post #11 of 11
Originally Posted by kaingosu View Post

Again you don't understand what i wrote. I said you shouldn't add prediction in that list until you have an accurate way of measuring it. When you find that, only then ppl can make a decision based on objective data. If you say a sensor has 3-5º prediction, how would that help the average Joe? How is he supposed to know what that means? Maybe it will help me and you, because we know exactly how you measured it. But you can't expect people to learn the methods you used because, like you said, the masses don't even care. I guess a better way is to attach pictures with lines drew in Paint, that way anybody can see the difference with their own eyes. But it's a tedious work and no one wants to do it.

I was contemplating the Paint pictures as a way to do so, but it would still require a very well documented method to get them and for them to have a real meaning.
Thinking of drawing some straight vertical and horizontal lines, a few circles and several squares and triangles would be a good way to show the tracking of the sensor for any characteristic that was there, as in angle snapping, or to show jittering.

I don't have any clear idea on how to measure acceleration in a repeteable enviroment though unless that's making a video of a game with raw-input and making a vid of that, even with all the inconsistencies and the lack of precision of that method. It could be achieved with the right hardware, much like what Sujoy did, but it would cost time and money.