APO Summicron 50/2 ASPH: Central veiling flare / fogging

[adan]

Advertisement (gone after registration)

The image in a camera is recorded upside down...

 

Yes, Erik. I know.

 

See diagram:

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]

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

[Erik Gunst Lund]

That is a rather simplified version of an photographic lens...

 

Top / bottom still applies.... The bottom of the image would have veiling flare

 

I think this doesn't apply to 'normal' photographic lenses and cameras like Leica M

 

Would you be able to make a similar trace of sunlight on a sectional view of a Leica M camera with lens attached, thanks!

 

Where did you find that illustration?

[adan]

I'm not sure what you are debating, Erik.

 

My diagram (I didn't "find" it anywhere - it is my own drawing, based on 40 years experience and knowledge, and an advanced degree in photographic arts and technology) is completely consistent with what you said - that photographic images are projected upside down in the camera.

 

It doesn't matter if the "lens" is a simple single element (used here for graphic simplicity), or a multi-element Summmicron, or a pinhole, for that matter. It will always project the world upside down, so if the sun is "up" and outside the frame in the real world, it will be "down" and outside the frame (i.e. on the floor of the camera chamber, unless blocked by a lens hood) - in the projected image.

 

I hope we at least agree on that.

 

If you go back and look at helged's 9 images spanning from the bottom of page 6 of this thread to the top of page 7: you will see that they all have veiling flare to one degree or another, BUT that there is a band across the top of each - about 8% of the image - that is mostly protected from flare. The tree branches are fairly dark and unfogged.

 

As you say - "The bottom of the image would have veiling flare" - or at least the bottom 95%.

 

Again, I hope we agree that in those specific pictures of helged's in posts 119-121, there is generally veiling flare, except for a narrow band with no veiling flare right at the top of each picture.

 

That protected band has a sharp edge where the flare begins. It corresponds to the shadow in my diagram marked "e".

 

The protected band occurs at the TOP of the pictures because the shadow responsible for it was cast at the BOTTOM of the sensor (again conforming both to your own statement, and to my diagram).

 

If helged agrees, I will borrow one of his pictures and combine it with my diagram.

 

Anyway, this is off-topic to the original problem of the central veiling spot some have found occurs with the 50 ASMA at small apertures, which is a whole different question.

[Erik Gunst Lund]

I agree with you on the veiling flare, and the way you have explained it now.

 

I think something was 'lost in translation' regarding top and bottom... Sorry for that!

 

I was not patronizing you image, just interestet in where it came from since I have never seen it before and I'm just interested in seeing your drawing of what happens, in a much more detailed drawing to scale of the Leica M with a lens.

 

This could lead to an answer for the central veiling/loss of contrast... IMHO

[Nicolas Soldati]

Thank you Adam for your interesting draft. It really helped me to easily understand the issue. Thank you again for your helpful post!

[Bill W]

Thanks Andy for the explanation and diagram. I think I see some spots on you sensor though

[lct]

Advertisement (gone after registration)

...It is my considered opinion (to which the 50 AA situation adds another data point) that these concave rear elements - whatever other optical magic they may perform - are inappropriate for use in front of a shiny digital sensor. I believe they are prone to reflecting back to the sensor (somewhat refocused, rather than dispersed, by the concave surface) bright areas of the subject first reflected by the sensor surface itself...

Interesting indeed. Do you know if central veiling flare has been reported out of other lenses with concave rear element as well? (35/1.4 asph FLE, 35/2 asph, 50/1.4 asph, 90/4 macro, 135/2.8, 135/4 IINW). The rear element of the tele lenses is too far from the sensor i guess but i have never noticed this particular flare out my 35/1.4 FLE, 35/2 asph, 50/1.4 asph and 75/2 asph so far. My experience is far from being at the level of yours though.

[Paul J]

It is my considered opinion (to which the 50 AA situation adds another data point) that these concave rear elements - whatever other optical magic they may perform - are inappropriate for use in front of a shiny digital sensor. I believe they are prone to reflecting back to the sensor (somewhat refocused, rather than dispersed, by the concave surface) bright areas of the subject first reflected by the sensor surface itself..

 

This is an exceptionally well considered opinion and I am betting you are right.

[01af]

I have never noticed this particular flare out my 35/1.4 Asph, 35/2 Asph, 50/1.4 Asph and 75/2 Asph so far.

The Apo-Summicron-M 75 mm 1:2 Asph is very prone to this particular kind of central veiling flare (CVF) at medium and small apertures and short subject distance—at f/16 & f/11: CVF at less than 2 m/6 ft; at f/8: less than 1.5 m/5 ft; at f/5.6: less than 1.2 m/4 ft; at f/4: less than 0.9 m/3 ft; at f/2.8 & f/2: virtually no CVF at any distance.

 

With the other lenses I don't find this CVF issue. So I guess Andy adan is wrong with his hypothesis about concave rear elements.

[lct]

I leave it to you as i rarely do closeups with the 75/2. Thanks for sharing.

[Rick]

I have seen central veiling on all of my Leica M lenses. I first became aware if it in 2010 with the 28 Summicron shooting in the streets in Vernazza on my M9. It has been an annoying random occurrence with several lenses since. I thought it might have to do with fog on the rear element or maybe the in-camera lens correction was not applied for some unknown reason. I was wrong.

 

It wasn't until I got my M and utilized the EVF that I was able to clearly identify and reproduce the effect of central veiling. Looking through the EVF and shooting across a dark room at a window I can reproduce the effect with all of my Leica M lenses. The EVF allows me to set up the condition that elicits this effect.

 

The condition is; a strong central light area with surrounding area of lower light level. This high general contrast situation makes it easy to see any amount of loss of contrast caused by this type of optical central light scatter. (Other shooting conditions elicit this effect to lower degrees as seen in examples posted here on this thread by members)

 

I believe Andy is very much correct that it is caused by the sensor reflecting light to the rear of the lens. The rear lens optics reflect a diffuse but centrally focused pattern back onto the sensor. This happens to some degree with all lenses I've tested.

 

One example is the 28 Summicron. Its rear element is convex . Yet it still produces this effect. But, if you look at the rear elements as a whole you will see that the rear element is convex, yes, but, the second element to the back is strongly concave allowing for the light to pass through the back convex element and reflect on the second concave element. I assume it is this second concave element and its distance from the sensor that focuses a diffuse(unfocused) reflection of the sensor back onto the sensor. Some ray tracing is in order, but I don't know the distances or the powers of these surfaces to really know if this is the path of the reflected light.

 

In the case of the 28 Summicron, it is as Andy postulates, but it is the second element that probably reflects the light back to the sensor.

 

Lenses that are more flare resistant and that exhibit higher overall contrast are easer to identify this characteristic. In other words, the better your lens, the easier it is to find the loss of central contrast due to reflections off of digital sensors. Once you are able to identify this effect it becomes easy to recreate this effect with stronger amounts of central loss of contrast. Obviously other characteristics, such as, element curvature would also paly a role as well.

 

The 50 APO Summicron would be the perfect high contrast lens to use to identify the effect of reflections off the sensor onto the rear elements of lens. Maybe, this lens is too perfect for its own good?

 

I am sure that some lens designs lend themselves to this effect more than others dependent on the curvature, placement and distance of the offending element surface to the sensor. Maybe, this also plays a role in the 50 APO effect, but again, I see it strongly on all my Leica lenses, including but not limited to the 35/1.4, 50/1.4, 50/2.0 old.

 

(It would be interesting to see if this effect is greater when lens recognition is turned off. In other words, does Leica know about this effect and try to correct for it. I assume that if they do it could not be correct in all situations with equal success.)

[lct]

Would you have some pics to show this flare Rick?

[Rick]

Would you have some pics to show this flare Rick?

 

You have the M and EVF. Sit on your couch, look through your own EVF and shoot towards a window across the room in a somewhat darkened room yourself. It would nice if others could report how easy it is to elicit this effect on all M lenses. The EVF just makes it easy to swing the camera around and find the perfect situation. You can see this easily in the EVF without even shooting the picture.

 

So far, I wouldn't be concerned with the effects shown here with the 50 APO.

 

I think I deleted all of these photos, but I will reproduce them with a dozen different lenses for you soon.

[lct]

Are you sure what you see is not flare due to overexposed parts of your images? Anyway it's too late to check this out here but i will do the test when time allows.

[01af]

I believe Andy is very much correct that it is caused by the sensor reflecting light to the rear of the lens.

Except this wasn't Andy's point.

 

 

The condition is; a strong central light area with surrounding area of lower light level. This high general contrast situation makes it easy to see any amount of loss of contrast caused by this type of optical central light scatter.

Umm ... now I'm under the impression that you have no idea what we're talking about. Or I have no idea what you're talking about.

 

 

Would you have some pics to show this flare, Rick?

You have the M and EVF. Sit on your couch, look through your own EVF and shoot towards a window across the room in a somewhat darkened room yourself. It would nice if others could report how easy it is to elicit this effect on all M lenses.

Rick, do us a favour and post a sample picture. Please.

[adan]

The Apo-Summicron-M 75 mm 1:2 Asph is very prone to this particular kind of central veiling flare (CVF) at medium and small apertures and short subject distance......

 

With the other lenses I don't find this CVF issue. So I guess Andy adan is wrong with his hypothesis about concave rear elements.

 

Well, I also included an example from the 35 Summilux ASPH. And we have the examples (not mine!) from the 50 ASMA. But I will certainly accept that my hypothesis is incomplete.

 

There are a lot of variables:

 

- the "focal length" of the mirror lens formed by a concave rear element. lct mentioned the 135 f/4 TE, which has a very slight concavity. The modern ASPH lenses that includes the signature Solms concave rear element usually have a more pronounced curve.

 

- the presence or absence of other sources of flare (from lens barrel, lens elements internally, filters, inside of the camera itself as diagrammed above)

 

- all rear elements, concave or convex, are, of course, momentarily "flat" - parallel to the sensor or film - at their exact center.

 

So there is always the potential for sensor-rear-element-sensor reflections, with any lens. The strongly concave rear elements may be more prone to (but not alone in) revealing this because: a) the curve focuses** the light to the center more than a plane or convex rear lens surface, and because these ASPH lenses are generally corrected so well for high contrast and control of the other possible flare sources, allowing the reflection to shine through more obviously.

 

**I note in reviewing my own 75 Summicron picture posted originally, that the flare spot is actually red-tinged at the top. It is thus even more clearly a blurred reflection-reflection of the white chicken AND its red coxcomb.

 

BTW - RickLeica - I also got a central flare spot occasionally with the 28 Summicron. But on the M8, before IR/UV filters became available, so I put it down to the "IR hot spotting" that was being reported around that time.

 

http://www.l-camera-forum.com/leica-forum/leica-m8-forum/32744-lenses-ir-hotspots.html

 

It is always possible that factors interact. If "A causes B" is a true statement - it does NOT therefore follow that "ONLY A causes B" or that "A causes ONLY B" or that "A ALWAYS causes B." The world is just not that simplistic.

 

My 135 TE (concave rear element) has never produced a center flare spot, that I am aware of. But since it produces other type of veiling flare sometimes, I just may never have seen the center flare spot even if it is there. it is masked by the "other stuff."

 

In the "chicken shoot" with my 75 sampled above, there are other frames, same aperture, same distance, that do not show the same flare. Same for the family shot with the 35 ASPH. A slight change in composition and positioning of the bright elements changed the amount of flare.

 

The acid test would be to try some of these "known offender" lenses - with identical light, subject, composition - on a film camera. If the central hotspots consistently go away when the sensor goes away, we can be more confident it is a sensor reflection effect. If they still show up, then I am probably wrong.

[01af]

Well, I also included an example from the Summilux-M 35 Asph.

That picture has some veiling flare but no central veiling flare. So that's a different thing, and most likely (I'm tempted to say: certainly) not caused by the rear element being concave.

 

The central veiling flare effect sure is a mirror effect between the lens and the sensor, and it hardly ever (I'm tempted to say: never) will occur on film because a film's surface is always matte and a sensor's surface is always glossy. If it occurs then it will be in a digital camera. And the rear element's shape has nothing to do with it.

[adan]

And the rear element's shape has nothing to do with it.

 

I hear you talking. I don't see evidence to back up your opinion. I do see fairly consistent evidence that concave rear elements produce these flares more often than other shapes.

[Advertisement]

Hallo,

Look here: Leica M

[Rick]

Except this wasn't Andy's point.

 

 

 

Umm ... now I'm under the impression that you have no idea what we're talking about. Or I have no idea what you're talking about.

 

 

 

Rick, do us a favour and post a sample picture. Please.

 

Leave it to you to start making it personal. Maybe, I do know what I'm talking about, but let's remain civil. This is a good thread, don't ruin it. I believe Andy is talking about light reflected off the sensor(look at a digital sensor, it looks like a mirror) onto the rear element and then back from the rear element onto the sensor. If, this is not what Andy is talking about, please explain your interpritation of what he said. I'm all ears.

 

I've already stated I'll look through my pictures and post, but you got to give me time to get off work and make it home.

 

Andy, thanks for the comment about the 28 Summicron. This is the lens I've spotted it from more than any other.

[Rick]

That picture (Andy's 35mm of family) has some veiling flare but no central veiling flare. So that's a different thing, and most likely (I'm tempted to say: certainly) not caused by the rear element being concave.

 

I agree.