Originally Posted by Kris Deering
I agree that it shows the full 0-255 range but black is certainly not mapped to 0. If that were the case you would never see below black information and I can assure you I have test patterns that show that it will retain below black information all the way down to digital 1. This is not the case with Normal, which clips the information. No matter where I set the brightness control you cannot recover the below black information WHICH IS INTENDED TO BE USED IN A REFERENCE VIDEO SYSTEM.
I think there is confusion here. Some of it is probably terminology.
1st - We agree Normal clips information below video black (as I explained above). But then you say Normal is for PC signals??? PC signals have black at 0, so if Normal clips at 16 (or 15) you definitely don't want to use Normal for PC signals. Expand shows the entire 0-255 range, which is the PC range. Expand is obviously for PC signals, not Normal which would clip the PC signals way above black.
You say Expand is not set for black at 0. Well it is. Put in a signal at level 0 and where is the Brightness control? It's at 0 (or -1). In the Expand mode Black is set in the projector for a digital 0 signal. Put in a signal with black at 16 and you can see everything below video black. I think that is what you are doing and it proves that the projector black is at 0. If the projector black were at 16 you wouldn't see anything below 16 by definition.
All of these signals are available from our AccuPel HDG-4000 video generator so it is simple to verify.
2nd - You statement that below black information is "INTENDED TO BE USED IN A REFERENCE VIDEO SYSTEM" doesn't mean anything unless you tell me what it is intended to be used for. Below black (digital 1-15) in all video standards, ITU Rec 601, Rec 709, SMPTE 274M, SMPTE 296M, etc. is absolutely NOT intended to be displayed. That foot room is only included for processing intermediate video signals before display. It avoids clipping intermediate processing values that could then propagate down the video signal processing chain creating additional, larger processing errors. But at the end of the processing chain, signals below black are NOT intended to be displayed. Imagine how absolutely horrific the picture would look if you set the projectors black level to digital 0 for video signals, so you could see everything from digital 1-15??? That's obviously absurd. Talk about gray blacks and poor contrast.
Originally Posted by Kris Deering
I have a question for you now. When you calibrate a projector do you calibrate to preserve head and toe room? Also, what do you use to define the contrast setting? Are you bringing contrast up until you are just clipping above 100 IRE or are you preserving the full range of white? This would have serious consequences on measured contrast with results being much higher than if the projector was calibrated with head and toe room preserved and according to what every ISF calibrator I've ever worked with would recommend, including Joe Kane.
You don't set the black level of the projector above the black level of the video signal (to reserve toe/foot room) as I said above. That would obviously destroy the image quality.
Video signals go above reference white (digital 235 for video signals) for several reasons. Most importantly, it is difficult to anticipate the maximum incoming brightness when capturing live video, or even when converting film to video without costly pre-scanning, so the reference white level is set to 235 to provide headroom that is intended to prevent white clipping. Therefore, you should leave some headroom when calibrating a projector, but how you do that depends on how the projector performs, i.e. what happens to the grayscale as you approach clipping on one or more of the internal projector RGB output channels. Consequently, a knowledgeable calibrator will handle this problem very differently for different projectors.