Originally Posted by jrp
I was talking to Craig Rounds (a great calibrator and Lumagen expert out of Chicago) this weekend and he pointed out something he figured out from his calibration efforts. He found that sometimes if he calibrated with a 4k source active he would get inconsistent measurement results. Eventually he tested calibration by selecting an unconnected source (a non active source would act the same). When he did this he never got inconsistent measurement results.
We came to the conclusion that the Pro was taking longer to put up the pattern with an active source and Lightspace was taking the reading too soon. No surprise it took longer to write with an active source as there is a lot going on. It is unknown if this would apply to Calman or Chromapure. However, the below recommendation would be a good idea for any calibration. Craig could not find a setting in Lightspace to further delay the start of the reading (if my memory is correct), but based on Craig's testing Lumagen can recommend the following:
When calibrating select a non-active source input (turn it off or unplug the HDMI cable). This will insure the patterns are written to the screen as fast as possible.
Even if you can delay the start of the reading, not having the extra work for the microprocessor insures a (slightly) faster calibration. Note: This may not be needed in most cases but it will not hurt to do this.
[EDIT] Note that the time it takes from the pattern command to being on the screen is increased by the TV/Projectors internal processing time. It might make sense to have the TV/projector in its "Game Mode" if one is available. Then you would also watch leaving the TV/projector in Game Mode. Since Game Mode has lower latency I have recommended using the TV/projector Game Mode for years, but you need to make sure there is no image degradation in Game as some TVs and projectors have had.
This confirms what I have experienced too (but not reported yet because I wanted to do more test to be sure). As you know I have reported issues with no source active and test pattern, 121217 fixed it. But even in this configuration, I recommand to add an extra delay in Lightspace CMS (extra delay=time for delaying probe measurement in between 2 displayed patches). I have reported this to Light Illusion as well and they told me that they are thinking to add a feature to help finding the best optimized extra delay (I encourage lightspace cms user to ask for this feature).
With no source active, the output video condition matches with OTHER but, in the case there is no active source, sometimes OTHER does not trig the signal output expected (i.e. the last output video condition when the source was active is applied instead). Patrick told me that it was a strategy to avoid switching output signal but form my experience (I use my PRO as external TP for my calibration services) it would be better that, in the case no source is active, OTHER works the same way as for other output video condition. This would allows setting exactly the output signal you want to perform the calibration.
(by the way, maybe it is already the case, I will check with last firmware)
The workaround method with no source active is to use TP with user mode but the problem is as soon as a TP is no more displayed, there is a switch for output resolution based on the output video condition. This switch is very annoying (but normal) when you do a manual calibration (as said above, I use my PRO for my calibration service as external TP generator)