Joined
·
277 Posts
I want to have the tools to calibrate grayscale on my Sony 1272 CRT projector myself, but I am not willing to pay $400 for an ISF optical comparator or more for a color analyzer.
I'm planning on using the Guy Kuo method of getting accurate grey tracking by using a lens cap with a solar cell hooked up to a voltmeter, to ensure that the ratios of red, green, and blue stay constant as the amplitude changes.
But getting the reference 6500K calibration (the initial ratio of Red:Green:Blue for my phosphors) is the part that I'm thinking about now. The KeohiHDTV site has pointers to 6500K flourescent fixtures from Cinemaquest ( www.cinemaquestinc.com ) which look like good references. They appear to use Chromalux tubes, which seem to be the best at mimicing true daylight spectrum. They have a color rendering index, CRI, of 98 which is the best I have heard of. I am planning on getting the 18" unit for $54.95 adding some neutral density filters (I found 4 ND gel filters: 1, 2, 3, and 4 stops from "Lee Filters" on the www.filmtools.com site for about $20 total). The ND filters will be taped onto the light fixture in a stepped fashion so that the flourescent light becomes kind of a grayscale from light on one end to dark on the other end, allowing me to visually compare both high, medium and low light levels directly with the reference. I'll probably use the grayscale pattern available from the www.displaymate.com free area as the image on the projector.
The question that is coming up for me is whether I need to diffuse the light to make it more uniform, or whether just looking at the bulb directly (Through the clear cover of the cinemaquest fixture) is okay. And if I want to diffuse it, what material will be optically neutral (known not to transmit any color more than others)? I found a cheap $8 18" GE fixture at home depot which has a built in diffuser that looks good, and so I might use that instead of the cinemaquest fixture, but I'm not sure that the diffuser material is completely neutral; holding it up to the sun it looks just a little yellowish.
Any ideas about whether diffusion is needed and if so what kind works well?
-Tom
I'm planning on using the Guy Kuo method of getting accurate grey tracking by using a lens cap with a solar cell hooked up to a voltmeter, to ensure that the ratios of red, green, and blue stay constant as the amplitude changes.
But getting the reference 6500K calibration (the initial ratio of Red:Green:Blue for my phosphors) is the part that I'm thinking about now. The KeohiHDTV site has pointers to 6500K flourescent fixtures from Cinemaquest ( www.cinemaquestinc.com ) which look like good references. They appear to use Chromalux tubes, which seem to be the best at mimicing true daylight spectrum. They have a color rendering index, CRI, of 98 which is the best I have heard of. I am planning on getting the 18" unit for $54.95 adding some neutral density filters (I found 4 ND gel filters: 1, 2, 3, and 4 stops from "Lee Filters" on the www.filmtools.com site for about $20 total). The ND filters will be taped onto the light fixture in a stepped fashion so that the flourescent light becomes kind of a grayscale from light on one end to dark on the other end, allowing me to visually compare both high, medium and low light levels directly with the reference. I'll probably use the grayscale pattern available from the www.displaymate.com free area as the image on the projector.
The question that is coming up for me is whether I need to diffuse the light to make it more uniform, or whether just looking at the bulb directly (Through the clear cover of the cinemaquest fixture) is okay. And if I want to diffuse it, what material will be optically neutral (known not to transmit any color more than others)? I found a cheap $8 18" GE fixture at home depot which has a built in diffuser that looks good, and so I might use that instead of the cinemaquest fixture, but I'm not sure that the diffuser material is completely neutral; holding it up to the sun it looks just a little yellowish.
Any ideas about whether diffusion is needed and if so what kind works well?
-Tom
