Film (35 & 70mm) Light Levels

[Jim Cassedy]

It's been a looooong time since I've wanted/needed to take screen light levels for
film (35 & 70mm) - -and my reference materials are in a storage locker that's not
easy for me to get to. So, my question is: WHAT ARE THE PROPER FT-LAMBERT
LIGHT LEVELS FOR FILMSTUFF? Obviously, I want to be ready or OPPENHEIMER,
but I'm also slightly dissatisfied with the 35mm light levels at my venue, so I'd like to
come up with some quantifiable numbers that I can pass to management, so that
somebody just doesn't say "gee, It looks OK to me!"

(It's not that my current light level is 'horrible", and there haven't been any complaints
that I'm aware of, but I'm not 100% satisfied and I KNOW it could be a little better.
70mm shows, like LICORICE PIZZA, INHERENT VICE, etc have all 'looked OK to me"
(lol) but I think that's at least partly due to the much larger working aperture & lenses.
So, I'd like to come up with some numbers. THANXS!

[Steve Guttag]

16fL in the center, open gate (no film) projector running (+6/-4 fL). The meter has to integrate over the period of the shutter. In cinemas, the sides can be 75% of center.

For Review (screening) rooms (center is +/- 2fL) and the sides are to be not less than 80% of center.

The Reference document is SMPTE S196. RP98 discusses the measurement method.

[Marco Giustini]

I would stress Steve's point "the meter has to integrate over the period of the shutter" - I'm not sure what that means for meters designed for digital projectors but better to make sure the setting is correct when reading.

Basically the meter has to do some sort of time average - but not many "instant" averages.

[Jim Cassedy]

<edited> I'm not sure what that means for meters designed for digital projectors but
better to make sure the setting is correct when reading. Basically the meter has to do
some sort of time average - but not many "instant" averages.

Fortunately, I own both a fairly new digital meter AND one of those older analog "spectra-spot"
meters which was designed to be used for film. The digital is fairly new, and I had the Spectra
re-calibrated in 2021. It would be interesting to compare readings between the two if I have time.

[Steve Guttag]

The official integration range is supposed to be 24Hz to 72Hz. To know if your meter is integrating, I'd test on the lowest xenon light source you have with the shutter not running (make it a fast reading). Note that number (let's say 32fL. If you have a 50% shutter, then if it is integrating properly, it should read 16fL. Shutters vary in their dwell based on make/model. Simplex's XL is an 84-degree shutter so it is open a bit more than closed, for instance. So, if you were to read 32fL with the shutter fixed open, then when running it should be closer to 17.14. There are other variabilities in there since the shutter is not right up on the focal plane but that is what is meant by integrating over the period. It should produce the "average" light level for the entire period of the shutter rotation.

One would think, nowadays, one could create a test on a computer to simulate the shutter now that there are very fast refresh rates in the monitors themselves that making a 24-72Hz flash at whatever percentage open-time desired.

[Harold Hallikainen]

I think an insufficient integration time would result in an unstable reading. If the reading is stable, I think the integration time is sufficient. I had to deal with this in the LSS where I set the sensor to its maximum integration time and then averaged several samples.

[Stefan Scholz]

I would stress Steve's point "the meter has to integrate over the period of the shutter" - I'm not sure what that means for meters designed for digital projectors but better to make sure the setting is correct when reading.

Basically the meter has to do some sort of time average - but not many "instant" averages.

To perform these measurements, a specific camera system, the PSA 200, existed, which integrated correctly, with option to also "read" 3 bladed systems. It was also a great help in adjusting lamps and make the required documentation.

The original specification for screen brightness was 100 asb (approx 40 cd/m2 12 fl) with a drop not more, than 10% towards the edge in Europe. Actually tougher than the studio norm for 20% dropoff at 16 fl.
In many theatres I would have been happy to find only 25% dropoff to the edges.

The specification for Todd AO was 100 cd/m2 (30 fl) for a better color reception.
At 50 candles(the magic 16 fl) you're on the border to night vision, and the human color vision is still somewhat limited.
To double light level, makes sense, you're again away from flicker sensitivity, color reception is great, and 70 mm gate area allows way more light to pass through from the lamp, so it's easier to achieve higher brightness. You can find the doubled light requirement in multiple projection standard documents from all over the world of the time.

Meters must do a correct integration over the shutter cycle, as mentioned by others before. I'm not sure, and have never tested against the film measurement kit, if the color spectrometers used in digital calibration do that.