am Bays wrote:(filmed with analogue technology then converted to a digital signal
not sure what you are getting at with that comment, doesnt make a lot of sense to me.
the ball tracking is pretty damn precise, it is the prediction that is the problem. because you are extrapolating over a potentially long distance, based on a small distance that the ball has travelled.
random example pulled out of my arse.
say you know the accuracy of the ball to within 5mm. (this is what i have heard quoted for the tennis hawkeye so i assume it is the same)
how we say the ball bounced say half a metre in front of the pad.
that would give us a 1% margin of error at that point in time.
Now the crease is 1.2m so we could say the crease plus a stride would be 2m.
That now means there is a 20mm margin for error which is a fair difference. which is why the half a ball thing is in place
nothing to do with the digitising of the signal.