IR Contamination Issues? (IR "hot mirror" filters))
 

Hi there --

Some camcorders are known to have problems with IR contamination and there are tons of threads on this board about this issue with the EX-1 and Red camcorders. Do you know if the JVC ProHD camcorders CMOS sensors have similar issues (i.e., lack of IR absorption in the light path)?

I am asking because under certain light conditions (bright tungsten or outdoor) my footage tends to be a bit magenta which is characteristic of IR contamination. So I am wondering if I should get an IR filter to see if this solves the problem.

Does anyone have experience with a IR Cut filters ("hot mirror" from Tiffen, Schneider or Redrock) on a JVC camcorder?

Cheers!

While I can't comment on IR issues, I can point out that the ProHD cameras use CCDs and not CMOS chips.

Stefan,

It sounds like you have an IR contamination issue but images to that effect are needed before any of us can comment for sure.

One note is that IR contamination issues are not bound only to CMOS sensors but certainly have a higher occurance than CCDs. I have not tested our Schneider True-Cut IR 750 on your camera but it is one option. We know that this filter works extremely well for the RED camera but not at all for the Sony EX1. We do offer the B+W 486 filter which works for the EX1 but has a possibility of creating a blue cast on your images at full wide angle only.

I did have one customer buy our True-Cut IR 750 for his JVC 200 camera and did not return it or comment otherwise but that is not to say that it worked or didn't because I have never talked to him since (about 4 months ago).

I suggest you post images so we all can see. If you are certain of this IR contamination then give our True-Cut IR 750 a shot but with caution. The 486 has the highest likelihood of working but you have to avoid wide angle shots.

Ryan Avery
Schneider Optics

Thanks for your reply, Ryan. Below are some (unedited) frame grabs under various light conditions to illustrate what I mean.

- The interview shot using three 640W tungsten lights. The skin tone is off.
- The shot of the obelisk was during sun set.
- The capitol and brick stone building were shot in the late afternoon.
- The guy in the street was shot after noon (note the wall which should be white).

Do you think this could be due to IR leakage?

Thanks!

P.S.: Shaun, you are right, the JVC uses a CCD, not a CMOS sensor.

This is not to say this is your problem. I have run into a magenta artifact when there has been green spill from reflection or diffusion from vegetation in some of my shots when I whitebalanced to a white card in that environment then subsequently attempted to grade my colours.

You have trees in one of your locations. Did you manually whitebalance there or use the daylight preset?

I am prompted to this comment because the green in the trees appears to be suppressed.

I don't see any problems. shot of guy has white balance thats cool / cyanish... the other shots are a little warm, but look quite fine and certainly acceptable. I've noticed the camera can be on the warm side, but thats ok

people complain about the CA with the stock lens. I can show you $10k new "broadcast" SD glass thats worse. the stock lens is really pretty decent. its just that since people know what CA is and can spot, they freak out about nothing. its just the characteristics of the lens and its not that bad....

Hi Stefan...........
 

This kinda reminds me of a parent wondering whether that rash on their 3 YO is Meningitis (insert any other certain death issue here) and it turns out to be Nappy Rash.

Maybe my screen lies and maybe your audience all carry colourimeters to every viewing, but I'd suggest a bit less introspection on the colour front and a bit more "go do it".

Don't know where this stuff of yours is headed for final release, but if the Monument is a bit pink in the afternoon sun I can't see an audience getting up as one and exiting the cinema/ whatever.

I can see nothing wrong with any of the pics posted.

Geesh, you seen CSI Miami lately?

There isn't anything there remotely related to "real life" colour from the opening credits, and it's still up there at the top of the ratings.

By comparison, any isue with your pics is pretty small beer and can be put down to all sorts of oddities with time of day/ poor WB technique (the interviewee has a incandescent directly over his left shoulder - was that in frame for the WB or not?) etc etc etc.

Relax, you ain't pregnant.


CS

Alright, you guys convinced me. I guess I better stop worrying and shoot my movie :-)

Hi Stefan,

I couldn't leave this page without saying I for one think your concern is valid. Those screen grabs look seriously off to me. A camera has to be calibrated to a baseline reference. Assuming those grabs are as unprocessed as possible off the camera there no doubt exists an abnormal magenta shift. Why settle for starting with less than 100% or as close as you can get it when it's only going to create more work for you in post. If after factory defaulting your camera that's what you get I'd send it for service, assuming you don't have a color test chart and scope to first check it yourself. I wouldn't want the camera to create a pink obelisk for me, I'd want it to capture the obelisk accurately so I can make it pink in post if I choose to rather than have to take out the pink from every shot if I don't want it.

Stefan,

I definitely see what appears to be color shift. A lot of the warm color shift is caused by unblocked IR, but I also notice you referred to some Cyan and in the picture of the gentleman in the first picture it appeared to be a bit blue. If the lighting in the area is emitting any UV (and most lights do, especially fluorescent lamps) then it's possible your camera is picking up the UV-A or the near UV Blue that is close to 400nm.

If you do pick up a hot mirror, it should also include a UV blocking aspect to the filter. It would also be in your best interest to request a spectrograph of the hot mirror such that you can see what it is actually doing. Not all hot mirrors are the same and comparing spectrograph data is a very good way to see what you are getting for your investment. An example of a spectrograph for a hot mirror is <a href="http://zcrcoatings.com/heat-control/hh-vis-1600-hot-mirror/">here</a>.

I hope your video shooting went well!

Hmm... the example I posted apparently didn't come through. This should work.

Z C & R Coatings for Optics, Inc. - Precision Optical Coating HM-VS-1600 Hot Mirror