AVS Forum Special Member
Join Date: May 2008
Location: San Francisco - East Bay area
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 188 Post(s)
The thing to keep in mind is that a spectroradiometer just by the nature of how it works, can measure ANY light source accurately as long as it is within the luminance range the meter can handle. Even though the light spectra delivered by plasma, CCFL LCD, LED LCD, and various projection lamps are very different, a spectroradiometer can read all of them accurately.
Lower-cost (and some moderately expensive) meters use combinations of 3 or 4 up to maybe 10 very specifically selected filters. This type of meter can be made accurate for one very specific light spectrum... that means you could make a colorimeter accurate for plasma, but because CCFL or LED light sources in LCDs produce very different combinations of light (primarily red, green, and blue, but there's variations in the amount of overlap (broadness or narrowness) and amplitude of each color when light sources are different. Those differences cause different readings from colorimeters... for example, if you had a "perfect" plasma display with zero errors and a similar "zero error" LCD panel, a spectroradiometer would produce the same readings for both displays, but a colorimeter would produce different readings for the 2 displays. To make the colorimeter measure the "different" display accurately, you have to provide some sort of correction. And that correction helps the colorimeter compensate for the very likely different light spectra emitted by, say, CCFL or LED-lit LCD displays compared to, say, plasma displays.
If you poke around online, you can find some fairly graphic examples of light spectra emitted by different types of video displays or projectors (and projection lamps). A laser display with red, green, and blue lasers emits only 3 frequencies of light... one for each color. Projection lamps, CCFLs, and LED produce a huge number of different frequencies of light and not all of them have equal amplitude. It's tempting to call the number of frequencies infinite because you can just keep adding decimal places to each discrete frequency to get a slightly different frequency. You could, hypothetically, produce a colorimeter with 4 different filter sets in it... one for plasma, one for CCFL, one for LED, and one for projection lamps and select each filter set as needed. That would be closer, but still wouldn't compensate for variations from display-to-display even if they have the same type of light source. And there are many "flavors" of projection lamps so to be REALLY accurate, even if a colorimeter had multiple filter sets, it could still be made to read better via characterization against a known-good spectro device.
"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA