Red or infrared?

[Casey Sanders]

We have one set of 35mm projectors with red-light sound readers, one with infrared. I have been reading up on which types of optical soundtracks to funnel to each.

I understand that cyan-dye tracks only play properly through a red-light reader.

I have read sweeping statements regarding red light and variable density tracks — that red light does not properly read them. However I’ve also read more specific anecdotes which indicate that it’s specifically hi-magenta VD tracks which are the problem for red light. A lot of what I’ve read about this lives in the archived forum, but the discourse seems to have taken place over many years and throughout the industry shift to red-light readers, so there’s a lot of speculation and uncertainty mixed in. Is there a consensus about this today?

I also have read about soundtracks prior to the early-2000s being optimized for white/infrared light vs. newer prints being optimized for red. Is there a “cutoff” date to keep in mind when choosing whether a print might sound better through a red or an infrared reader?

If anyone has any insight, rules of thumb, etc. I’d appreciate any tips.

(FWIW the red-light readers are capable of digital sound as well, so we funnel a lot of prints with SRD to those for that reason.)

[Ryan Gallagher]

Are you at an archive, museum, or university where you are asked to show a lot of the older types of prints like VD?

[Steve Guttag]

All non-cyan or high-magenta tracks will have lower distortion on the infra-red readers. The red light will get you "crossmod distortion." It sounds a bit light a distorting amplifier. It is a harsh type of distortion/cracking. How bad it is varies with the quality of the printing process.

High Magenta still used the silver doping but just used the magenta layer of the soundtrack. This was a compromise track. It lowered the quality for white light and infrared while knocking down the distortion for red readers.

For cyan tracks, only red readers can be used. infrared and white light pass through the cyan track. Cyan depends on an exact complimentary wavelength for its opacity. However, that doesn't change the solar cell itself, which remains more sensitive to infrared than visible red wavelengths. As such, red light readers needed the extra pre-amp (more noise) to work with typical optical preamps designed for white-light soundheads.

While some, I'm sure, will still sing the praises of the whole cyan debacle. It was nothing more than a last gasp attempt to make film seem more environmentally friendly by eliminating a nasty process in the development process. Mind you, they only applied this to 35mm. Any lab supporting 16mm sound prints still must support the original yellow/cyan/doping process. Has anyone tried to source the required wavelength LEDs with the proper distribution pattern lately? They are unavailable except by what someone may have stockpiled. Our industry is too small for anyone to keep them in production, particularly now that people are not going to consume them at the rates they once did. It was a bad idea, in my opinion to have ever gone down the path, particularly in the last 5-10% of film dominance. It would have been one thing had it happened in the 1950s/1960s...but in late 1990s/2000s? In 1999 we had the first DCinema showings with Star Wars EP1...anyone that saw one of those knew it wasn't there yet but it was FAR closer than anyone suspected.

So, now, we have to contend with with two types of tracks...cyan and the prior 80 or so years. I have some screening rooms where we have 4 projectors. Two with red readers and two with infrared readers...just to optimize the results. One nice thing about IR readers is that their level barely fades. I haven't had to turn them up in over 15-years.

[Carsten Kurz]

There is also a difference between red laser and red led readers.

[Steve Guttag]

The laser readers I dealt with were nasty. I want to say that Earnamen used them for analog (but red LED for Dolby Digital). I want to say they were riddled with noise and a less-than-optimal phase response. I seem to recall seeing artifacts when I was doing A-chains on them.

Mind you, IR LEDs are not without their short comings. They are very collimated in their light...which makes the shadows caused by scratches more pronounced. I had some luck using 3M "Magic Tape" to make them more diffuse (and compensate by turning the LED up some). That could indeed knock down that sort of noise.

[Marco Giustini]

My limited experience with laser readers were also riddled with noise. At least on one system I replaced them lasers with Cinemeccanica Red LEDs and the difference was night and day.

[Juergen Kellermann]

Has anyone tried to source the required wavelength LEDs with the proper distribution pattern lately? They are unavailable except by what someone may have stockpiled. Our industry is too small for anyone to keep them in production, particularly now that people are not going to consume them at the rates they once did.

So does someone know, if there are red LEDs available anywhere, that can be used to convert a soundhead to red light? What would be the exact wavelength required? What would be the next best LED?

Someone also suggested to me to put a red filter in front of the (white light) exiter lamp - would that actually work?

[Casey Sanders]

Are you at an archive, museum, or university where you are asked to show a lot of the older types of prints like VD?

Ryan: yeah, museum/archive. A comparatively small but fair amount of prints with VD come through our booth. We deal with a somewhat diverse array of material.

There is also a difference between red laser and red led readers.

Carsten: I am not familiar with laser in this context so thank you for making me aware of it. But we have red LED.

All non-cyan or high-magenta tracks will have lower distortion on the infra-red readers

Steve: this is regardless of VA vs. VD? I’ve seen VD mentioned as a bigger problem for red light, but maybe that is misinformation?

I have some screening rooms where we have 4 projectors. Two with red readers and two with infrared readers...just to optimize the results.

Do you know if these rooms by default put all prints with NON-cyan optical tracks on the projectors with IR readers, then?

IR LEDs are not without their short comings. They are very collimated in their light...which makes the shadows caused by scratches more pronounced.

I’ve read that there were early reverse-scan readers that used an exciter lamp rather than LED. Would that theoretically provide the benefits of reverse scan while avoiding the pitfalls of collimated light?

Thank you for all of the detailed info, Steve.

Relatedly: do all digital readers use red light?

[Josh Jones]

660nM 1W led's are available from china cheap, and could likely be retrofitted to an existing reverse scan kit. Some gymnastics would be involved to spread the light across the track, but it should be possible with some effort. I tested a sample against a real reverse scan reader LED and couldn't really see a difference in contrast between the two.

The Dolby cat700 penthouses, and some early basement readers used white light. The SRD tracks were always dye only, but black I believe.

[Gordon McLeod]

The Spanish company SAX used to make red readers for many euromachines and they used a dual element LED We installed several on the Wassman Eurokino 35/70 machines and they seemed to work well

[Stefan Scholz]

A lot of crossmod distortions has to do with improper loading of the photodiodes used for conversion. A zero ohm input is required to minimize those, but in reality voltage amplifiers with low iinput impedance were used.
Actually, you might say, eventually you don't need a red light source, and you can keep you incandescent bulb with it's broadband IR pickup. Nearly all movies employing a Cyan track actually carry a "digital backup" in one to three digital track formats. These can be used to screen the movie. And those films that do not have a cyan track, they are IR lamp compatible.

Laser are special. The slit width is extremely small on industrial laser creating a sharp rectangular pattern. The intensity on a small portion of the converter diode surface is extremely high, and often beyond the photoelectric capacity for linear operation. I assume the one used on the Ernamann 15 Raytheon projectors is different, than the one I used since 1990, where my brother, at that time a young laser professional after graduation, suggested the use of these tremendously expensive laser optics projecting a sharp rectangular pattern (slit). The laser at Raytheon wasn't intended to replace the light bulb. There wasn't even serious discussion on silver free tracks at the time. It was intended as a measure to bring projector build cost down, as the HQ Zeiss slit lens they were using for decades meanwhile started to cost in excess of 1800 bucks each.
And not to mention, there were cyan colored ferrocyanide tracks in war time. An idea to save silver on color prints. These tracks were metal based, an IR compatible.

[Josh Jones]

By crossmod, are we really referring to intermodulation distortion?