Now that I got home, I tested this further extensively: I can confirm that with smaller objects the issues you described here are noticable.
For the developers:
In summary, smaller objects are much more harder to grab while moving, especially in scenarios where “next frame” would have the entity further away than the mouse current location, instead of in current frame. This can also occur with larger objects, if the object is moving fast enough.
The line of code I pointed to also confirms that at some distances the small objects cannot simply be interacted with if they are far enough from the camera, but the distance shouldn’t still be an issue as the objects would be too small for you to accurately grab due to the above case: but this should really be by avatar distance instead of camera distance, considering that default inspect script. unless this will change?.
The main issue however, which I think Balpien is frustrated about is that
Initiating the grab doesn’t actually stop the inertia of a dynamic entity. Only once the mouse if moved while the mouse button is hold down, does the object stop, and then update to the mouse position. This means you have no feed back if you actually grabbed an object or not and leaves you guessing. Since there is no beams from our hands, we are left guessing even further, and simply release our mouse to click again. This can be simply replicated by rolling an object and then grabbing it and not moving the mouse.
Unfortunately, I have been unable to repeat the spastic movement with the small objects being moved via grab. Perhaps it is caused by something very specific?
I personally do not see you as “no help”, @Balpien.Hammerer it was still a valid issue you found. You should have mentioned the tiny objects first and make a small test case. Take a break though every now and then, it helps.
also what is this. an interface for ants?