Hi Moti, > Suppose that you want the outside sensors to be not too sensitive to > walls in a maze, but the center sensor to be very sensitive to the > absence of an obstacle ahead of it. (I've probably got that wrong or > backwards, but I hope I'm conveying the idea of different sensors > having different properties.) > > Also, there is an additional cognitive load involved with deciphering > the meaning of the block with the linked sliders. I suggest _not_ > trying to prevent a "meaningless" situation. At worst it is an > occasion for learning!
I understand your point, and almost agree. I still think that as the sensor buttons are three-states (gray, white, red), these two values are conceptually linked. But we can switch to two separated sliders. > There isn't anything wrong with the non-linearity. This property of > the sensors can be used to teach the concept. When you have such a > small slider ranging over a such a large range, it doesn't help that > the values have high precision. The sliders work perfectly OK for the > range 0-500 for the motors, and I think the sensor sliders should also > have 10 or so settings that can be reliably set. So you would have the sliders be linear? I am worried that with linear and discrete sliders, there will not be enough precision in the regions of interests. > I meant that the little triangle that you drag should be the same > color as the sensor image on the block: red or white, not both white > as it is in the current design. > > I'm attaching a rough drawing of two independent sliders in two > colors: I think that would be sufficient for providing the ability to > adjust the sensor thresholds and yet much simpler to learn. Actually I tried to keep a consistent look with the other blocks, hence the coloured background and white handle. I will wait for a while before touching these sliders again, as maybe some people have additional feedback, then I will explore your suggestions. kind regards, Stéphane -- http://stephane.magnenat.net _______________________________________________ Aseba-thymio mailing list [email protected] https://mail.gna.org/listinfo/aseba-thymio
