What is this green dot matrix next to sunbursts? a7II

[TierraCosmos]

I just tested my new a7II and a new Zeiss 18mm Batis lens outside, and some of the shots had the Sun in the frame. In every one there was a very distinct  "matrix" of bright green dots visible right next to the sunburst. I doesn't look like typical lens flare but rather is a symmetrical grid of bright, evenly spaced points. It ruined every picture with the Sun in it; I can't remove the artifact decently in Photoshop. Picture is attached. Anyone have any ideas what's causing this artifact? I am guessing it's the camera and not the lens, but I could be wrong. 

 

Steve

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!

[Colorado]

I really don't have a solid idea to what caused this but I am interested. What were your settings and did you try to vary them? Nice shot by the way.

[LiveShots]

Do you have any filter attached to the lens ?

[TierraCosmos]

Do you have any filter attached to the lens ?

No filter.

[TierraCosmos]

I really don't have a solid idea to what caused this but I am interested. What were your settings and did you try to vary them? Nice shot by the way.

Thanks. Settings varied as I took a bunch of exposures, but for this image they were 18mm, f/18, ISO 50, 1/160 sec., manual mode, Raw. Single-shot AF, No IBNR, center metering, DRO/Auto HDR Off, most everything else Off.

[Guest all8]

I've seen this before, I think its reflections from the sensor because the pattern is particularly regular.

 

But at the same time, I've not seen it for a while ... but then its been a bit cloudy. It might be possible to use a point source light to try to induce the problem and better understand what it causing it.

[TierraCosmos]

One thing I'll try is taking the same shot with the same lens but different camera bodies (I also have an a7S and Canon 6D), and see if I get the same artifact. Not that that solves anything, but at least will give me some more info. Will also try same shot with other lenses on the a7II. At least all this will help to isolate the problem.

[Dollanganger]

No filter.

I'd strongly advise a circular polarizer for any outdoor shots with strong daylight. A photo directly of the sun with a snow white background is an extreme situation.

 

Also, try shooting darker and then brightening in Photoshop. 

[Guest all8]

Google a bit and you find this is a common occurance. You might be able to mitigate it with a larger aperture and/or a slight underexposure, then fixing the exposure in post processing.

 

The more simple (single dot) sensor reflections I have encountered can often be avoided with slight changes to camera angle. And some other dots which occurs at the same time I tracked down to the UV filter, most likely reflections from the sensor to the filter inner surface and then reflected again into the sensor.

[TierraCosmos]

I just tried a few things. I tried shooting darker (shorter exposures), but the pattern was still there, and came out bold as ever when I upped the exposure in post. I then swapped the same lens onto my a7S camera body and the grid pattern was gone. There were a few faint lens flare blobs on some of the images (I took a bunch of different exposure lengths), but nothing like the distinct grid pattern on the a7II. Very interesting!

 

Then I put the lens back on the a7II and shot at different apertures, having read online that the problem is worse at smaller apertures. The was inconclusive because some thin clouds appeared behind the Sun, which seemed to mask the dots somewhat. That is,even at my original settings of f/20 and 1/160s the dots were more muted, harder to make out vs. a plain blue sky. But at wider apertures the effect did seem less evident, even when I shortened the exposure so the Sun was not super blown out.

 

Lastly, someone online also said the dots went away when he switched from AF to MF. I tried that but, again, the clouds made it inconclusive. It may have lessened the artifact, but I'll have to test again (tomorrow hopefully) to see for sure.

 

Not sure I want to keep the a7II if the dot grid is something I have to live with when shooting wide-field shots of sunrises/sunsets and sunbursts. I suspect it's not a matter of individual copy variation; that is, if I were to exchange this a7II body for another one, I'd expect to see the same thing.   

[Username]

I'd strongly advise a circular polarizer for any outdoor

shots with strong daylight. 

    

There is no need for a Circular polarizer using  

a mirrorless live view camera such as the a7x

and a6xxx or Nex series. All camera functions

are unaffected/normal. 

[TierraCosmos]

I'd strongly advise a circular polarizer for any outdoor shots with strong daylight. A photo directly of the sun with a snow white background is an extreme situation.

 

Also, try shooting darker and then brightening in Photoshop. 

I thought circular polarizers were not effective when looking into the Sun. Ideally you want 90 degrees, right? Also with wide lenses like an 18mm, polarizers can be problematic, causing non-uniform brightness gradations.

[Guest all8]

Just tried it now with an A7m2, using a LED torch, and the pattern is there for Batis 18 & 25 as well as Loxia 50. So I would think its a problem caused by the sensor. These kind of things should be picked up by manufactures, its not exactly hard to test for acceptable behaviour of lens flare and sensor reflections.

 

An ND filter might help, that would reduce the intensity of the light entering the camera and possibly reduce/stop the reflection or the circumstance which causes it.

[addicted2light]

Timde is right, this is sensor reflection (well, actually these should be the microlenses). It happens quite frequently with several Fuji bodies (never, at least to me, with the A7r), so I've seen it a few times.

 

Usually all it takes to eliminate it is *very slightly* changing position (we're talking a couple of millimeters).

 

Using a tripod with a good head helps, because you can more easily refine the position of the camera watching for the disappearance of the reflections (like I said all it takes is the smallest of changes, and doing it handheld is not so easy)

 

Also I would strongly advise to avoid any filter shooting in this kind of extreme backlight, as they can only make things worse...

[Guest all8]

Perhaps using the LED Torch is not completely a fair test (or perhaps it is), however, what surprised me is that no matter the angle of the camera to the Torch I could not get rid of this reflection. It was more visible in a darkened room, and almost invisible in a well lit room. The only other camera I have is a Ricoh GR - I could not reproduce the pattern with that camera (large green blob sensor reflections were easy to induce).

 

Someone with an A7ii and A7rii might want to do something similar, I would be interested to know if the A7rii is handling this situation better. 

[Username]

One of your lens elements has an unfortunate characteristic. 

This is not a defect, but can be considered a less desirable 

design. The concave curvature just happens to magnify and

focus a reflection of the micro lens array on the sensor.  A 

lens design thaz been very thoroughly computed for use on  

digital cameras will almost never do this. It might still reflect

the sensor occasionally, but not a sharply magnified image

of the micro lenses ... more of just a vague shapeless green

area. Your particular problem MIGHT be reduced by working

at a wider aperture. I expect that another lens of exactly the

same FoV, at the same f/stop you've used above, would not

deliver the result you posted, cuz your result is a fluke of the 

particular curvature of one of your lenses elements. There's 

apparently no actual magic to the Zeiss name. I use an off 

brand 17/3.5 on my a7-II and never encounter this. I do get 

the occasional single blueish blob which is a lot easier to get 

rid of in post. This is an old film era lens without the latest in 

anti-reflection coatings. It just doesn't happen to have some 

unfortunate curved surface that acts as a magnifier.  

  

IOW, altho it's a combination of the two, it really IS the lens,   

and not the camera, that is to blame for this effect. You could 

have used a different camera and recorded that same effect, 

but if you had used most any other lens, it would be gone :-(   

I can see by the diffraction pattern on the sunburst that some  

rather small f/stop was used, which does make things worse. 

[Username]

I thought circular polarizers were not effective when looking into the

Sun. Ideally you want 90 degrees, right? Also with wide lenses like

an 18mm, polarizers can be problematic, causing non-uniform

brightness gradations.

  

Thaz all 100% true. The right side of your image already 

shows this effect, as blue sky is self-polarizing and using 

a polarizer only exaggerates that. With a really wide lens,

and with the sun well toward the left side, the right side is

already approaching that 90 deg viewing angle. Acoarst 

the fall-off toward the corner with a really wide lens only 

increases the effect, but at the small stop you used, it's 

prolly mainly the self-polarizing effect that we see there. 

[Dollanganger]

There is no need for a Circular polarizer using  

a mirrorless live view camera such as the a7x

and a6xxx or Nex series. All camera functions

are unaffected/normal. 

 

That's certainly true for UV filters and I've heard other say similar things about circular polarizers and mirrorless, but I personally have noticed night and day differences with and without a circular polarizer.