F3 Nissan Spot - Discolored Seat Belts (IR Issue?)

[Philipp Eierund]

Just shot a Nissan spot and had a very strange problem with the black seat belts being captured red by the F3. Do you think this is an IR issue? We did not use any IR filters, however, I'm not even sure if that was the problem.

I've attached a screenshot.

What do you guys think?

Thanks,

Philipp

[Henry Epstein]

If you were using ND.6 or ND.9 it should be very possible...

IR Filtering is almost mandatory at that ND level for CMOS sensors.

[Philipp Eierund]

Yep. We we definitely using some high ND filtering here. Could you explain why ND filters increase the need of IR filtration?

[Henry Epstein]

Here there is a short explanation about IR contamination. Hope it helps.
You might find a lot more and better explained in this forum.

H

[Philipp Eierund]

Thanks, Henry. Now do you think the Tiffen T1 filter is sufficient with the F3 in combination with the build in ND filter or Tiffen ND filters, or do you think one has to go with the Tiffen Full Spectrum ND IR filters?

[Henry Epstein]

Very good question and very hard to answer for me since I do not have yet my F3.

Nevertheless, many at this forum might answer that... or have already!

Someone, please?

[Brian Drysdale]

I gather each IR ND has a precise level of IR filtration for its neutral density factor.

With the Tiffen T1 with filters over ND 0.9, Tiffen recommend using an IR ND combination filter. You can see that in the note at the bottom of the T1 data sheet.

I'd be interested to see if the variable ND filters suffer from IR on these cameras.

[Philipp Eierund]

This is definitely something that needs to be investigated. It sucks that this is an issue because it means that we will always have to shoot with a mattbox system attached to the camera for filtration.

Especially when shooting cars it's nice to be able to quickly pop off the mattbox system and use the build-in ND filters for the tight interior shots.

Of course that means you'd also loose your IR filtration... :/

[Thierry Humeau]

Aside from the IR effect, I kinda like the red seat belts. Very cool!

T

[Henry Epstein]

I must assume that those Internal ND filters in the PMW-F3 are already IR corrected.

If that's the case, with a plain ND.9 + the T1-IR @ 1/64ND in the F3 we should be able to shoot at 0db/800ISO @ 24p/25p (1/48 NTSC -1/50 PAL) at f2.8 in a sunny day... according to the Sunny 16 Rule.

"On a sunny day set aperture to f/16 and shutter speed to the [reciprocal of the] ISO film speed [or ISO setting] for a subject in direct sunlight."

Please do the math and correct me if I'm wrong...

1/64ND, 6 f-stop reduction
ND.9, 3 f-stop reduction
T1-IR, .5 f-stop reduction

???

[Leonard Levy]

I did some of the initial testing with the Tiffen rep that led to the T1 filter.

Yes that is "far red contamination" ( its not actually IR). Frankly I think anyone shooting an EX or F3 without owning a T1 is just setting themselves up for this- dare I say "irresponsible." Wait till you shoot a CEO and he has red pants.

The F3 is better than the EX1/3 but still has some problems.
However ND filters create a host of additional "Far Red" problems. So here are some thoughts:

1. Different cameras have problems with red or IR contamination in different parts of the spectrum. Thus an ND filter with IR cut added may be irrelevant to an EX-1 or F3 but solve problems on a RED or F35 and vice versa. The EX and F3 seem to have problems with far Red not IR. (Far red is nearly IR but still visible - outside of the spectrum that most photographic film was sensitive to, hence ND filters made for film didn't need to extend into those parts of the spectrum.)

2. In principle if your ND filters don't also filter far red or IR then as you use increasing ND the amount of Far Red increases logarithmically. So shooting with ND filters can be a big problem. So what ND filters should you use? Here's where in my opinion it gets complicated.

3. In my testing I found NO INTERNAL filters on any cameras (including the F3, Ex-1, HVX200, and a number of other cameras) that generated any Far Red problems when they were engaged. So presumably they included Far Red in the ND spectrum. (But these were just my tests and I didn't check every camera)

4. Thus - if you use a T1 on an F3 or an Ex-1 you will not have additional contamination when using the internal ND's.

5. If you use additional ND's it is a crap shoot that I have no clear answer to. While we were testing very expensive Tiffen, Format and Schneider ND filters with IR cuts supposedly in them I picked some up held them to eye and said - Hey this thing is creating red contamination of black fabrics to my naked eye!
Every traditional Tiffen ND filter I've ever tried does this and you can test it yourself just my holding a black fabrics in front of your own ND's.

On the other hand I own some 82mm B+W and Heliopan filters that showed no issues - black remained black. If buying round ND filters I would stick to those brands because I know they work and don't cost anything extra.

5. However Those brands are not made in 4x4 type filters and so I've been reluctant to buy any until i can test them. Theoretically Schneider makes B+W so perhaps they should be the same, but i want to see before I buy, and I already have seen problems on some Schneider ND's. - I would love to know what brands are problem free if anyone tests.

Hope that helps. its a complicated little issue. If you do any testing yourself try a range of black fabrics since some will have problems and some won't.

Lenny Levy

[Dean Harrington]

Far Red contamination ... I had some serious problems with this on a shoot a few months ago. You can correct this problem in post in Apple color if you use final cut pro. If the problem is stationary ... it's not a person jumping around ... you can also use color correction in Magic bullet colorista 2 to change to black. Nasty stuff and a pain in the arse to do but hey, it's apart of the job!

[Philipp Eierund]

I'll have to run some tests soon. This is a nasty problem and Nissan is certainly not going to accept red seat belts. ;)

Dean, any pointers in correcting this issue in post?

Thanks!

[Peter Moretti]

Essentially you're going to have to do a secondary color correction based on selecting the brownish-red black and replacing it with a more palitable black. You might also have to use a garbage matte to keep other parts of the image w/ the same brownish-red black from being affected.

Hope others chime in.

[Aaron Newsome]

Originally Posted by Peter Moretti

Essentially you're going to have to do a secondary color correction based on selecting the brownish-red black and replacing it with a more palitable black. You might also have to use a garbage matte to keep other parts of the image w/ the same brownish-red black from being affected.

Hope others chime in.

Peter has the right idea. This should be fairly easy to fix. The garbage matte doesn't need to be too tight so tracking it should be fairly simple. The color remapping is not too extreme so it should blend fairly well.