

[wijsbroek]

16 hours ago, lct said:

I guess it could be a focus/recomposing issue if the OP did reframe to take the shot, instead of moving the focus point as the M11 allows to. The out-of-focus issue would have been visible though if the OP did use the focus magnification feature of the camera. As for the CoC value, even 0.008mm seems unrealistic when calculating the critical RF base length with the formula b = e*f^2/k*z where b is the critical RF base length, e the visual acuity, f the focal length, k the aperture and z the CoC value. A good colleague here used to suggest 0.015mm if i remember well but i can't seem to recall his name and even then, the critical base length would be too high according to the said formula IMHO.

No I didn't recompose...

[Nitnaros]

On 5/23/2023 at 8:32 AM, 35photo said:

It’s either a focus issue or a calibration issue… 

It’s not a hand holding or shutter speed issue, if I would need to guess 

I had no blur issues when shooting with the Fuji GFX 50R - 50 megapixels, no IBIS. But of course auto-focus. So what it tells me is that we can hand hold a 50 megapixel camera and get sharp images.

my speculation is that it relates to some kind of focusing issue 

best - Peter 

[lct]

Advertisement (gone after registration)

2 hours ago, wijsbroek said:

No I didn't recompose..

But you were using your Visoflex, weren't you? In such a case, you did not move the focus point and/or you did not use focus magnification. Besides user error, i cannot figure out any other logical explanation.

Edited May 30, 2023 by lct

[jaapv]

Or were you using focus peaking? That can be quite imprecise. 

[wijsbroek]

9 hours ago, lct said:

But you were using your Visoflex, weren't you? In such a case, you did not move the focus point and/or you did not use focus magnification. Besides user error, i cannot figure out any other logical explanation.

I did move the focus point and focus magnification (10x) with the Visoflex.

My conclusion is... this is only happing at f/1.4. I guess I just always have to stop down a little bit. I have had some great results with the M11/35 summilux and I am absolutely happy with it. Just not at f/1.4.
 

[jaapv]

All lenses perform slightly less at the widest aperture - even Leica ones (albeit minimally). I would suggest taking and printing a few photos with your Summilux wide open. You'll see that the problem is pixel-peeping a 60 MP camera, not the lens or the camera themselves.

[lct]

•

Edited May 30, 2023 by lct

[algrove]

OP, kindly make sure you have adjusted the diopter to your eyesight. You should turn it until you get the best focus on the words in the EVF, not the scene. Kindly confirm.

Edited May 30, 2023 by algrove

[wijsbroek]

39 minutes ago, algrove said:

OP, kindly make sure you have adjusted the diopter to your eyesight. You should turn it until you get the best focus on the words in the EVF, not the scene. Kindly confirm.

Yes, already done that. Thank you 🙂

[Photoworks]

3 hours ago, wijsbroek said:

I did move the focus point and focus magnification (10x) with the Visoflex.

My conclusion is... this is only happing at f/1.4. I guess I just always have to stop down a little bit. I have had some great results with the M11/35 summilux and I am absolutely happy with it. Just not at f/1.4.
 

Did you ever do a lens test for flat focus area?

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!

as you can see in this image the focus does not across linear , but makes a little curve and and fall of to the corners.

this is a normal behavior lens with this character. Leica implemented more of a waveform of lens design to give you the 3D pop in many lenses.
This would be just to see how your lens render.

If I look at my Noctilux, the images are very sharp in the center, with quick DOF falloff, and in focus again near the borders.

You can read more about it here.

https://www.lensrentals.com/blog/2016/09/fun-with-field-of-focus-part-1/

 

[algrove]

OP, have you tired another lens?

[lincoln_m]

Some more M10 plus 35 summilux ash fle tests on 3D scenes first at F1.4 than at F5.6 focus is the white dot at the back of the fire pit 1.5m away according to the lens marking. Rangefinder focused. I don't have EVF and the Live View doesn't allow a zoom-in (the Sony A7 allows for a zoom-in for focus). Also I tried focus peaking in Live View which like the Sony is also very inaccurate based on contrast I think. So I stick with RF focusing which also isn't perfect but as I'm not doing model portraiture so it's close enough.

I think my LightRoom does automatic lens corrections for CA vignette and distortion, but I never really know if the M10 adjusts the DNG based on the lens coding/setting in camera and then LR does it all again.

The focal plane does seem to be flat as far as I can tell.

@F1.4 1/3000th, first image, I think the DoF seems to be at most +/- 10cm = 2 table slats ( each slat is 5cm) but might be +/- 7cm that the Leica data sheet (DoF=1.425 -> 1.584m at F1.4 and 1.5m focus) suggest you get at 1.5m. The fire poking stick just happened to be in the foreground in front  the DoF and we see the Bokeh (or is Bokeh only background out of focus?) out of focus effects which actually do look similar to motion blur. The leaves in the background also look to have motion blur from the bokeh, but it can't be at 1/3000th.

My impression is this 35mm F1,4 ASPH FLE lens sometimes gives the appearance of motion blur @F1.4 for out of focus areas both foreground and background particularly for naturally curved objects.

 

 

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!

[lincoln_m]

@F5.6 1/180th ( don't believe the M10's aperture guess in metadata as it's often wrong +/- 1 stop). Now we have about 8-9 slats in focus 45-50cm. Both were handheld so I could have moved my head back a little between shots. 1.5m +/- 0.25 DoF is similar to Leica's DoF chart @5.6 1.247 -> 1.889m when you're not using a tripod and tape measures.

There are no artefacts on the stick now.

That's my unscientific review of the 35 FLE at close focus.

The data sheet MTF are for long distances (infinity) which is why I use this lens at F5.6 at infinity for very detailed landscape shots, well I'm happy with it. It will take the rest of my life to understand (imagine in my mind's eye) the behaviour of the out of focus regions of close focus (<5m) images at wide apertures.

I encourage Wijsbroek to also experiment with 3D scenes to understand how objects in the foreground's out of focus region appear in the image. I'm thinking of say a group of people at a party, wedding, dance floor for instance where the subject would be in focus but other people in the foreground would be blurred and also appear bigger (closer = bigger) in the image if you were using this lens at F1.4. to give a 3D effect but we have to be careful that the foreground blur isn't too distracting. The fun of photography I guess?

 

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!

[Roderikvn]

There is definitely a ‘issue’ with the unsharpness caused by movement.

This image with a 28mm and a shutterspeed of 180 is sharp in most of the image except where the people walk. They are not sharp. Very frustrating but probably solved by using a lower resolution…

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!

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Stevejack]

1 hour ago, Roderikvn said:

There is definitely a ‘issue’ with the unsharpness caused by movement.

This image with a 28mm and a shutterspeed of 180 is sharp in most of the image except where the people walk. They are not sharp. Very frustrating but probably solved by using a lower resolution…

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'm not sure if I'm understanding you correctly, but if it's the motion blur you're not wanting in the people who are walking then you just need a higher shutter speed to freeze them. I like the slight blur in moving objects but if that's not what you want - up your shutter to 1/250 or above.

[lct]

I like your photo as is but 1/180s is not fast enough a shutter speed if you want to avoid motion blur on moving subject matters.