well i thought I had replication all figured out. I’ve worked with much larger implementations recently playing with blueprints and it all worked no problem. The difference then was that I wasn’t casting to the animbp at all and tick event in the animbp was sufficient. i never thought changing how I set the vars in the animBP would matter so much in terms of replication. because I have X amount of characters and when primary atk is pressed depending on the character different things happen completely, including length of animation, size of sphere traces, delay of unsetting the variables, how many traces are run, if a trace is run etc. I wanted to make it so basically the same functions work no matter what character you are, so future changes dont need to be done in multiple code branches.
This was working fine. The issue though is that there are tons of functions after the top layer setting variables in each step of the way add to that timers that unset things when needed, since the animation length or movement disables vary depending on the action. Because this was all working before I didnt think twice about it. Now I’m like “you get a multicast… you get a multi cast… you get a multi cast…”
of course its not working though. I’d consider these animations important, since its the feedback the user gets that the atk is happening. When the atk doesnt play, the attacked character still takes damage and the rest works. (except the getting hit animations of course, same problem now.)
I don’t know how to use interfaces well. Might be helpful.