Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Red and the infraRED

J. Beale, I'm confused. What test produces a broad-spectra spectragraph from point-spectra sensors (granted with overlapping spectral ranges) and three-point color data? How does this work?

Here's the sample snapshots I mentioned in a previous post:

1.8ND/F, no adjustments, ~6760K
264_1204859629.jpg


1.8ND/F2.8, attempt at color correction
264_1204859649.jpg


.3ND/F11 no adjustments
264_1204859675.jpg


.3ND/F11, 1.2 saturation + light s-curve
264_1204859698.jpg
 
Michael Morlan said:
J. Beale, I'm confused. What test produces a broad-spectra spectragraph from point-spectra sensors (granted with overlapping spectral ranges) and three-point color data? How does this work?
Click to expand...

I'm not sure I understand the question but I'll try to answer... if you're asking how I got the spectral-response pictures on this page, http://www.bealecorner.org/red/test-080108/page2.html#Test_6 I can tell you.

That is one frame of a normal image from the Red, while it is looking into a custom-built spectroscope. I designed and built the spectroscope for that test. It uses a broadband light source, a slit, an enclosed optical path with light baffles, and a high-efficiency diffraction grating of 19,050 lines per inch (750 lines per mm). You may have built something similar in 6th grade science class using a cardboard box with one of the Edmund Scientific gratings. You look into it and see a rainbow of color. This is basically the same thing, just bigger.

The diffraction grating spreads out white light from the slit into a rainbow of colors. Exactly where which wavelengths appear in the image depend on the lens focal length, but they can be calibrated by using a secondary discrete-line source. I masked off the continuous-spectra light from part of the slit, and put a fluorescent bulb at the very bottom of the slit. Fluoro light bulbs have discrete and well-known wavelength peaks which serve as convenient markers to calibrate the spectrum image. (Just to be sure I also checked calibration using solar Fraunhofer lines... the flouro peaks are accurate.)
 
Heat absorbing glass?

Heat absorbing glass?

Someone might try Heat absorbing glass in combo with ND to cut the far Red and IR. Heat absorbing glass does not have so many angle problems as an interference filter.

Some kinds of Heat absorbing glass cut into the red around 650nm so may cut the red exposure of the blue pixels. The red leaking into the Blue Bayer filters looks very bad, I would have hoped that that would not start until 700nm, RED TEAM should look into getting a better dye for the blue filter on the sensor. The blue filter trys to have high transmission to get a high EI, but seems to let too much red through.

Here is the curve for Schott KG5 glass,

http://www.optical-filters.com/kg5.html

One problem with Heat absorbing glass is that it may not be ground optically flat, and so may work better on wide angle lenses than long lenses.

Lower cost Heat absorbing glass does not cut the near IR as well and has more of a green cast.

Putting the IR mirror behind a retrofocus very wide angle lens should work better than in front since the angle of the light is smaller. High speed lenses can make a wider angle on the IR mirror, so it would transmit more IR in the corners, so when you use NDs and open up the IR filter may transmit more IR in the corners.

It might be possable to make a concave IR mirror (behind the lens) to reduce the wide angle problem, maybe coat the front element of wide angle lenses with an IR mirror if they have a convex front element.

I have some Heat absorbing glass, if a RED owner in the Bay Area would like to try some Heat absorbing glass with NDs I might be able to bring my glass collection over, let me know...
 
jbeale said:
That is one frame of a normal image from the Red, while it is looking into a custom-built spectroscope. I designed and built the spectroscope for that test.
Click to expand...

Ah ha! I was under the impression you had somehow generated the broadband spectral chart from a digital frame. I couldn't figure out how that could be done. You shot a picture of a spectragraph, not generated a spectragraph from a picture. Duh!

Thanks,

Michael
 
Just for fun, I've ordered a 4x5.65 cut of Rosco hot mirror. $53.

264_1204995955.jpg


I know it can't be optically flat like a Schneider but it may well serve for a short film coming up soon. I don't have time to wait for optical filter manufacturers to ramp up production.

Michael
 
Has anyone considered going the opposite direction with this issue by slapping some wrattan 88, 88c or 89b etc... on the camera? Does the built in IR filter take out too much to make this course of action useful?

I'm a huge fan of IR photography and would love to have this option on the Red.

I'll probablly test this myself when I get my hands on a Red, but if someone gets the chance to do it before me, I'd love to see the results.

-Rob Kraetsch
 
Okay so I was wondering yesterday. The still lense industry has obviously worried about this before (just see all the round screw on 486 filters at B&H). But why has the film industry never fixed this? Does film not have this problem? What about F900 or Viper? Surely they've gotten into heavy NDs and theyn wondered "Why does the black shirt look purple??"
 
Wow Mike, really great post.

My contrarian artist instincts tell me that Rob K and whoever it was who suggested using the RED for IR photography have a cool idea here. I'm thinking of the haunting IR in I AM CUBA and that "night" ride in THE WILD ONE.
 
Rob_K said:
Has anyone considered going the opposite direction with this issue by slapping some wrattan 88, 88c or 89b etc... on the camera? Does the built in IR filter take out too much to make this course of action useful?

I'm a huge fan of IR photography and would love to have this option on the Red.

I'll probablly test this myself when I get my hands on a Red, but if someone gets the chance to do it before me, I'd love to see the results.

-Rob Kraetsch
Click to expand...

Hi Rob, I tossed on an R72 just to play around with that and you can see the results by jumping over to the InfraRED thread which has a link to some footage. I didn't go to a full on IR filter just so that I could get a little of the visible light to help with exposure time.

Jarred also teased with the info that he was experimenting with the OLP on a Red to find the optimum setup... there is no way that I'm dismantling my Red to get a little more IR to the sensor!
 
Shawn Nelson said:
But why has the film industry never fixed this? Does film not have this problem? What about F900 or Viper?
Click to expand...
Normal colour neg stock has no serious problems with IR that I've come across, but then with film stock you don't usually ND that heavily, as you simply switch out the stock for a lower ASA one - typically 50D for daytime exteriors. You generally want to use slower stocks anyway, as they exhibit less grain.

Some B+W stocks have very high red sensitivity (SFX200 - from the stills world - springs to mind) that will also be very IR sensitive, but then it's a different issue when you're dealing with monochrome images, and IR bleed can often be quite pleasant (it's one of the big selling points of the SFX stock, in fact, and most people shoot the stock with a full red filter to increase contrast and IR effects).

F-900's don't seem to have problems, at least when using the internal filter wheel - whether or not those internal filters are specially designed to reduce IR problems though I don't know. Never used a Viper, so can't comment there, either.
 
I'm just guessing, but I'd imagine the prism block alone would negate any IR issues for something like an F-900.

Corcuera, water tends to swallow IR light.

Clint, thanks for the link!
 
Prism blocks react differently to both IR and visible light because they use very different materials/processes to make them (since a bayer color filter is ~2000x smaller than an optical block filter).
 
Rob_K said:
I'm just guessing, but I'd imagine the prism block alone would negate any IR issues for something like an F-900.
Click to expand...
Yes, of course, that will be a factor - good thinking Rob!

So between that and Deanan's info, I guess that's just a growing pain for single CMOS sensors. The fact that Red doesn't employ an analogue gain that can be turned down is probably exacerbating the issue a little as well, at a guess.

Haven't heard stories of Genesis or Dalsa problems with IR though - can anyone comment on if they have similar issues or how they solve them if not?
 
Michael Morlan said:
Just for fun, I've ordered a 4x5.65 cut of Rosco hot mirror. $53.

264_1204995955.jpg


I know it can't be optically flat like a Schneider but it may well serve for a short film coming up soon. I don't have time to wait for optical filter manufacturers to ramp up production.

Michael
Click to expand...

Has anyone had a chance to test the Panchro mirrored ND's talked about earlier in this post...I've got a good feeling about it!

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
dmacleod said:
Has anyone had a chance to test the Panchro mirrored ND's talked about earlier in this post...I've got a good feeling about it!

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
Click to expand...

I just did a very quick test today with an ND9 Panchro vs. reg ND9 under HMI lighting at PS today. Mr. Koch had tipped me off earlier. I will post when I get the .r3ds tomorrow from Morning. Congrats on the nom.
 
You must log in or register to reply here.
Back
Top