Why quitters can't be replace by new joiners that get team average skill tree

Battleborn Battleborn Competitive Multiplayer Discussion
#21

One thing I’ve noticed in private matches is that the bots actually play relatively good (like mediocre players), but fall back and ‘reduce’ their contribution to the team, once human players start to ‘do most of the work’.

(Also, the bots seem to be balanced for defensive first. To a point, where they won’t leave the base after you took the first enemy sentry on Monuments)

But if all human players are down, the bots actually start to do things again.
So maybe they could perform better as a replacement for a lost player, if they didn’t have those ‘shackles’.

As @dannyorke01 said, there are many ‘roles’ a character can fill. Maybe GBX will find a way to balance the individual bot AI for the roles of their character.
But this would actually revolve around the metagame and would therefore be a very costly thing (from both a money and time perspective), as the bot AIs would need to be rebalanced on a regular basis. And proper class/role sensitivity is something we’ll probably never see from bots, given the fact that even some players struggle with it.

#22

Sometimes my friends or I lag out. I’d rather be able to rejoin and play after I reset my game so that would kind of break that. The real issue is many people in general don’t want to join a game that’s almost over and it makes leaving a game by rage more “acceptable” since there’s someone to replace you. That being said if someone doesn’t return after say 3 minutes then allow it and just see how well this will work and then take it out if it hurts it but not for a big patch a hot fix is the onl way to not break the game for a while but it wouldn’t fit in one.

#23

UI adjustments. You mean “that one dude has left” and “what’s their name has joined”?

Score? Who cares about score when you are down several teammates? The people that leave on purpose wouldn’t have been contributing greatly to the team’s score anyway.

I actually believe it would be easier to get a new player in the game instead of a bot with next generation AI. Designing something that can think as well as a human? Now that sounds like it would take longer than getting a replacement player.

#24

Well, that would be a part of it. Among other things, there are also the Score Board and the team display in the lobby.
While it may not seem obvious to a human, the question ‘where to put which player?’ in all those screens is something that would need changes on a deeper code level than just putting out text notifications in chat.

Well, the game does. At least, that’s how it’s built.
While you’re making a good point from a player’s perspective, the game itself cannot deal with uncertainty, being a system of modular processes. It needs hard numbers.

If you read my post again, you will notice I didn’t say anything about developing new AI, let alone some kind of next-gen AI.

All I said was, that for the time being, bots only seem to spring into action in they’re deemed needed (by the game). They will (try to) defend their own sentries. They will even follow in advancing to the mid, if human players decide to push.
They are however so much balanced along the lines of ‘helping’ the players that most of the time they don’t actually help the team.

Prime example: Thralls. Bots will attack Thrall Camps but wait with capturing until a human player enters the immediate vicinity. This was clearly designed to allow the human players to profit from capture experience and keep the control over which thralls are captured when in the hands of players. While working ‘as intended’ in that regard, it’s actually detrimental to the gamestate in most cases and would have to be changed for drop-in bots in Public matches.

Also, if the human player’s ‘power level’ (for a lack of a better word) exceeds a certain point, the bots fall back and slack around at the base. So, basically this seems to be their default defensive directive (be at/near own sentry, build any turrets left and stand guard).

To even consider using the Bots as ‘fillers’ for dropped teammates, this default behaviour would need to be re-balanced based on character (e.g. Miko’s default behaviour should be
[‘move to where most teammates are’ / ‘heal injured teammates’ / ‘use slow/stun on enemy battleborn’],
whereas Pendles’ should look something like this
[‘move behind enemy lines’ / ‘focus turrets if allied wave/thralls are past the middle’ / ‘engage enemy battleborn with >20% HP’]).

I hope this dispels any misconceptions about my prior posts.

Anyway, with all that being said, I still see drop-in bots as the most realistic option to replace a quitter.

#25

There is no need for any deep changes, no deeper code, at this point it is all a simple scripting really. Total score would count in both score of player who quit and of player who replaced him. Adding player to the board is also not a problem, even if some UI graphics would have to be changed (and I dont think there would have), they already have the templates and should be able to change it quickly.

#26

Sure, it’s all scripting anyway. I’m not gonna argue that.

But there are adjustments to the game that can be done via Hotfix.
This however would require a ‘real’ patch, as they couldn’t just adjust some existing variables, but rather would have to add a couple of new methods to the whole game.

With what Vizard said earlier, I just felt like I should clarify this can’t possibly come as hotfix.

I just wanted to make clear, that if there was a solution to add new players or bots to an existing team, it would need to come with a full patch, due to the extend of changes.

#27

Now, I be damned. It seems like I proposed similar solution few hours earlier, and suddenly it got it’s own thread!
Well, I look at the case in a less nuanced manner. No bots needed, nor lvl calculation, just a simple hostile takeover on one player’s game session.
In that case “the substitute” inherits quitter’s (AKA loser) match stats ti lvl, mutations, k/d ratio, shards and a Bb of course.
This has many seemingly negative implications, but also many obviously positive so I think it’s worth it - we got no Bb pick screen but also 5vs5 match, we got no free stuff but also one person not waiting in a long matchmaking line, and we piss on loser’s mischievous action :wink:
Having in mind these inconveniences I’d also propose some rewards for gamers voulenteering to do it like honorable titles, skins &/or credits.

#28

They would need to work some things out, but I think it’s more of an undertaking than you believe. I mean, they genuinely have to change SOMETHING about the structure of PvP for joining in to make sense

1 Like
#29

I know that. It would be huge. What I meant was I don’t want it bc it requires a full patch and that means if it breaks the game it’ll be broken for too long. Unless a hit fix can remove it. Which is also not possible.

#30

The timeline isn’t bad and I don’t know how complex the code is nor how they divide up work but 4 months seems about right 35 people though it could vary either not at all or a lot depending on the above.

#31

I would be against getting stuck with stats and character of a quitter since the stats may not even exist like shards collected. The volunteer idea isn’t bad though it’s still and undertaking even if they did it your way. How matches are searched and process will be changed quite a bit. A box fox wouldn’t work for this either since it’s not like a number change.

#32

I think this is an important point! Big, complex and/or fundamental changes can lead to bugs, glitches and errors beyond imagination. Even small changes have this potential.
To avoid such incidents it needs alot of time to test and refine new added code and changed code.
Many people like to think “one guy could do it over night!” - yeah, maybe.
But what if the code is flawed or leads to further bugs? To assure quality many eyes have to overlook the matter.

Regarding my example with 35 people working for 4 months: These 35 people would consist of everyone who is partial or fully working on this fictional project.
They´d consist of the programmers, character designers, testers and the people who coordinate their work (managers, secretaries, publishers, bureau staff).

So alot of workers we´d not consider in the first place. Its easy to imagine the crazy code-smitting dude in the basement, but often we forget the various other jobs that are contributing to a game-project.

I hope hat makes my example more clear :heart:

#33

I get where you’re coming from, but I would be really disappointed if they got rid of the scoreboard. That’s usually how I know who to focus and who to be wary of (a good galilea or Benedict). Likewise, it’s nice to see which of your teammates is most competent. As a healer, I like to know who I can trust with my healz when the going gets rough.

But honestly, if it meant fun, playable games for everyone, it would be hard for me to argue against it

#34

No problem, nobody force you to substitute quitter. But some ppl would be that desperate, just to play instead waiting, and I have them in mind…and qutter’s team ofc.

As a matter of fact I think that that is how it would look like: just changing some numbers. Actual matchmaking is basically game server taking player’s ID and pairing it with match’s ID. Subtitution would have to mess with these only so that specific volunteer would be able to get assigned to qutter’s match.
The only objections that come in my mind would be loadout/Bb (skins, taunts) case.

#35

There is a functionality in session interface of Unreal Engine 4 to call searching for sessions, and there are even tutorials on how to deal with finding, listing and joining sessions in Blueprint, and they are rather short. A lot of things is handled by the engine, and most likely by Steam API as well.

And there is no need to get rid of scoreboard, it is the simplest thing to change in all of that.

I would say no more than 2 to 3 days of work for 2 people to get first build for testing. However I could be mistaken as Im not a coder and I havent worked with Unreal Engine 4.

#36

Still seem like you know more about it than me. I still share the worry that certain changes could be too buggy in an already pretty buggy game

#37

Battleborn was made using an altered version of UE3, not UE4