Double equals roughness precision

The stuff I have written there is the reason why people should not use floating points for money computations. If you are working a little with floating points, then it is quite important stuff to learn, at least a little, because there is a lot interesting stuff going on, for example numerical stability, as in floating points it can happen that (a + b) * c != a * c + b * c (you should use latter as addition can loose precision, while multiplication doesn’t). Actually I should check if it would be possible to create compiler tracer that would warn about such things.

I think many decades ago when I used MatLab it did warn me from such equations working with float vectors and I was going insane, as it is mathematically correct theorem wouldn’t translate into computational.

Reading the help pages of MatLab gave some insight and stayed away from additions wherever I could.