Jump to content

Any known bugs on followers packages?


Guest

Recommended Posts

Posted

Hello,

 

Does anyone know if there's some known bug and how to solve it, concerning the "follow" package? I'm using it on a npc that is not set as TeamMate.

 

This is what's happening. I have a npc following me inside a interior cell (quest.var sets her behaviours, if 0 waits, if 1 follows etc.)

Everything is ok, but SOMETIMES after a cell change, it goes wrong. Problem usually starts when I go out in the wastelands (the npc was born inside, I'm bringing her outside)

 

I placed a lot of redundancy in these places, when I noticed it was randomly happening. Essentially, I placed trigger zones with a script, it sets again the quest.var to the value of the "follow", then npcREF.resetai. It seems that after this the problem isnt' happening anymore in this third cell; before that, it was randomly happening...

 

But when I go out it's worser. Just outside the exit there's a trigger zone Onplayerenter, with a script, it advances the quest stage and triggers a "npcref.startconversation player"

 

And sometimes this doesn't happen. Worst, if I speak with the npc she answers "What do you want?" like a generic greeting that doesn't meet conditions. And this shouldn't be true: if she was following, the condition is setted (same condition for the package is for dialogue)

 

This is absolutely random, it happens like once every 5 tries. I'd like to understand if there's some "known issue" on this. I.E. I've seen every follower has a onload/doonce blocktype that re-evaluates their package. Now this isn't something I need in this issue, because she doesn't need to re-evaluate, she doesn't have a package anymore... -_- I can see it if I disable/enable her by console...

 

The last info I can provide is that to reach the third cell the player passes through a hatch with an OnActivate script on it. There are no "player references" on it, it means that anyone could be affected OnActivate. It is simply a OnActivate -> Activate, so it's like passing through it (script is needed because there were other conditions that are not affecting now). Anyway, after that hatch, I placed another trigger as redundancy to move her to me if something was wrong, re-set the value to the follow package etc. in fact she still continues to follow me after that hatch. Till the last step, when I go out, when randomly she goes completely crazy (no more AI packages on her even re-evaluating by console, resetai, disable/enable)... Reload / redo everything, everything goes well... -_-

 

Now I could place some more redundancy outside too, re-setting again the value of the quest.var... But really, I'm wondering why this is happening. Has someone already had this kind of random issue?

 

EDIT: found actually which behaviour she was having when things weren't working. I couldn't figure it because it's something that shouldn't really happen, it's another package she had in another cell. What I can't understand is that behaviour is given by a certain script in a completely different place, on a trigger zone. This is the block, the only one that set that behaviour to 5:

 

begin ontriggerleave player

    if (getstage aaPNowheretorun == 40)
    set aaPinkyAptSet.aaPKatPackage to 5
    aaPKatREF.resetai
    endif

end

 

How this "ontriggerleave" is supposed to trigger when I change cell, if I am in a completely other cell? more, I'm 3 cells away from there... Shouldn't be require that I first enter on it, then I leave it? And I didn't create a new Trigger starting from this one then changing it. I always create new things since I noticed duplicating existing things bring the scripts with them. So, or it is triggering for a reason I don't know, or the whole thing is going crazy and me with it :)

Posted

There are numerous issues with packages and cell changes. So much that the engine itsellf was changed to force teleportation of the teammates if they ever fall behind.

 

Any package added by addscriptpackage gets deleted if some other regular package gets elected to be run. So if you see your NPC loosing her package, you need a script to check for this and reapply it.

 

 

I blame this on the code for optimizing ressources but its just a hunch, I have no definite answer. Just like having scripted effect shut down as soon as the player is out of range.

 

Posted

Thank you Jaam, I edited with the last thing I've found.

 

There are numerous issues with packages and cell changes. So much that the engine itsellf was changed to force teleportation of the teammates if they ever fall behind.

 

Any package added by addscriptpackage gets deleted if some other regular package gets elected to be run. So if you see your NPC loosing her package, you need a script to check for this and reapply it.

 

 

I blame this on the code for optimizing ressources but its just a hunch, I have no definite answer. Just like having scripted effect shut down as soon as the player is out of range.

 

And these were exactly the things I wanted to know. Maybe they are not the reason of this peculiar issue above here, after what I've found, but it is something I had many times and I didn't know why it was happening. I agree on the continuous re/check of things, but also sometimes I don't know how to check it. I.E. sometimes the conditions were satisfied but the evaluation wasn't evaluating when requested (ok I could wait 15 seconds, but I didn't want to), so I tried to force it until the result of GetCurrentAiPackage was my effective package, but a cycle / condition like that was only leading to my npc getting stuck in a neverending evaluation. I don't know, I really get messed up with packages, they works 90% of the times and then there's a 10% that stops working without an apparent reason. Just like navmeshes. All I'd like to do is placing some redundancy so that I could avoid any possible issue.

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...