Well, I found a workaround for this, it's still not quick enough, but way better than my original solution.
If someone is intrested in this in the future...
When you loop through an object's instances in the event editor, it's picking the other objects in the same container, so this way it works.
+ System: For each a
SOL.b.x = SOL.a.X
SOL.c.x = SOL.a.X
I tried this with 15 object, if I only use events it's running 2200-2300fps, this way 900-1200fps, and in pure python I got only 200-220fps(ouch).
Note: these values can be a lot higher(I've got the same fps with the mixed solution when I only checked them every 20-30ticks) if you not loop through all of them every thick, but that is only in option if you don't use vsync of course.