Looks like a combo of lens and sensor. The last one looks more like typical lens flare. Were these all shot at f/2.8 as the EXIF indicates, or is that just a default value for a non-coupled lens? Considering there is no starburst type effect around the sun, and the flare has no strong edge definition it kind of does look like these were wide open.
It's interesting that you got a ghost reflection of the reflection of the sun in the water. Looking at the ZE/ZF 21's block diagram, there are a couple lens surfaces that appear to be perfectly flat, or very close to it, and might be enough to bounce back a weak reflection of the sensor reflection. This was one of the big 'features' of the previous generation Canon super-teles with a slightly curved front protective element to prevent direct reflection of sensor reflections back into the image path...
I don't own the ZE/ZF 21, so can't speak from experience with it. If indeed you were shooting wide open, or close to it, stopping down a fair amount would considerably diminish the size of the flare, though not sure it would affect the reflection. The Canon TS-E 17 is susceptible to flare if a strong specular light source shines directly on the large front element. My experience with it has been that if the situation is unavoidable, it's best to stop down a fair amount, such as f/8 or 11, and the flare spots will be small enough to usually be easily spotted out of the image in post.
Both have some flare. The first image it's just under the 'er.com' in the watermark, the second image it's at the lower left third in the water. Both shots were stopped down to around f/11. But I don't see any indication of a ghost reflection. Do you remember Luka's trip to the US when he was in Antelope Canyon and got a really nasty flare with the ZE21? It could be the lens is susceptible to flare in certain suboptimal situations...
OK, that definitely looks like a normal lens flare string with a pretty big, but faint one, in the dark part of the water. I wonder if it's going to be more of a problem if the bright light source is nearly dead center...?
In your initial post you indicated you got a few images without the flare/reflection. Any chance you can post one or two of those?
Well, my understanding of sensor reflection problems is that the lens is also a factor. Sensors are very reflective. If a reflection is visible in the image, some of the lens optics are responsible for reflecting the light back at the sensor again.
Your first images don't look like typical lens flare, as seen in the later example with the trees and water.
I believe this is one aspect when manufacturers state their lenses are optimized for use on digital cameras. Not only are there considerations such as the angle at which the light reaches the pixel sites, but also how the lens elements deal with light reflection off the sensor surface.
I have mentioned it before, but I have had issues with sensor reflection with the a900 (and I would guess the a850 would be similarly susceptible). In my case, it has shown up with many different lenses but only when doing critical copy/ reproduction work under controlled studio situations, typically at distances of say 20ft or closer. I have never seen it outside at greater distances when shooting general subject matter - though I don't shoot a lot of sunsets and the only time I shot into extreme light was for lens testing (and I never saw it in that situation either). Some of what I see in your first series does look similar to the reflection shapes I saw but mine were much more subtle - not noticeable in reduced sizes- and in every case they were red/ magenta in color, regardless of subject or lens used. Nothing I tried prevented them (and I have a LOT of professional lighting experience) when they showed up and the only solution was removing the reflections in PP. Given this experience, I suspect the sensor in the a900/a850 is more susceptible to sensor reflection than most other sensors). It's really the only fault I have ever found with the camera.
FlyPenFly wrote:
And in the majority of cases, where the flare isn't shown is when the sun is closer to the borders.
Although in the first example, I guess I got lucky.
My guess is, the flare/reflection is still there but when you put the sun closer to the borders, that image is off sensor...
In the first example the flare is there, but right on top of the bright streak in the water. Therefore it goes unnoticed.
You need at least two reflections for flare patches, so reflection off the sensor alone cannot explain your observations. The flare has mirror symmetry with respect to the image center, which normally characterizes reflections between sensor and (protective) filter. In the absence of a filter the culprit has to be one of the lens elements, presumably one that is faraway from the sensor and reasonably flat. Reasonably flat but not entirely, as evidenced by the comatic smearing of the blur patch as the source moves away from the image center.
The fact that the flare disappears when you put the sun close to the borders is expected. If the reflective element is in front of the aperture stop, you will find that the vanishing point depends on the f-stop.
Toothwalker wrote:
In the first example the flare is there, but right on top of the bright streak in the water. Therefore it goes unnoticed.
You need at least two reflections for flare patches, so reflection off the sensor alone cannot explain your observations. The flare has mirror symmetry with respect to the image center, which normally characterizes reflections between sensor and (protective) filter. In the absence of a filter the culprit has to be one of the lens elements, presumably one that is faraway from the sensor and reasonably flat. Reasonably flat but not entirely, as evidenced by the comatic smearing of the blur patch as the source moves away from the image center.
The fact that the flare disappears when you put the sun close to the borders is expected. If the reflective element is in front of the aperture stop, you will find that the vanishing point depends on the f-stop. ...Show more →
Interesting. I was wondering why the reflection of the sun was smeared in that manner. As far as nearly flat lens elements... take your pick. My bet is on #10 (counting from the left).
Lenses for Digital have a coating also at the back element, so that reflections from the sensor aren't reflected back to the sensor. Our alt lenses therefore are more prone to flare arising from sensor reflections.
Probably true. I use only alt lenses mostly older film lenses, and I've never seen a sensor reflection that I know of. I guess I've taken something close to a quarter of a million shots over 4 GH1 bodies and not seen one yet.
I've seen buttloads of filter reflections, lots and lots of lens flares, lens reflections at night, extension tube flares, adapter light leaks, and so on but no sensor reflections yet.
phuang3 wrote:
This is not sensor reflection. A sensor reflection doesn't have comet-like tail, it's just a mirror of point source on the frame.
Did you notice this point:
Toothwalker wrote:
In the absence of a filter the culprit has to be one of the lens elements, presumably one that is faraway from the sensor and reasonably flat. Reasonably flat but not entirely, as evidenced by the comatic smearing of the blur patch as the source moves away from the image center.
The fact that the flare disappears when you put the sun close to the borders is expected. If the reflective element is in front of the aperture stop, you will find that the vanishing point depends on the f-stop.
H.Lux wrote:
Lenses for Digital have a coating also at the back element, so that reflections from the sensor aren't reflected back to the sensor. Our alt lenses therefore are more prone to flare arising from sensor reflections.
Compound lenses have coatings on all glass/air interfaces. That includes the great majority of alt lenses. It makes no sense to coat only the rear element.
I use only alt lenses mostly older film lenses, and I've never seen a sensor reflection that I know of. I guess I've taken something close to a quarter of a million shots over 4 GH1 bodies and not seen one yet.
I've seen buttloads of filter reflections, lots and lots of lens flares, lens reflections at night, extension tube flares, adapter light leaks, and so on but no sensor reflections yet.
Just thought I would mention it.
I bet you have seen many sensor reflections without realizing it, but as I said before a sensor reflection alone does not lead to flare. You need a second reflector. Many cases of 'mirror ghosting' can be traced back to reflections between sensor and (UV) filter, or film and filter in the old days.
rscheffler wrote:
Interesting. I was wondering why the reflection of the sun was smeared in that manner. As far as nearly flat lens elements... take your pick. My bet is on #10 (counting from the left).
Want to make any prediction we'll eventually see something similar from the new ZE15?
No bets or predictions. EURO 2012 is difficult enough.
H.Lux wrote:
Lenses for Digital have a coating also at the back element, so that reflections from the sensor aren't reflected back to the sensor. Our alt lenses therefore are more prone to flare arising from sensor reflections.
Nice theory, but all today Zeiss lens are for digital at first place. Btw. Zeiss coatings didnt have much trouble with sensor reflections even in C/Y age.
Btw. Zeiss coatings didnt have much trouble with sensor reflections even in C/Y age.