Major issue with HVX and ND

[smelni]

First let me preface this with saying I love the HVX and we are shooting our feature with it.

But we have run into a major issue - it appears that the HVX, for whatever reason is not compatible with large amounts of ND. I have seen this with both filters and ND on windows.

In low abounts (.3) it is fine and the filters and/or gels cut the light correctly. But in higher amounts (.9 and higher) the ND turns the light pink. So much so that it is noticable. If you use white balance to correct it, the scene turns green to offset the pink.

This has happens with me with tiffen ND filters and with roscoe nd gels on windows. My understanding is that ND is color agnostic and should not cause this.

So this limits creative use of bright exterior lighting with interior scenes and is causing huge issues for our production.

Has anyone else seen this, have an explanation or a suggestion.

Jan - panasonic - any reports of this from other users?

[Barry_Green]

What you're describing is fundamentally not possible. An ND filter is, by definition, neutral -- it cannot impart any sort of color cast to the footage.

Cheap filters are made from so-called "green glass" and could impart a small green tint to the footage; if you're using screw-on Tiffens I'd suspect this could be what you're running into.

You've got six stops of ND filtration built into the camera (on 1/64) and there's no pink or green cast when using that, right?

[smelni]

Barry - beleive me - I understand it is theoretically fundamentally impossible. But it is happening for sure and is repeatable and has been seen by our AC and another DP on set.

We have seen it with tiffen 4x4 - only at ND .6 or more.

We saw it today with 1.8 total of ND on the windows- and it was roscoe gels - the skylight was pink - very obviously pink - and if we white balanced on that the sky was white but the room was green (cause the green and red channels are opposite i guess).

It is baffling - try it - on a sunny day (very sunny) put 2 layers of .9 on a window and then point the camera at it. It is repeatable - and very disconcerting - we had to frame windows out of the shots and even black them over cause they light coming through was tinted

Fundamentally impossible - in theory only - cause it is happening on our set.

Any frustration in this message is not towards you Barry - you know I have the upmost respect for you - the frustration is at a seeming impossible and therefore insolvable problem

By the way - i would LOVE to be proved wrong as this is killing us

Oh and yes - the internal ND does not have the issue

[Barry_Green]

I know you're not frustrated at me, no problem, don't worry 'bout that.

But this line is telling:

the internal ND does not have the issue

That in and of itself leads me to say that it's not any sort of issue with using ND with the camera, obviously. So I'm left to wonder about the quality of the ND stuff you're using. And the only thing I can think of is to say "try a different camera for a test" -- preferably a different brand. Would a Sony or a JVC also go pink in the same circumstances? If so, it lets you know what I suspect to be the case: that the color tint is due to the ND filters you're using, not due to any fault of the camera itself. They're supposed to be color-neutral, but apparently they're not.


Not that I'm blindly defending the camera, but that it simply doesn't make sense. The camera doesn't know whether you're using ND or not; if you use a .6 ND, or you just close the iris by two stops, the net effect to the camera is the same.

If the HVX footage is turning pink, then I'd expect that any other camera in the same conditions would also turn pink.

I'll dig through my stuff and see if I have any window gel here and try to replicate it. In the meantime, I'd suggest trying to add a bit of whatever the color complement of pink is -- would that be blue? blue-green? anyway, whatever it is, a mild shade of that should compensate and get you back under proper color conditions. Might keep you up and running until you can get to the bottom of it.

[Lenilenapi]

It would help if you could be much clearer about what you are seeing and doing. I am confused about whether you are using rosco window gel or ND filters on the lens.
I have definately seen problems with rosco gel not being truly ND. In my case I remember seeing significant magenta casting and I have a vague recolllection of seeing that compound with heavy layers of ND. Also have seen it on 85ND and blue correction. Its probably a general problem with that kind of material, simply doesn't have the same standards as glass filters and even those differ in their standards. I've had bad polas and bad ND's . Its not uncommon and Tiffen may be more prone than say heliopan. Usually you can tell just by looking carefully at the ND with your eye

Seems to me I started using Lee and Cinemills gel because the colors were more reliable but that was some time ago.

I'd be very surprised if it was an HVX problem, and the first thing to do before making those kinds of claims is to compare it to another camera.

[billiamfilms]

why not instead of trying to make the outside (EXT) less bright by using ND filters, make the indoors more bright and expose for that. That way you wont need ND filters really if the exposure of the indoor and ext is decresed? Just a thought. I could be way off from what you wanted. Just trying to give a suggestion that wouldnt involve even using the ND as much.

[Anders Holck]

What kind of lightsource are you using for the room?

If you are using fluorescent and white balance to that, you might introduce a slight magenta cast. When the windows re blown out you won't notice this but as you bring the intensity down it might become obvious.

To test the camerafilter, try holding it with your hand in front of the lens. Then as you tilt the filter towards the camera, see if the cast intensifies. If it does it's the filter alright.

In my experience Tiffen and Format polarizers can introduce a slight green cast while schneider doesn't.

[Luka_Brazzi]

The problem lies with filters rather than the camera. I have had this problem on a regular filmset. We have used around 1.5 total ND and the colormeter started showing CC+sthG. The point is that sometimes filters (even glass filter for the camera) act like that below the normal "magenta introduction limit" (it is normal that after you reach 2.4 ND magenta may show even with new filters) because they are old or cheaply made.

What you should do is check the filters with a colormeter. If this problem shows on the meter not only on the camera than you now where is the problem.

[rgbuser]

Technically speaking, there is no truly neutral density filter if by that one means that all wavelengths of the visible spectrum are equally attenuated (absorped) across the entire visible spectrum - the power spectrum is perfectly flat. The nearest to true neutral are filters used in optics labs which rely on semi-silvering or a scattering of carbron particles (also seen in centre-filters for wide-angle lenses).

So let's assume that the best of normally manufactured filters will stop most wavelengths equally but let some through, er, a little less equally. In this case it looks like there are dips in the blue and red absorptance which become visible when you pile a lot of ND together. If there is imbalance in two axes of the chrominance, then I'd guess it is very difficult for the camera to correct - which could be why Smelni is getting green in response to his attempts to correct the magenta. It could be that Panasonic's built-in filters are matched to the absorptance bands of the sensors, so they are truly neutral but only relative to the triple-sensor set-up (in fact that is what we'd have every right to expect) - which means you may not replicate the error with another camera which uses a different sensor/Bayer filter set-up.

Solution? What about setting 1/64ND on the camera, and a lighter ND on external filters or windows - sorry, I know that won't solve the luminance balancing problem? Thanks for bringing up an interesting one.

[smelni]

Thank you all for your responses - it has been helpful to have ideas

First I am sorry if I am blaming the HVX without further understanding - after reading this I think I would have to blame the gels and filters - 2 different cases with the same results. Its odd that both the Tiffen filter and the roscoe gels (each used independently) have the same cast but that seems to be the case

To those whose suggested the internal ND and using more internal light - that really isnt an option in our case - to make up the 6 stop difference we would need so much light - and we are working on 15 amps circuits (no tie-in) so we dont have enough power - and if we did we dont have the right lights to have the proper quality of light in that intensity - besides the fact that we would cook the talent.

It is very frustrating and has led us to framing windows out of shots - but production is always a series of comprimises.

I am surprised more people dont run into this - i would think more people would be filming scenes in bright daylight indoors with windows in the shots- but if you do beware that you may not be able to get the correct ratios just by using ND.