Jump to content

IR Contamination...


Jim Jannard

Recommended Posts

Have you considered some sort of IR warning system?

You could have a function that detects a color balance shift between two different levels of ND, indicative of excessive IR "pollution", and post a suitable warning.

 

Keith... difficult trick but the boys continue to look at it.

 

My thought in the meantime was to do a White Balance with no filter and another with the ND in to see if there was a major change. The problem is you'd have to look at the whole image because some parts of the image may show no IR contamination (like a white polyester fabric) and other parts could show alot...like black cotton dyed fabrics. I still like the idea because it could also be used to check if your filter has a color bias in addition to being an IR filter.

 

One thing is certain... and tests have proven out, the new sensor and OLPF filter does a much better job with IR contamination. 1.2 ND should be no problem and 1.8 is significantly better than with the original sensor.

 

Jim

Edited by Jim Jannard
Link to comment
Share on other sites

  • Premium Member

In the meantime, perhaps using something like a "black balance" card. It would be made of a material black in both the visible and IR spectrums. Have someone hold it up against what you believe should be black in the coposition and eyeball the difference on the monitor.

 

W/ the newer touch screen cameras, maybe having the ability to isolate a small part of the composition and view its R, G, B vlaues, an RGB histogram, or a vecotorscope trace.

 

This would allow you to evaluate how the camera is seeing that portion of the composition which you believe should be black.

 

[i edited this post after thinking about the problem more.]

Link to comment
Share on other sites

  • Premium Member
Keith... difficult trick but the boys continue to look at it.

 

My thought in the meantime was to do a White Balance with no filter and another with the ND in to see if there was a major change. The problem is you'd have to look at the whole image because some parts of the image may show no IR contamination (like a white polyester fabric) and other parts could show alot...like black cotton dyed fabrics. I still like the idea because it could also be used to check if your filter has a color bias in addition to being an IR filter.

 

One thing is certain... and tests have proven out, the new sensor and OLPF filter does a much better job with IR contamination. 1.2 ND should be no problem and 1.8 is significantly better than with the original sensor.

 

Jim

It occurred to me that another way do this would be to turn the problem on its head: Put in a filter that is completely opaque to visible light, but passes infrared. That way you could roughly estimate the IR content by simply noting the amount you have to stop the iris down to get a “normal” (albeit weird-looking) B&W picture. 100% blackened colour film is almost completely transparent to IR, so interested persons could probably fashion an experimental filter out of some of that.

Link to comment
Share on other sites

  • Premium Member
Why not just get the new Formatt or Tiffen ND's that block IR? Problem solved, no? Compared with the other ideas, buying the right filters isn't such a big cost....

Yeah, but all of the different manufacturers' IR ND, IR, and hot mirror filters have different properties, so testing and picking the right one/combination is a pain. And then each rental house probably carries a different brand. So it would be great if we could just eliminate them altogether and make the AC's life easier. ;)

Edited by Satsuki Murashige
Link to comment
Share on other sites

Yeah, but all of the different manufacturers' IR ND, IR, and hot mirror filters have different properties, so testing and picking the right one/combination is a pain. And then each rental house probably carries a different brand. So it would be great if we could just eliminate them altogether and make the AC's life easier. ;)

 

Agree. Which is why we spent a lot of time to design a new OLPF that is better at IR blockage. We can confirm, as others now have, that the new sensor combo is much "safer" with NDs than the original sensor.

 

I have to say that we really had a lot to learn when we started this program. Many upgrades later... our program is much better than 3 years ago. RED is finally coming out of infancy. Many of the problems are being solved every day. While it is a bit premature to say that our program is mature... it certainly is much better. RED is not the same as yesterday. We still have many ways to improve, but I am comfortable that we are in a good place now. David's post seems to indicate that. Trust me when I tell you that we have worked very hard to improve every aspect of RED. And we haven't stopped just because we are getting a few good reviews.

 

This is a RED thread. It is about sharing info and suggestions. If we weren't interested to do better, we should be ignored. Our "cowboy" entrance was probably not the best way to begin, but I hope that our improvement over time on a regular basis will, at some point, earn real feedback and suggestions.

 

Jim

Link to comment
Share on other sites

  • Premium Member
Why not just get the new Formatt or Tiffen ND's that block IR? Problem solved, no? Compared with the other ideas, buying the right filters isn't such a big cost....

 

 

 

 

 

-- J.S.

 

Also, regardless of what the graphs show, the filters do change the coloration of visible light as well. The colors lose some pop. So if the camera can help you know if there is a problem before you take the cure, that would be nice.

 

BTW, I wonder if people are using black lens caps that block both IR and visible light when black calibrating their Red Ones? If not, it would seem that their black calibration could sometimes be off due to IR contamination.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...