35 APO-M: Grainy bokeh issue?

[onasj]

Advertisement (gone after registration)

Received my 35 APO-M.  It's an optical masterpiece in many ways, but one potential issue I noticed is that point sources of light when out of focus appear as grainy bokeh.  I compared with my 50 APO, and my conclusions are:

- My 35 APO has much grainier bokeh from point sources of light than the 50 APO, with a semi-random pattern resembling stucco walls.

- My 50 APO has more of an onion ring effect and more color fringing in the bokeh from the same point sources of light, but is overall smoother than the bokeh from the 35 APO.

This might only be an issue with point sources of light, but then again, those are common sources of bokeh and common picture elements.

See the attached crops, which are 100% crops of bokeh from the same point sources of light taken with the 35 APO at f/2 and the 50 APO at f/2.

Original DNG files:

https://www.dropbox.com/s/d1gw3yaueqpdn9p/Leica 35 APO-M bokeh.DNG?dl=0

https://www.dropbox.com/s/xa0xy7k31x08lwn/Leica 50 APO-M bokeh.DNG?dl=0

100% crop of 35 APO-M bokeh:

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!

100% crop of 50 APO-M bokeh:

Have any other 35 APO-M users noticed this issue?

Importantly, I don't notice graininess in the bokeh of other out-of-focus elements—only from point sources of light.  Indeed, the bokeh of regular out-of-focus areas seems fairly smooth and blends in even at the edges with other picture elements.

 

Edited March 8, 2021 by onasj

[Lelmer]

Brownish background is also different. Same ISO? 

[onasj]

1 minute ago, Lelmer said:

Brownish background is also different. Same ISO? 

You can check out the DNG files.  Same subject (a black ceiling with fiber optic lights).  ISO may have been a bit different since I shot these with auto-ISO, but nothing hugely different and certainly nothing approaching the point at which the M10-R shows lots of noise (>ISO 10,000+).  The bokeh grain definitely isn't coming from sensor noise, as some of the elements within the bokeh are reproduced between different bokeh balls.

[sebben]

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 grainyness is coming from the attempts to polish away the onion rings. It looks like this has not been perfected yet. The Pre-Asph lenses don't seem to have onion rings.

[BastianK]

The 2/50 has one aspherical surface whereas the 2/35 has four.
So instead of one pattern of rather distinct onion rings (2/50) you see four slightly different patterns of that kind overlap
and leading to a "messier" (or grainier) pattern (2/35).

[colint544]

Yes, it looks very much like the 'onion ring' ring effect. As someone above mentions, you don't get it with the spherical type lens elements, or with the early aspherical designs, which were hand ground at great cost. What you're seeing there is the fine machine pattern on the surface of the tool which pressed the blank disc of glass into the aspherical shape of the element(s) 

You probably won't see it often. My 50mm Summilux ASPH does it too. I used to find it irritating, but have just come to accept it. They're not hand grinding lens elements any more.

I did read some time ago, that Panasonic were working on a technique to banish the 'onion ring' effect, but don't know if it ever came to anything. If it did, seems they didn't pass the secret on to Leica.

Here's what it looks like on my 50mm Summilux ASPH.

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 March 9, 2021 by colint544

[onasj]

Advertisement (gone after registration)

Yes, I agree that the superimposition of four sets of microscopic ridges left in the four aspherical surfaces in the 35 APO is causing the issue, as nicely explained here: https://www.imaging-resource.com/news/2014/05/02/the-end-of-onion-ring-bokeh-panasonic-beats-the-curse-of-aspheric-lenses

I’m hoping another 35 APO-M user can confirm that they also see grainy bokeh from out-of-focus point sources of light so we can know if it is my particular copy or if this is simply an inherent aspect of the lens.

[colint544]

Apologies if I overexplained something you already knew there!

Was not entirely evident to me in your original and follow-up question.

Best wishes,

C

[tightsqueez]

Ah, the dreaded onion ring effect. I’ve noticed this a lot on the new 35 APO image examples. Too bad. 
 

Has anyone noticed this on the newly released 1.2 Noct? I’ve seen it quite bad on the original. 

[onasj]

1 hour ago, colint544 said:

Apologies if I overexplained something you already knew there!

Was not entirely evident to me in your original and follow-up question.

Best wishes,

C

No worries at all, no offense taken, and no need to apologize!  

@Steven, have you seen this kind of "stucco ceiling" bokeh (aka grainy balls :D) in your 35 APO from out-of-focus point sources of light?  I see them when I focus well in front of the point sources of light.

[onasj]

One thing that's odd about these 35 APO grainy bokeh balls is that the pattern of grain within each ball differs much more than the pattern of onion rings within the 50 APO's bokeh balls.  See my crop above and the original DNG file for many examples.  The 50 APO makes highly reproducible onion ring patterns in each ball, while the 35 APO makes semi-random bokeh grain that share the same general form, but not fine details within each.  Makes me wonder if interference from some wavelengths of light is contributing to the issue?  The point sources of light were generated by an LED illuminator I made long ago that uses LEDs of several different colors (cool white, warm white, orange, red, and blue), with each fiber optic getting a slightly different color of light.

Edited March 9, 2021 by onasj

[tightsqueez]

7 minutes ago, onasj said:

One thing that's odd about these 35 APO grainy bokeh balls is that the pattern of grain within each ball differs much more than the pattern of onion rings within the 50 APO's bokeh balls.  See my crop above and the original DNG file for many examples.  The 50 APO makes highly reproducible onion ring patterns in each ball, while the 35 APO makes semi-random bokeh grain that share the same general form, but not fine details within each.  Makes me wonder if interference from some wavelengths of light is contributing to the issue?  The point sources of light were generated by an LED illuminator I made long ago that uses LEDs of several different colors (cool white, warm white, orange, red, and blue), with each fiber optic getting a slightly different color of light.

It does differ however the 35 is noticeable, the 50 is not. That’s what matters. 

[Guest JMF]

27 minutes ago, onasj said:

One thing that's odd about these 35 APO grainy bokeh balls is that the pattern of grain within each ball differs much more than the pattern of onion rings within the 50 APO's bokeh balls.  See my crop above and the original DNG file for many examples.  The 50 APO makes highly reproducible onion ring patterns in each ball, while the 35 APO makes semi-random bokeh grain that share the same general form, but not fine details within each.  Makes me wonder if interference from some wavelengths of light is contributing to the issue?  The point sources of light were generated by an LED illuminator I made long ago that uses LEDs of several different colors (cool white, warm white, orange, red, and blue), with each fiber optic getting a slightly different color of light.

Here is a photo taken with 50 APO Summicron on SL :

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!

Tocopilla, Chile by JM__, on Flickr

 

[Guest JMF]

Taken with the 35 Summilux AA on SL:

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!

en Tocopilla by JM__, on Flickr

Best, JM.

[tgray]

5 hours ago, colint544 said:

I did read some time ago, that Panasonic were working on a technique to banish the 'onion ring' effect, but don't know if it ever came to anything. If it did, seems they didn't pass the secret on to Leica.

I believe the Sony GM lenses have made a lot of progress in this area. Also seems like many newer Voigtlander Aspherical lenses seem to fair better than these examples. 

[epand56]

That's the most expensive onions I've ever seen in my life.

[cboy]

On 3/9/2021 at 1:54 AM, onasj said:

100% crop of 35 APO-M bokeh:

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!

100% crop of 50 APO-M bokeh:

 

Some nasty grainy bokeh there. Im sure it can be overlooked given its seem to be a fantastic lens otherwise.

All the best

[darkinners]

M10-R with APO 50 Summicron, 50% zoom. For your reference. 

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!

[haikos]

Would these effects show up on prints?

[Advertisement]

Hallo,

Look here: Leica M

[darkinners]

7 minutes ago, haikos said:

Would these effects show up on prints?

They will but whether you can see it, totally depends on how large you print and viewing distance.