Dot product strangeness

[noc2]

Hi;

I have quite a simple question on a seemingly simple task, yet I'm kinda confused. If you check out the snapshot attached, you'll see that I tried to create Outward facing normals on a circle and also vectors facing along the positive Z axis in a parallel fashion. When I go ahead and get the dot product of these 2 vectors, I know I should get scalar values between -1 and 1. Yet, when you check out points 0, 2, 4 and 6, not all the values are as they should be. The dot product on points 0 and 4 seem to be correct, yet I'm getting off values for the other 2. What do you think I'm doing wrong?

Thank you so much in advance;

 

AJ

 

[Aizatulin]

Nothing wrong with the results (the small values near zero are due to the numerical accuracy)

[noc2]

Hi @Aizatulin Thanx for your reply. The problem that I'm seeing is that the values go beyond -1 and 1. Shouldn't they yields be confined to that bracket only? And with both vectors normalized, I find the results to be off

Edited January 17 by noc2

[animatrix]

The values you outlined are not outside [-1, 1]. Some of them are just really small values.

[noc2]

17 minutes ago, animatrix said:

The values you outlined are not outside [-1, 1]. Some of them are just really small values.

Oh jeez... Thanx @animatrix. So, I guess I'm add odds as to how to read those values. Lastly, I really want to be able to read those values correctly. So, the dot product at point #2 reads: 4.37114e-08 as is. I'm aware that it should be very closely biased toward 0 as the angle is seemingly almost 90 degrees between the 2 vectors. What is the right way to interpret such numeric yields, for someone like me, who expects to see literally something like -0.00001 or something?

Lotsa cheers;

AJ

[Aizatulin]

You can do something like this: if (abs(f@dot) < 1e-6){f@dot = 0;}