View Full Version : some theoretical questions about brightness and fL


Heboil
08-25-09, 12:38 PM
I have been trying out a few different projectors recently and brightness has been a big issue (coming from an older IN76 that was kinda dim). I have read much on screen size and fL in a light controlled room. I have a few questions though.

Seeing as fL are measured when viewing a white screen (maybe I am wrong on the color), it takes the whole screen into effect. I have a 96" screen and viewing a full image produces said fL. Using good old math, if I use the square inches that I get for my image (3942 sq. in) and reduce that to a 2.35 image, I get 2979 sq. in. If I take that 2979 sq. in. and convert that to a 16x9 image, it is roughly 90 inches. If I take the Projector Central calculator and punch that number in, I get a higher fL calculation... because my screen is smaller and I pack the light into a smaller area. But in reality, I am not packing the light into a smaller area when I watch 2.35 material... I am merely cutting off roughly 25% of my screen in total from top and bottom. That would then reduce my fL by 25% when watching scope movies, no?

Following that logic, if I am "packing" the light into a smaller frame when you reduce the size of an image, are you not doing the opposite when you blow it up? Using the calculator, it states that you do. So can you take that logic one step further and state that when you take a DVD image and blow it up to 720 or 1080 to fill your screen, aren't you losing some fL on the blow-up? Less densely packed pixels give an opportunity to introduce something... is it less brightness? Or does the projector interpolate the image into other pixels of the same intensity (just in a digitized stair-stepped pattern near the original pixel), thereby NOT lowering fL?

The red ATTENTION screen at the beginning of Fox DVDs made me think of this. The BD version of that screen seems so much brighter using the same equipment.

Anyway, after doing a bunch of reading, research and trial and error, the fL seems like a good guide, but other than a guide, it doesn't seem to offer much (at least to me). The fL seems to vary based on the source (we'll see if others agree with me on this one) and definitely the aspect of the image. If you have lights on in the room the fL also needs to be different to achieve that desired goal (whatever it may be). I know that after trying a few different projectors, some were too bright (couldn't produce good enough blacks or I had eyestrain) but looked good with ambient light... and others looked great in total darkness, but literally paled with ambient light.

So it appears to me that fL calculators are good to let you get an idea of how bright an image is next to its competition, but unless you factor in the source, scope and lighting environment of your room into the equation, it doesn't mean that much on an individual basis. Ultimately, it is your eyes that need satisfying, not some spreadsheet calculation.

JOHNnDENVER
08-25-09, 01:25 PM
True that conclusion...

Ftl's are only good on the calculators to relate one projector to another.

Generally I shoot for 18 ftl's with new lamp and 16:9 image, this gives enough lee way to allow for other factors to lessen it.

Now, totally dark / light controlled is all I care about and all that should really be cared about if it's a dedicated theater.

jrwhite
08-25-09, 02:01 PM
Hi Helboil

Yes, you measure foot lamberts with a totally white field. You are measuring the maximum output from the projector reflected off the screen. Both the lumen output of the projector, and the gain of the screen material contribute to the brigtness of the reflected image.

If you are 'masking' your image to 2.35, you're losing brightness, as you're only using 75% of the projectors panel area, and thus only have 75% of the lumen output. Aside from the resolution benefit, this is why people use anamorphic lenses ... the projector ( or external processor ) scales ( vertically stretches ) the 2.35 image to fill the 16x9 panel, and the anamorphic lens scales ( horizontally stretches ) the image back to 2.35. This way, you're getting the full lumen output of the projector.

Scaling of the image ( upscaling from 480 to 1080 for example ) doesn't have an effect on lumen output, as long as the scaled image completely fills the panel. The number of 'pixels' in the panel don't change, just how the image is mapped onto them.

Footlamberts don't vary with a source device, as long as the source / projector are properly calibrated for gain and offset ( contrast and brightness ). This can be easily accomplished by eye using calibration disks like AVIA or even the THX Optimizer incuded with some movies. If you're using HDMI or DVI, you must also make sure the normal or extended range settings on the source and projector match.

Most projectors allow for individual calibration settings per input. Many also allow for individual calibrations on the same input based on detected resolution. Movies, however, can vary wildly .. even different editions of the same movie.

Another issue in choosing a projector is how it performs with different brightness settings. For example, I have a Mits HD1000 and Epson 400. The Mits in it's best mode is brighter than the Epson in it's best mode. The Epson, however, is much brighter in it's brightest mode compared to the Mits in it's brightest mode. In a dedicated theatre with a large screen, the Mits would be a better choice, as the enviornment ( ambient light ) is constant, so the Mits will give you the ability to use a larger screen with good colour accuracy compared with the Epson. However, in a mixed-use room, I find the Epson with a smaller screen to be better. For critical viewing you can black out the ambient light, or wait for nighttime, and have a picture as good or better than the Mits. In uncontrolled ambient light, you can put it into torch mode to fight ambient light, and live with less accurate colours ( non critical viewing like TV and Sports ). From my reading, the Infocus models tend to have the highest D65 calibrated brightness.

Hope this helps,

Jonathan