Field curvature on 50mm 1.4 Summilux ASPH?

[IronLion]

Advertisement (gone after registration)

Hello,

 

I'm a new member of the M rangefinder club, and I recently acquired my first M body and a used 50mm Summilux. 

 

I was out on a walk today taking random photos, and when I was looking over the photos I took today, two of them stood out to me as a bit odd due to the presence of what looks to be what looks to be some fairly odd field curvature, at least to an extend that I wasn't expecting from a 50mm.

 

Below is a photo that I shot wide open at 1.4, with the lens focused on the stop sign, which was a just bit before infinity (I could tell because I hit the infinity stop, then backfocused a tiny bit to bring the stop sign back into critical focus). 

 

 

As you'll see the stop sign is in focus, but on the left hand side of the frame, the brick building that is across the street from this stop sign and on the left edge of the frame appears to almost be in full focus as well (if you zoom in, you can see it's not critically sharp, but it is definitely sharp enough to draw your attention there).  

 

Is this type of field curvature normal for the 50 Summilux at this focus distance? I'm guessing that the fact that it was wide open AND close to infinity made this happen, but I'm just wondering if other users of this lens have noticed this type of field curvature as well?

 

Frankly I find it pretty distracting once I noticed it- from a sharpness perspective, two totally separate parts of the image are competing for my attention (more than anything though, this is probably because this is a terrible photo that I took just to test the lens out).

 

 

[erl]

Something weird is appearing here. The car is closer than the building to the camera, yet it is out of focus.

I think the builing is OOF as well, but the little bumps on the foof may be creating an impression they are ar sharper than reality.

 

Is there a misaligned element in the lens? I guess you don't know.

 

Try shooting a brick wall, square on, at different apertures. Camera on tripod of course.

That will be a more definitive test.

[Advertisement]

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

[IronLion]

Something weird is appearing here. The car is closer than the building to the camera, yet it is out of focus.

I think the builing is OOF as well, but the little bumps on the foof may be creating an impression they are ar sharper than reality.

 

Is there a misaligned element in the lens? I guess you don't know.

 

Try shooting a brick wall, square on, at different apertures. Camera on tripod of course.

That will be a more definitive test.

 

Yeah, I don't know because I purchased the lens used! Something definitely looks weird though. 

 

What distance should I be from the brick wall for the test? 

[erl]

Try several different ranges.

Say, 5m, 15m, 30+m.

Should not make must difference though.

[pop]

The image exhibits several strange phenomena. 

 

Firstly, the distance of the traffic sign appears to be about 4m or 5m from the camera. That's nowhere close to infinity. Does the lens focus at infinity, i.e. can you take sharp pictures of a building or tree which is more than 10m distant?

 

Second, elements in the image have double contours; you can see those quite well when looking at the image at 100% zoom. The doubling of the contours appears to restrict itself to the upper half of the image and only close to the left and right hand edge of the frame: the twigs of the tree on the right hand side have it, so have the white painted parts of the buildings (door jambs, garden fence) on both sides. The offset of the double contours appears to occur in the vertical direction only.

 

The first thing I would check is whether the lens can take pictures of flat things perpendicular to its axis: take a shot of brick wall or a large sheet of graph paper. Are flipcharts still being used? Use a tripod or another firm support for the camera.

 

I fear the lens might be badly misaligned; however, drawing conclusions from but one image is not a good idea.

[microview]

Odd because the wall to the left looks like camera shake with double-imaging/blurring. Yet the street sign is OK and focus on the drilled metal post tails off slightly towards ground level as you would expect. Some green fringing too around the white painted windows etc.

[Shu]

Advertisement (gone after registration)

I'm the previous owner of the lens and searched through some photos close to the point of sale that offered similar composition. Michael, if there is an issue with the lens, we'll make it right! Appreciate the community weigh-in, guys.

 

In this photo, I think the background is too close to my wife and daughter to draw a conclusion:

 

The plane of focus is pretty straight here, but I'm close to the focal plane:

 

Now, things get a little interesting in this photo. I wish I had more foreground so we could see more of the focal plane, but we can see trace it from my daughter back towards the tree behind her. It's hard for me to tell if the focal plane is supposed to be diagonal because of my perspective or if it's an issue with the lens.

[IronLion]

Thanks so much Shu for posting, and to the others in the forum for their quick responses! I really really appreciate it so much. 

 

I'll see if I can do some more testing with the lens this week and will post images to this thread shortly thereafter. 

 

Pop, I think you're absolutely right- the stop sign wasn't as close to infinity as I recalled in my original post. I think I'm confusing this image with another similar one that I took, which also showed some weird behavior on the left-hand corner, that I will post here in a bit.   

[spydrxx]

In all honesty, the stop sign isn't razor sharp, given it is purportedly at the plane of focus. I would do some further testing of the lens as others recommended and be sure to use a tripod to eliminate one factor of unsharpness.

[01af]

At full aperture, there's significant vignetting which not only reduces illumination but also increases depth-of-field and decreases background blur at the frame's edges and corners. So the background at the frame's center, at or near full aperture, will always be more blurred than the same background near the frame's edges. This phenomenon often gets mistaken for field curvature. Of course, if there is field curvature then it will interfere with the vignetting's effect and, depending on the curvature's orientation, add to or subtract from it.

[uhoh7]

Just a nice long infinity shot is best. Brick walls and anything very close can mask what's really happening. The charts measure infinity.

 

M9_28cron_F28 by unoh7, on Flickr

Notice I have details in the corners which are closer, but still pretty far away. Ideally the corners are equidistant.

some like the sideways method:

 

V2_CV35/2.5_5.6 by unoh7, on Flickr

 

You want to take multiple apertures, WO 2.8 4 5.6 and 8, and do it again with focus backed off slightly. This method is fast and as good as any optical bench if you are careful and take multiple sets with focus variation.

 

Also you would take a set with lens correction on and off.

 

You can do the whole operation in 10 minutes shooting. Then you really will know. WO the 50 Lux has big mid-frame dip. You would be able to see this clearly in an infinity test shot. You can download the chart and compare directly.

 

These shots are on flickr and the fulls are easily seen. That way others can check them too. De-centered lenses are not uncommon, and that includes M lenses. Certainly worth checking.

 

Another fast check is the Zeiss Seimens Star chart:

https://www.lensrentals.com/blog/2012/05/testing-for-a-decentered-lens-an-old-technique-gets-a-makeover/

 

You can download a star chart and print as well: do google image search and find a big one. Many out there.

Edited March 27, 2017 by uhoh7

[tobey bilek]

There is also possibility sensor is not sitting properly/slanted.

 

Take some pics of the wall straight on and then turn camera over and see if things change.

 

Without a second lens,  it is difficult to tell what is out.

 

Also is this the new 50 1.4 ASPH or one of the older ones.  Post a pic.

 

I never could find an older on I liked, but my ASPH is really nice.  

 

ASPH elements sometimes cause abnormal depth of field as you focus , sometimes curving in, sometimes out, sometime partially in and then out.   If you focused and recomposed this can make for strange photos.

 

Just reread the title.   No not normal for new 50 `Lux.     Do the brick wall straight on, then at 45 deg making the focus point with tape or chalk.  Do not focus and recompose.  Then 45 deg the other way.

Edited March 27, 2017 by tobey bilek

[Fgcm]

Question for IronLion: I downloaded your file to see it better.

It's pretty weird.

5976 X 3984 pixels? 

Which software did you use convert the DNG?

I see no Exif data.

Franco

[lct]

All f/1.4 lenses i've used had field curvature at full aperture so far. It is more visible on sharper lenses though. I have no idea what role FLE can play but both my 50/1.4 asph and 35/1.4 FLE are affected to various degrees. 

[IronLion]

Question for IronLion: I downloaded your file to see it better.

It's pretty weird.

5976 X 3984 pixels? 

Which software did you use convert the DNG?

I see no Exif data.

Franco

 

Hey Franco, I use Adobe Photoshop to open my DNG's and to tone them, then I just save them as a JPEG. 

[Guest]

All f/1.4 lenses i've used had field curvature at full aperture so far. It is more visible on sharper lenses though. I have no idea what role FLE can play but both my 50/1.4 asph and 35/1.4 FLE are affected to various degrees.

 

May I ask: convexity or concavity? Photo examples at far distances and infinity with both lenses would be sincerely appreciated.

[pgk]

It is possible that the apparently 'sharper' areas background might just be as a result a phenomenon know as 'spurious' resolution, which can apparently afflict highly corrected lenses (one definition found online is:  "Optical phenomenon in which a highly corrected objective that is slightly out of focus appears to resolve detail, particular where the detail is high contrast". Just a thought  .

[jaapv]

Hey Franco, I use Adobe Photoshop to open my DNG's and to tone them, then I just save them as a JPEG.

 

if you choose "save for the web" EXIF will be stripped.

[lct]

May I ask: convexity or concavity? Photo examples at far distances and infinity with both lenses would be sincerely appreciated.

 

No idea sorry but field curvature at fast apertures is pretty common IMO. Perhaps a bit of it on the statuette at f/2 below:

50/1.4 asph: http://lctphot.smugmug.com/photos/3878259694_cZN3DN2-D.jpg

50/2 apo: http://lctphot.smugmug.com/photos/3878262917_vFtxJSS-D.jpg

(M240, tripod, self timer, focus on "IX" with EVF and 10x magnification, 11 MB files)

 

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: Leica M

[Fgcm]

Sorry, but in my opinion this is not a matter of field curvature.

 

This is a crop of a corner of the photo posted by IronLion:

 

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 left corner is as bad as the right one. 

 

This is another crop from the same photo:

 

 

Double Images in the background with a Summilux 50 Asph? Really?

 

 

IMO this is too weird to be normal.

Franco