The problem I had was not with dt or slowing time, but with modifying velocity directly instead of modifying the parameters in function. I figured out the solution to my problem even before anyone else posted and I even added [SOLVED] to topic name. I don't need modifying global dt, because I don't ...

Modifying dt directly is bad practice. Also I already figured out the solution.

Ok, I figured it out. I just needed to modify x and y values in setLinearVelocity(x, y) instead of modifying the velocity itself. This way velocity remained true the whole time and only it's application was slowed down, which is what I wanted. Still would leave this here for anyone wandering about t...

I want to make a mechanic, where time slows for a few real seconds. Actually modifying time flow is easy after the fact, I just need to multiply any changes to velocity by a value I choose (0;1). The hard part for met to figure out is this: I need to modify every velocity of all entities at the begi...

And if I want to have collision callbacks between two bodies and them only, while not blocking bodies from overlapping, then what? Why can I only set up filters to ignore category completely, but can't set filter to allow category to act as a sensor in the specific filter I choose.

Thank you for quick reply. Yeah, that sounds like it should work. But your answer sounded, like this is not a frequent thing people do, although possible. For me it seems kind of essential to have overlapping events for objects. Is it not common to do this in Love 2D? Then what do people do instead?...

I need to implement behavior as follows: There are two dynamic fixtures A and B and there is another static fixture G which acts as a ground. I want both fixture A and B to be affected by gravity and they need to be stopped by fixture G. Simple enough, I did it. The next thing is, both these fixture...

Too bad you can't commend people on this forum.

Ok I was able to replicate it. Your example obviously worked. So basically the point was to apply drag based off of persons velocity, not dogs velocity. Thank you.

I also applied damping to the person.

I actually already tried that. I set person's mass to be 100000 and dog's mass to be 0.0001, just to test if it would make a difference. It did not. Even with absurd difference like that. Also I printed out the mass and it was correct. Console log for print of getMass of both bodies: 9.9999997473788...