The method, no. Video with, say, a 12bit intensity range, would increase that 256 to 4096.Hmm, that's a subtle point. I believe that that doesn't impose a limit to the method
I don't think averaging influences the mechanical theoretical limit.at least if it's somewhat expanded to do some averaging.
Slightly. There had to be enough "room" for the pattern to repeat (a full alias cycle), so to speak. That is where the NIST 100px translation came from.The practical limit is the height in pixels of the façade.
The driving factor limiting fidelity is primarily the angle of the line. (or more specifically, the ratio between line angle and relative pixel size)
The method forced NIST to select a point fairly low on the West edge, rather than a more useful position such as the NW corner.
It's possible, yes. Perhaps theoretical maximum was the wrong phrase...practical maximum...in theory, with a large enough video frame, full spectrum usage, and the right line angle, almost limitless, in theory.I can see the possibility of achieving, for example, resolutions of 1/300 px even with 256 tones per pixel.
Again, for clarity, the actual NIST example had a 1/100px maximum fidelity.
Pretty good still, and unfortunate that similar quality results were not gained by NIST for their vertical motion study.