Originally Posted by crakarjax
Any digital source would have a spikey PSD, only analog would be smoothe. Pixels are discrete RGB sources, after all.
Not entirely true. Even an analog CRT is going to be a spikey source, as it still uses descrete R, G, and B phosphors...
I'm unsure as to how this can skew calibration though. Is it just a matter of compensating for the spikiness of the Power Spectral Density? If so, that could be fixed in their software.
That's what you do when you calibrate the probe. You store in the software a matrix of values used to convert whatever readings you get... the more specific you get with the display you calibrate on to the one you're measuring, the more accurate these values will be. Each MFR is free to use and change their filters and phosphors, and may change from time to time... I'm simply unsure if doing this once per generic display technology (CRT, Plasma, LCD, LED-LCD, etc..., and going brand-specific could get you even closer) gets you as close as consumer grade COTS units, or if those somehow can accurately compensate more generically... there's also an issue of refresh rates, but that is why the HCFR software has you choose the display tech for certain COTS probes...
Is it open source? If so I might take a peek and see if I can do something with it.
Not sure. Google "HCFR" and you'll come up with the website, which has an English version... also, Patrice (laric) and Dave (dlarsen) (who I think are somewhat involved, and at any rate certainly know infinitely more than I) show up here once in a while...