Originally Posted by Wizziwig
Those are some impressive results indeed. Have you ever verified the calibration at the start of a movie and then again at the end? I found that on the RS40, it would drift quite a lot over time. I'm not sure if it was caused by the bulb or panels warming up. Made it hard to calibrate - you can't get accurate colors for the entire duration of a typical 2 hour movie.
Good point. It's not only the rs40, it's all UHP projectors. If you care about accuracy, you do need to warm up the projector for at least one hour before calibrating (watch some TV or play a movie for the kids) otherwise the lamp/panels drift too much to get consistant results. Two hours is better, as that is the point where the drifting becomes insignificant, 30 minutes is the minimum if you are in a rush. Same thing before watching a movie, although 30mn to one hour warm-up is enough for critical watching as most of the drifting will be done by then. But again, if you're very picky or are doing a demo to an expert friend, two hours warm-up time is necessary for both calibration and watching.
This drifting mostly affects greyscale, but gamut too (as getting 100% white to D65 has a huge effect on the gamut, so should be done before any gamut work). One simple way to check for this is to display your RGB balance at 70% or 100% white (avoid 80%-95% as these levels are not always very stable on JVCs anyway), in continuous measurements mode, right after you switch the unit on. Balance it perfectly. Then you'll see as it changes over the first hour. Balance it every 15mn or so, and when it stops moving significantly, you'll know how much warming up time you need before calibrating/critical viewing/checking a calibration.
It's also not a bad idea to leave your meters connected for at least 15mn before calibrating, especially the i1pro/i1pro2 need that time before doing a dark calibration (or even more important, before using them to profile a colorimeter) or they tend to drift too as the electronics warm up and they adjust to ambient temp, especially when measuring low light levels. Then of course you have to take regular dark calibrations with these when the software tells you to.
It's also very important to look for the room temperature if your room doesn't have air conditioning. With the heat from the projector (after all, a JVC is like a 250W+ heater), especially in summer, the ambient temperature rises in the room - and therefore in the projector - and this affects the color temp of the lamp. In some cases it can lead the greyscale to drift way past two hours, until your room temp gets stable and the lamp temp gets stable (which might never happen).
If you don't check for all this, your calibration is impaired and you never watch a movie with a calibrated picture
. You still get say 80% of the benefit of a good calibration, but you do get significant and visible differences when watching a movie as the picture drifts between the beginning and the end.
That's why good calibrators often spend a significant amount of time with you watching clips before calibrating (or doing stuff they know they'll check again later). It's not only to asses the performance of the projector "before", it's also to warm the projector up (and their tools, and the room) before starting the calibration. Otherwise you can never reproduce these nice looking little graphs...
You also need to touch up greyscale and gamma at least every 200 hours on these projectors if you calibrate yourself and want an accurate picture, although since last year and the new lamp they seem to be much more stable over time.
Now with autocal, people just plug the (usually not so accurate) meter in, switch the projector on, run the autocal, get the nice looking graphs, and think they have an accurate picture