Help us understand spectral transmittance

[apendleton]

My boyfriend and I are trying to wrap our heads around some charts describing lighting gels used in theater, like these:

http://rosco.com/images/filters/roscolux/3315.jpg
http://rosco.com/images/filters/roscolux/4830.jpg

Each has a chart with a spectral transmittance curve on it, and also has a number for the overall transmittance of the gel, and we can't figure out how that number is calculated. In the first one, for example, the overall transmittance (at the top) is listed as 90%, but there's no single wavelength with a transmittance that high, so it doesn't seem to be the mean transmittance, or any other calculation based on the integral over the curve that we can come up with. Any help you folks could offer would be appreciated.

Oh, and we're not students (he's a theater lighting technician and I'm a web programmer), which is why we didn't post in the Homework section, though if this would be better asked there, feel free to move it; we're new here. Neither of us has a background in physics, but if you would prefer to answer by pointing us in the direction of reasonably laypeople-accessible books or other resources, feel free. We just haven't had much luck with Google or asking other theater people, so asking physicists seemed like a decent next step.

Thanks,
Andrew

 

[K^2]

Total transmission cannot exceed transmission on anyone wavelength, regardless of how they measure it. And since the 61% on the other chart is very close to what I'm getting by just averaging the numbers from the table, I would assume an error in the first chart.

If I understand correctly, you are looking at these for filters on stage lights. The exact effect will depend on both the lights and the exact pigments used for stage decorations. What exactly did you need to know?

Edit: Don't know if it helps, but I integrated over these with color-matching functions, and converted to RGB. That should give you some rough idea of what the light passed through these should be like.

R G B
151 167 120
187 112 132

Or #97A777 and #BB7084 in HTML respectively. These are VERY rough, but better than description from text on these charts.

 

[apendleton]

Thanks for your response. We're working on a database of information about gels from multiple vendors, each of which provide slightly different stats about the gels that they sell. Overall transmission is something that two of the three vendors we're including provide, but the third doesn't, and we were trying to figure out the process for calculating it from the curve so that we were consistent across all three. If averaging it is the way to go, though, perhaps I'll just do that and not worry about the fact that it's not the same as the official stats.

Andrew

 

[K^2]

Yeah, I'd go with that. If nothing else, it'd give you a consistent criterion across the three. Because even the two vendors that do provide their "total transmittance" aren't guaranteed to derive it the same way. One might use white-noise, the other incandescent light. That would give you different transmittance data.