Welcome to our community

Be a part of something great, join today!

  • 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

Thank you all for the amazing info on this thread. I overlooked this thread for too long and after reading all of the posts realized how big an issue this is. All of your comments have been a huge help for me in understanding the problem and how to correct for it. One other question that I have pertains to ND grad's. How will the IR issue play out when using grad filters?
 
I ordered a custom 4x5.65" cut of their current hot mirror which cuts IR to around 925nm. The RED One is sensitive to around 1200nm so there is some room for an improved dichroic formulation.

264_1204995955.jpg


Rosco's common dichro glass is borasilicate (like picture glass) and chips a bit too easily in the matte box filter holder so something harder is needed.

I spent Friday morning with Henry Cowen at the Round Rock, Texas plant that produces all of Rosco's dichroic glass (as well as steel and glass gobo patterns) and we discussed the development of IR filters with one or more of the following characteristics:

Development priority list
1.Block IR to 1200nm
2.3.3mm thickness
3.Harder Glass than Borosilicate glass
4.Dressing of edges and corners or develop a ‘bezel’ to finish the edges of the glass
5.Labeling of filter (possible laser etch of bezel)
6.Flat and Parallel surface treatment

I'll be having a discussion with Joel Svendsen early next week about what market exists for Rosco in the optical glass world. My first impression is that there is no reason to compete directly with Tiffen and Schneider if they are prepared to bring IR solutions to market. There may be, however, an opportunity to bring a lower-priced filter to market that includes some, but not all, of the above features.

As I noted in my visual observation of the borasilicate glass I used this week, there was no reflection perturbation of fluorescent light when observed at twenty feet from the glass. (Of course, this is hardly scientific) When I consider the horrors imposed on the image by lens optics, I feel the over engineering of Schneider filters to be largely unnecessary (even though I have them in my matte box right now.) Would a lesser piece of glass be flat enough for most shooters?

Michael

Very good to know. I am calling my Rosco rep tomorrow to get a hot mirror "for now". What you describe is an excellent idea...tell Rosco...3000+ cameras and counting...all needing hot mirrors!

Thanks Michael.

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
Edmund Scientific has a 4x5" hot mirror for $189 that would also do the job. Not sure what glass that is made with. I believe we can bring the Rosco version in for less.

I was just talking to my Rosco rep. My only concern is that this glass is meant for putting in front of architectural displays. It is not "optically" the ideal glass!

I am likely doing a shoot where we are going to great expense to get quality glass (master primes, cooke S4's, etc). What is the point of having all that expensive glass if you are going to throw a "not so optically clear" piece of glass in front of it all.

This particular "hot mirror" solves one problem and creates another.

Hopefully Schneider or Tiffen will step up and offer "optically clear" hot mirrors in useful sizes and thicknesses...

Still no word on the mirrored Panchros huh?

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
I was just talking to my Rosco rep. My only concern is that this glass is meant for putting in front of architectural displays. It is not "optically" the ideal glass!

I would normally agree. My thinking at the time was that, to the best of my knowledge, Edmund optics uses glass supplied by Schott for most if not all their optics. In this case I was right.
I called one of the engineers at Edmund optics and he tells me that the glass is quite adequate. I asked him if the glass is commercial grade and he said yes. However, he mentioned there 'might' be a slight change in color temperature. He didn't seem too sure about the color temp thing so it might be worthy of an experiment. I also went to the Schott website and "optical filtration" is on their list of recommended applications for BOROFLOAT (the glass used to make these filters).
The question now is, does the IR coating create a color shift or not?

BTW, Dylan. I checked out some of your reels. Nice work!
 
I've never seen a decent Hotmirror filter designed for cameras that caused any color temp issues, beyond what I've already mentioned earlier about Cyan shift in the outer corners of the frame when using wide lenses.

Still trying to track down a set of the mirrored ND's, without buying a set.
 
Dylan, your concerns are valid, to a point... The borosilicate float glass used by Edmund and Rosco has not been ground flat like that offered by Schneider et al. However, I observed zero deflection in a 40' reflection of a fluorescent light in my shop. (Read, I stood 20' away from the IR glass in a matte box and took a picture of the reflection of the light hanging over my head.) Is that flat enough? Each of us must decide.

Do we have an IR solution from Schneider, Tiffen, etc. now? Not that I know of. The present Rosco solution may serve until then. It certainly allowed me to shoot now.

I used the Rosco hot mirror for a four-day shoot last week (left it in my matte box) and noticed zero effect on the image other than cutting the IR during heavy ND filtration. If you want to point your camera at a test chart and find the minuscule deflection that it may produce, by all means, do so. But, will you notice that deflection in your creative images? I think not. You'll discover far more sins created by the lens before you discover a problem with the IR glass.

That said, I'm working with Rosco to produce a better solution than presently offered. We'll see.

Michael
 
I would normally agree. My thinking at the time was that, to the best of my knowledge, Edmund optics uses glass supplied by Schott for most if not all their optics. In this case I was right.
I called one of the engineers at Edmund optics and he tells me that the glass is quite adequate. I asked him if the glass is commercial grade and he said yes. However, he mentioned there 'might' be a slight change in color temperature. He didn't seem too sure about the color temp thing so it might be worthy of an experiment. I also went to the Schott website and "optical filtration" is on their list of recommended applications for BOROFLOAT (the glass used to make these filters).
The question now is, does the IR coating create a color shift or not?

BTW, Dylan. I checked out some of your reels. Nice work!

I think a slight "shift" in color temp is not a big issue. This can be easily corrected. I am more worried about clarity.

I have a shoot coming up where we may be lucky enough to do a 4K finish and even project in 4K as a sort of "showcase" for a digital version of what IMAX does. Picture clarity is of the utmost priority...this ain't no youtube viral!

We are talking about large vistas on sunny days, so I am going to need it all...no IR contamination, no UV and no image degradation from a piece of glass whose clarity is unknown - i.e. the hot mirror.

I am hoping to be able to test the mirrored Panchros this week.

I am still going to get a hot mirror...don't get me wrong...I just want it all!

Thanks for the props Jeff...new reel coming soon!

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
I thought I'd drop a quick note in this thread. As many of you know, I optically white balance my Andromeda camera... but I've found some similar IR pollution going on when I go full bore and filter for shooting under tungsten... and none of my filters are NDs... I don't have issues with daylight filtering though, which requires less light loss.
 
quick pancro test on X98

quick pancro test on X98

I ran 2 quick tests earlier this month, but by my eye I can't see IR contamination. Here's the 1st test:

March 10, 2008 Red X98
ND 9 Schneider vs ND 9 Pancro
Cooke S4 50mm@T4
fps 23.98 1/48 shutter ISO 320
530 fc
Black 8 1/3 EV
Grey 11 1/3 EV
White 13 EV
Light source HMI
 
2nd test

2nd test

Here's the 2nd test:

March 21, 2008
ND 12 Schneider vs ND 12 Pancro
Zeiss 85mm@T5.6
fps 23.98 1/48 shutter ISO 320
3000 fc/2900 fc
didn't take spot meter readings
Light source Daylight
White Balance manually set to 5600K

These stills haven't been tweaked in Red Alert or Redcine. The only thing I can discern is that Pancros warm the picture up and lower the contrast. I also noticed that shooting at 23.98@1/48 is not safe under HMIs. There is a slight rolling shutter effect across the picture. 23.98@1/60 shutter if safe though.

Looking forward to your results Dylan.
 
Here's the 2nd test:

March 21, 2008
ND 12 Schneider vs ND 12 Pancro
Zeiss 85mm@T5.6
fps 23.98 1/48 shutter ISO 320
3000 fc/2900 fc
didn't take spot meter readings
Light source Daylight
White Balance manually set to 5600K

These stills haven't been tweaked in Red Alert or Redcine. The only thing I can discern is that Pancros warm the picture up and lower the contrast. I also noticed that shooting at 23.98@1/48 is not safe under HMIs. There is a slight rolling shutter effect across the picture. 23.98@1/60 shutter if safe though.

Looking forward to your results Dylan.


Just got back from shooting the test.

Don't have anything to post, as I just got in the door.

We were not able to reproduce the IR contamination either? Mind you it was fairly overcast. We could see and feel the sun through a light layer of cloud, but there was no IR colour contamination visible.

We used a Harrison ND 1.2 and compared it to a Panchro mirrored ND 1.2. Then did a third exposure with no ND and we compensated on the "shutter".

Agreed that the Panchros look warmer and less contrasty. But in discussing Panchro mirrored ND's with the much wiser and experienced Rene Ohashi, CSC, ASC, I do believe that the opposite is true; that "non-mirrored" ND's create more contrast and introduce a colour shift.

Because the mirrored ND's are "reflecting" light to reduce it, there is no dye in the glass to introduce a potential colour shift. I have no idea why there might be more contrast with a heavy ND...but it seems to be true.

Crewpix, my trusty partner in crime today, suggested that perhaps Infrared appears more strongly at different times of day. Perhaps Mike P and Michael M can comment on what time of day they shot their tests at.

Just re-reading this post and I am now kicking myself for not taking Crewpix's advice of trying to simulate it back at the studio with a 1K.

I have to return the borrowed Panchro mirrored ND's tomorrow, but still really think they may be a cure.

I think the way to do it would be how Mike did his "studio test". Then try introducing a Panchro mirrored ND.

Any takers?

Dylan Macleod
Cinematographer
Toronto, Canada
www.dylanmacleod.com
 
A 1k tungsten openface unit is what I used for the Sweatshirt tests, it puts out a TON of IR.

In the first shot that got me thinking about this, the sun was setting early in it's late winter fashion. I can't say anything about why it would be worse other than it was direct sun at a steep angle.

Gregor's Tests with an HMI and a cloudy day certainly aren't super IR rich environments, BUT they are good real world tests.
 
Agreed that the Panchros look warmer and less contrasty. But in discussing Panchro mirrored ND's with the much wiser and experienced Rene Ohashi, CSC, ASC, I do believe that the opposite is true; that "non-mirrored" ND's create more contrast and introduce a colour shift.

Because the mirrored ND's are "reflecting" light to reduce it, there is no dye in the glass to introduce a potential colour shift. I have no idea why there might be more contrast with a heavy ND...but it seems to be true.

The Pancro ND9 is only subtley warmer then the Schneider, but the Pancro ND12 is significantly warmer then the Schneider. This makes me suspicious of the Pancro, but these are filters from the 80s so who knows how well mirrored glass holds up over that much time. I understand the theory of Pancros advantage over regular NDs, but I have to base my use of them on real world tests. I don't doubt that they reflect IR like they're supposed to (Doug Koch, CSC spoke at length with me about his need to switch to Pancros when he was shooting IR film quite heavily in the 80s and 90s), but maybe the OLPF does a good enough job of blocking IR for Daylight shooting or maybe there is a QC issue with some OLPFs? I've never been in a situation where I've had to use NDs under Tungsten lights (except when testing latitude and I use NDs to bring down exposure instead of changing the iris).

Are there any 23.98 HMI flicker tests out there? I only have anecdotal evidence that this is a problem.
 
Are there any 23.98 HMI flicker tests out there? I only have anecdotal evidence that this is a problem.
Yes, and it doesn't flicker -- it bands. You get scrolling bands of orange in your image if you're using a magnetic-ballast HMI or fluorescent; the brightness/darkness of the bands depends on the shutter speed, and the higher the shutter speed the more noticeable the bands are. You have to stick with a shutter speed that's a multiple of 120 (1/120, 1/60, 1/40, 1/30 etc) to avoid it. Or, go with a high-frequency electronic ballast; those avoid banding (although there can come in some flickering at 1/1000 or 1/2000).
 
Back
Top