Cause of lens flare/streak and rainbow fringe from lightsource in primetime drama

[Christopher Lefchik]

Below are a couple screenshots from episode 7 of the American/CBS version of the television drama Eleventh Hour. I noticed the horizontal light streaking and rainbow color fringing on strong light sources and was curious if anyone might know the cause.

I've seen the light streaking effect in movies, which if I'm not mistaken is an effect of the anamorphic lenses used to achieve extra wide aspect ratios. But this is a television show in the 16:9 ratio. And the rainbow fringing is something I can't recall seeing in other shows or in movies.

Eleventh Hour screenshot 1
Eleventh Hour screenshot 2

You can view the video clip these screenshots are taken from here: Eleventh Hour Video: Eyes on the Target - CBS.com. The outdoor night scenes start at 1:24.

[Perrone Ford]

CCD technology and glass. Film reacts differently.

[Christopher Lefchik]

But if the streak was caused by an IT-type CCD sensor, shouldn't the resulting smear be vertical?

And note that if one looks closely at the first screenshot, you can see a rainbow fringe on the horizontal streak in addition to the main point source. If this were a smear caused by CCD storage register light contamination overload, I wouldn't think that should be the case.

As an experiment, I pointed my old Sony DCR-VX2000 at a flashlight. If the iris was set at around F2.0/OPEN or above (with gain added), I could create a lone vertical smear. If the iris was set around F2.0 or below, I would get a vertical smear, but it would be accompanied by an X cross-type smear that was strong as the vertical smear.

[Perrone Ford]

Interesting... maybe someone knows the true answer.

[Brian Drysdale]

Originally Posted by Perrone Ford

Interesting... maybe someone knows the true answer.

It looks like they're using a 2 star filter to simulate the horizontal flare you get on anamorphic lenses.

I don't know what they shoot this series on, so I can't comment further. However, the rainbow effect could be from the filter itself, depending on the pitch of the star markings, so that these could be acting like a prism.

[Jeff Donald]

Originally Posted by Christopher Lefchik

But if the streak was caused by an IT-type CCD sensor, shouldn't the resulting smear be vertical?

And note that if one looks closely at the first screenshot, you can see a rainbow fringe on the horizontal streak in addition to the main point source. If this were a smear caused by CCD storage register light contamination overload, I wouldn't think that should be the case.

As an experiment, I pointed my old Sony DCR-VX2000 at a flashlight. If the iris was set at around F2.0/OPEN or above (with gain added), I could create a lone vertical smear. If the iris was set around F2.0 or below, I would get a vertical smear, but it would be accompanied by an X cross-type smear that was strong as the vertical smear.

What happens when you stop your old Sony camera down to f/16 etc. Pointing camera at a bright light source would cause it to stop down (small opening) not open up to f/2. Small opening cause diffraction and other optical defects that cause color fringing.

[Christopher Lefchik]

Originally Posted by Brian Drysdale

It looks like they're using a 2 star filter to simulate the horizontal flare you get on anamorphic lenses.

I don't know what they shoot this series on, so I can't comment further. However, the rainbow effect could be from the filter itself, depending on the pitch of the star markings, so that these could be acting like a prism.

Thanks for the feedback. That would make sense as I did notice that the rainbow fringe didn't appear around light sources in some shots/scenes.

[Christopher Lefchik]

Originally Posted by Jeff Donald

What happens when you stop your old Sony camera down to f/16 etc. Pointing camera at a bright light source would cause it to stop down (small opening) not open up to f/2. Small opening cause diffraction and other optical defects that cause color fringing.

Normally pointing the camera at a bright light source would indeed cause it to stop down—unless the iris was in manual mode, as it was in my test.

Stopping the lens down to F/11 (the smallest aperture available) gives me the same results described for F/2 in my second post: a strong vertical smear accompanied by an equally strong X shaped smear.

[Ashley Briggs]

I'm a bit late here but worth my input;

Recently tested shooting through tights (stockings, hosseire etc) and found that man-made fibres gave this effect. Get a pair of 10 denier or there-abouts black tights, cheap ones make more of a rainbow effect, and stick them either on the front of the glass or behind the glass - stretching them to different opacities. Should give this look - I got it by mistake.

ash.

[Brian Drysdale]

Originally Posted by Ashley Briggs

I'm a bit late here but worth my input;

Recently tested shooting through tights (stockings, hosseire etc) and found that man-made fibres gave this effect. Get a pair of 10 denier or there-abouts black tights, cheap ones make more of a rainbow effect, and stick them either on the front of the glass or behind the glass - stretching them to different opacities. Should give this look - I got it by mistake.

ash.

It could be a combination of filters. Nets usually give a small star effect rather than a streak, but there's no reason why they couldn't add a 2 star filter in addition to the net that the production may be using as their "standard" filter.