Sensor reflection/ghost images

Rick Burnett said:
Not at all. It's why, for night shooting, for me anyways, filters are an absolute no. Definitely keep us posted!
Click to expand...

That's not always possible and for a camera without any built in ND that's a big problem.

I'm pretty sure the reflections are from the front, but i really don't understand why. What is it about the FS that is making them more obvious.

Perhaps a decent coated ND would help, and not a variable ND either.

Or are there any adapters (for SLR lenses) that can have a filter slot in them. I don't know whether that's a realistic possibility?

cheers
Paul
 
Or maybe those heliopans are not very good?

With Frank's examples it would look like it indeed, but I can assure you the reflections can appear without the Heliopan (or any other extra glass), as well...
 
paulcurtis said:
That's not always possible and for a camera without any built in ND that's a big problem.

I'm pretty sure the reflections are from the front, but i really don't understand why. What is it about the FS that is making them more obvious.

Perhaps a decent coated ND would help, and not a variable ND either.

Or are there any adapters (for SLR lenses) that can have a filter slot in them. I don't know whether that's a realistic possibility?

cheers
Paul
Click to expand...

Hi Paul. When you say the reflections are coming from the front, do you mean a front element inside the lens? Because most of those sample shots were without a filter.

I sent an email to Sony Support but I have a feeling my request will need to get bounced around a while before it gets to the right place.
 
danstone said:
Hi Paul. When you say the reflections are coming from the front, do you mean a front element inside the lens? Because most of those sample shots were without a filter.

I sent an email to Sony Support but I have a feeling my request will need to get bounced around a while before it gets to the right place.
Click to expand...

I think from the front element, you can make the reflections a lot worse as Frank showed with that ND. My reasoning is:

- In your tests the highlight wasn't stabilised.
- On my 21mm at an angle i could see the window as a clear reflection and the only way that could happen is if it hit the curved front when it was facing away
- It's the right way up (I think)

Of course entirely prepared to be completely wrong, these are just guesses. Of course i don't understand WHY or what it is being reflected ON. Because all cameras would have the problem, so there's got to be something.

I've not seen this with my NEX 5N which is my preferred stills camera at the moment. I use some of the same lenses on both. Both the 5N and FS100 have a very short flange depth but it can't be that alone.

So i would guess, either something on the OLPF or the Sensor itself. Of course the F3 shares the same genetics for the sensor - i'm not aware of it being a problem on this?

Good luck with Sony, Rick seems to have had some good communication from them so fingers crossed.

cheers
Paul
 
paulcurtis said:
I think from the front element, you can make the reflections a lot worse as Frank showed with that ND. My reasoning is:

- In your tests the highlight wasn't stabilised.
- On my 21mm at an angle i could see the window as a clear reflection and the only way that could happen is if it hit the curved front when it was facing away
- It's the right way up (I think)

Of course entirely prepared to be completely wrong, these are just guesses. Of course i don't understand WHY or what it is being reflected ON. Because all cameras would have the problem, so there's got to be something.

I've not seen this with my NEX 5N which is my preferred stills camera at the moment. I use some of the same lenses on both. Both the 5N and FS100 have a very short flange depth but it can't be that alone.

So i would guess, either something on the OLPF or the Sensor itself. Of course the F3 shares the same genetics for the sensor - i'm not aware of it being a problem on this?

Good luck with Sony, Rick seems to have had some good communication from them so fingers crossed.

cheers
Paul
Click to expand...

I think you're right on the money -- the reflected image is right side up and wasn't stabilized.

Just talked to someone at Sony who is forwarding this to a service engineer. We'll see.
 
I had also corresponded with a Sony service representative before I saw DANSTONE's post, and he is having the engineering department look into it, too, but for today he says:

My suspicion is that you are seeing the same type of lens flair that I have seen in 35mm and Super 35mm film cameras. With a large surface area of the film or CMOS Sensor and the shallow flange focal distance used, you are picking up the reflections off the sensor or film being picked up by the rear lens element that are reflected back into the sensor.

The strong light overpowers the coatings on the lens elements and the inside of the lens barrel that is supposed to help to prevent this from happening. Unlike higher end cameras that have shading and flair adjustments to compensate for this type of problem, the NEX-FX100 relies only on the lens itself, just like film cameras.
 
Kraut69 said:
My suspicion is that you are seeing the same type of lens flair that I have seen in 35mm and Super 35mm film cameras. With a large surface area of the film or CMOS Sensor and the shallow flange focal distance used, you are picking up the reflections off the sensor or film being picked up by the rear lens element that are reflected back into the sensor.

The strong light overpowers the coatings on the lens elements and the inside of the lens barrel that is supposed to help to prevent this from happening. Unlike higher end cameras that have shading and flair adjustments to compensate for this type of problem, the NEX-FX100 relies only on the lens itself, just like film cameras.
Click to expand...

This is consistent with what I have observed. I was shooting a street scene the other day in late afternoon. I noticed that when a car's headlights reached a certain brightness (the car got close enough), then the reflection popped up. Under that level/size and there were no reflections. ND on or off filters made no difference.
 
Kraut69 said:
Unlike higher end cameras that have shading and flair adjustments to compensate for this type of problem, the NEX-FX100 relies only on the lens itself, just like film cameras.
Click to expand...

I wonder what kind of 'shading and flare' adjustments these high end cameras are supposed to have (not that i'm aware of anything)

I did some more research and yes, it's a problem even in film days but then film was a lot more forgiving and rarely cropped up.

But something is making it more pronounced on the FS100, something that other cameras don't seem to have. So whilst i completely accept this is part and parcel of optics etc,. i don't accept why it is shown so strongly on the FS, compared to, the F3 for example (I've not found much evidence on this configuration). There's nothing in the F3 that would compensate for this, perhaps the problem is the coatings on the OLPF (although i would imagine the F3/FS would share the same filter). Also it seems particularly bad for danstones camera compared to mine for example, that's what kicked the thread off.

The short flange distance doesn't seem to make sense as i've seen the effect more pronounced with contax zeiss lenses on the CY mount which is at least 40mm from the sensor.

I hope that sony can shed some light (pun intended) and explain what the optical path is, so we're more aware of how we can avoid the situations it crops up in.

cheers
paul
 
I was wondering the same thing - why is this not a problem with the 5D (even bigger sensor) or the F3? So weird! I'm tempted to try a piece of polarizing glass between the sensor and lens.

Talked to a Sony tech who forwarded my video to the engineering department and said he'd get back to me. I'm crossing my fingers but have a feeling I'm going to get a "that's just the way it is" answer like Krout69 got.
 
I really noticed this problem for the first time yesterday. Very similar to the examples posted here. I had a subject against a very bright set of square windows, which were totally blown out and above the subject. The reflections were quite pronounced, almost like I was shooting through glass and I was seeing the reflection of something on the surface of the glass. I had to change the angle of the subject to the camera to move these reflections off the frame. I seems if you're aiming nearly straight at a bright defined light source like this it happens. I experience this with a Rokinon / Samyang 35mm EOS lens using an adapter, so the back glass was far away from the sensor. I'm not necessarily buying the 'distance to the sensor' issue. I also had a Heliopan variable ND on the front. I don't know if this contributed to the problem, though it seems this has been ruled out here? I was also shooting pretty 'open' about 1.4-2.0, maybe this also contributes to the problem?

Anyway, it's a disturbing problem, I've never really had to deal with this issue before with another camera without some type of simple explanation. It seems pretty easy to reproduce, and I will when I have time, just put a well defined light source behind a darker subject, it should be pretty obvious. So I think it is a problem endemic to the camera, a little depressing because I like to shoot with strong backlights. I'm interested to find out what exactly is causing the problem and what we might do about it, other than shading light sources, sometimes you want that source in the frame or it's not possible to put a shade in because of constraints.
 
Well I can definitely say when I was shooting with a 30mm lens with a Genus or LightCraft ND on the front of the lens, I had reflections. I took them off and they went away.

I think it's a combination of factors causing it. To get a reflection back at the sensor, there needs to be something reflecting back. That's the question. A combination of distance and reflective surface curvature probably both play a role. As does the reflectivity inside the lens mount. Someone already showed this on the AF100 that having a very reflective interior causes this problem to be worse (and it makes sense, more light bouncing around).

What I want to know is why does the AF100/FS100 bounce more light than say the 7D. One person I talked with believed the sensor coatings might be an issue. I don't have my 7D at the moment to compare the sensors.
 
After more googling and reading and thinking. I wonder if the alignment of the OLPF is causing more or less reflection issues that could explain a difference from camera to camera? Perhaps if aligned perfectly flat on, the effect is minimised but with some error (angled) it can make it worse.

Also on CML recently there was a discussion about the OLPF dividing light into 4 targets to cover the sensor locations. This is the first time i'd read about anything like that and if true could mean that alignment is paramount when these things are being manufactured.

cheers
Paul
 
This almost sounds like the purpose for anti-halation backing that is applied to photographic films. Does anybody know if there is a similar layer applied to the imaging chips?
 
I think the OLPF being a little off would help since the reflection wouldn't be right back from the lens from light being reflected off of the sensor.

As for the location of the OLPF, I don't think that's a problem. If the OLPF is like any others I've read about, they use birefringent materials to shift the image in multiple directions to cause it to spread over the surrounding pixels (basically the same thing diffusion would do). That said, it's not the actual location of the birefringent material matters, but the photons hitting it. For this reason, there is no need for it to be in an exact position. Unless of course you mean rotational, which in that case, it could affect where the copies are, but, we'd see that as more aliasing and I've not seen more aliasing on other examples of the camera. But even besides that, that rotation isn't going to cause the reflectance of the surface to be more in my opinion.
 
I just did a shot with Leica M lenses on FS100 in hard contrast situation. Low sun behind the subject and it was the first time that I noticed the green flares. First I thought there was some kind of burn in on my monitor but then remembered that Frank Glencairn mentioned the ghosting on Twitter just the other day. I have footage and a frame grab on my blog.
http://blog.domisljije.si/2012/02/20/leica-m-lenses/
 
Now that I am looking for it, I see it on a lot more cameras then before when watching a lot of event coverage footage (Like NAMM).
 
The problem is that it has to be really strong light. If you are shooting into the sun it is only a small dot that can be also interpreted as lens flare. But if you shoot at the large reflective surfaces like I did, low sun behind the subject in the water, than the highly reflective surface is really big and the problem becomes very visible.

Green-Ghosting-White-Hordes-.jpg
 
I am surprised this hasn't come up before now!!!!!! I would expect something like this to be caught by the first reviewer?????????!!!!!!!!!!
 
You must log in or register to reply here.
Back
Top