A correlated color temperature is basically a line tangent to the blackbody curve, and that curve moves from reddish (low temperatures) colors to more bluish colors (high temperatures). The subtlety is rather obscure, but shifted with a slight change in planck's constant, which meant that D65 no longer has a correlated color temperature of 6500K, but I believe it is 6504K. It's a situation where everyone can communicate, but the folks who really need to know understand what is really meant, even if 99% of the time the people saying it will never know the difference, or really ever need to know. Anyone who is actually calibrating anything hopefully knows that when people say 6500K, they mean D65. A spectroradiometer is like characterizing a complex instrumental note with a complete graph of frequency response.Ĭlick to expand.DK are used interchangeably, but they are not really the same thing. A spectroradiometer measures the exact physical spectra of light, and hence avoids metamerism (seeing different physical stimuli as the same color).

It reduces the full spectrum of infinite frequencies of light into only three values (this would be like characterizing a complex instrumental note as only quantities of bass, midrange, and treble). In other words, a tristimulus colorimeter measures an approximation of what color an average human sees. A spectroradiometer is a much more objective measurement device that can plot the exact spectral power distribution of any stimulus, which can then be characterized using human vision curves or any other viewing curve you'd like (bugs, dogs, whatever). A tristimulus meter (slightly a misleading label since they almost always use four sensing elements, not three) is in comparison a much more rudimentary tool which is also designed primarily to mimick human vision. For more difficult sources, particularly with very peaky SPDs (like LEDs, for example, which in theory are close to monochromatic), the inaccuracies of the response curves of a tristimulus meter become much more problematic (they don't kind of average away as much as they do when measuring light that has much smoother spectrum), and using a spectroradiometer is increasingly important (and also orders of magnitude more expensive). Phosphor-based displays like CRTs and plasmas have fairly even SPD curves which are relatively easy for a tristimulus colorimeter to measure with good accuracy. Click to expand.There are various measurement tools with varying degrees of accuracy, and that accuracy also will sort of vary depending on the particular SPDs (spectral power distribution) of what they're measuring.

Avs free tv calibration disc