Jump to content
Dante Unbound: Share Bug Reports and Feedback Here! ×

What can be done about quitters?


(XBOX)Erudite Prime
 Share

Recommended Posts

1 minute ago, kapn655321 said:

A new form of matchmaking that prioritizes available players relevant to our style and interests.

Great, and how exactly would you implement that? What system would you use to collect the necessary data, and what kind of algorithm would you code, and how would you code it, to make sure the matchmaking system keeps such data into account?

Link to comment
Share on other sites

3 minutes ago, TearsOfTomorrow said:

Great, and how exactly would you implement that? What system would you use to collect the necessary data, and what kind of algorithm would you code, and how would you code it, to make sure the matchmaking system keeps such data into account?

We'd need a profile option to get as specific as needed to find relevant points in psychometrics.
A system to rate our experience with other players with some specifics.
A categorization from DE on what points of value there are in each mission.

As for what kind of algorithm, I don't know. I'm not a programmer, nor do I study types algorithms.. and I wouldn't presume to know what would work for the Devs. That's up to them should they chose to pursue something like this.

I know for sure public is not satisfying, and the answer to what is differs by every player. The only way to ensure all player's needs are met is through a flexible AI that can measure which factors are causing the specific conflict for the player, and mitigate them.

If you happen to know algorithms, feel free to suggest a few to look into.

  • Like 1
Link to comment
Share on other sites

3 hours ago, (XB1)Erudite Prime said:

People who join Endless Fissures like Defense or Survival, but leave after only one Rotation. How can we solve this problem?

You can't cure the human condition, sorry...
I mean, sometimes, it's excusable, like that's the only fissure available, or there's an unforeseen event they need to attend to...
But 7/10 times people are just doing a endless mission only to inconvenience other players (on purpose or out of ignorance), and waste their own time, when they would have faster options.


That said, as many people said before, although it's inefficient behaviour, those people are free to do what they want, so if you want better companions, check recruit chat, or get a clan or something. It's not the game forcing you to do anything. This IS a social game, it's meant to be played socially with other people. You refusing to get a clan or find a few people to play in these occasions is just as misguided as those guys that leave the endless missions at the first wave.

  • Like 2
Link to comment
Share on other sites

7 minutes ago, kapn655321 said:

We'd need a profile option to get as specific as needed to find relevant points in psychometrics.
A system to rate our experience with other players with some specifics.
A categorization from DE on what points of value there are in each mission.

As for what kind of algorithm, I don't know. I'm not a programmer, nor do I study types algorithms.. and I wouldn't presume to know what would work for the Devs. That's up to them should they chose to pursue something like this.

I know for sure public is not satisfying, and the answer to what is differs by every player. The only way to ensure all player's needs are met is through a flexible AI that can measure which factors are causing the specific conflict for the player, and mitigate them.

If you happen to know algorithms, feel free to suggest a few to look into.

Welp, I was expecting the standard "I dunno but I want this, the devs will figure it out" kind of childish, entitled response. Instead, you made an actually intelligent post: you got me there, kudos for that.

So, let's proceed in order: you proposition in regards to profiles is reasonable, but has the major flaw of being based on volatile, non-objective data. Like, to pair together two guys who won't quit after Rotation B, you'd need to keep track of how many times a person chose to stay until Rotation C, right? But as others have mentioned, people have different reasons for staying or leaving: a guy who usually would quit early might decide to stay in longer when paired with you because they can't get the drop they want. Viceversa, a person who usually would stay a lot might get that gold rare drop on first try, and decide to extract. This is an aspect of the human thought process no machine can predict. 

The issue with rating other players is that people are petty,  and might decide to ostracize a person for the most worthless reasons. Overwatch had to remove a similar system because salty losers, upon being beaten by very good players, would exploit said system to ensure they'd never end up facing them again, and it eventually got to a point where those good players had nobody to play with.

Points of value are another problematic issue, because fissures (the topic at hand here) skewer those. Normally in Capture you'd kill nobody but the target, but with fissures you also need to look for reactant. That's just an example but you see what I mean here.

So yeah, your suggestion is actually pretty damn clever, but sadly I'm afraid creating an algorithm capable of keeping track of all those kinds of data AND dynamically interacting with the matchmaking system AND compensating for all the issues I outlined would simply be unfeasible under and realistic circumstances. That is my professional opinion as a developer, albeit an admittedly small one.

Link to comment
Share on other sites

11 minutes ago, TearsOfTomorrow said:

you'd need to keep track of how many times a person chose to stay until Rotation C, right? But as others have mentioned, people have different reasons for staying or leaving: a guy who usually would quit early might decide to stay in longer when paired with you because they can't get the drop they want. Viceversa, a person who usually would stay a lot might get that gold rare drop on first try, and decide to extract. This is an aspect of the human thought process no machine can predict. 

It certainly gives you a ballpark. Let's say, you can measure that player tends to stay on average, 30-50 waves. ...matching them with a player who tends to leave at 10, More certain to be a conflict of interest. It's not about making a perfect human experience.. which is totally impossible. It's about putting us in touch with that potential consistently enough that we generally have the experience we want.
 

14 minutes ago, TearsOfTomorrow said:

The issue with rating other players is that people are petty

So if they rule themselves out of your games, so be it. They'll be hurting themselves if they do.. but if they still feel they must, the end result is a more enjoyable experience for them and the people they clash with. Ideally, your rating should impact your results, but also account every so slightly if it detects a common pattern. "People who like these things just seem to run into problems., problems from the users with X preference Against the others, not the other way around... let's marginally adjust priority to reflect that, should it be consistent."

Each point of correlation should not wholesale shift, but gently inform the AI.
Your personal rating should affect your outcome... but as for influencing all other decisions in game for others, should be measured to others as the fraction of a % of the player base that it is. Even if petty people are petty all day, it may never have an effect on your experience.. and if they are petty, they're still a customer that seeks an experience they enjoy.

You're totally right that I have No idea about it's feasibility, or how this would be implemented. What I do know, is this is the only approach we're all universally able to use to have the matchmaking we want be reliable every time.

Link to comment
Share on other sites

@TearsOfTomorrow You're going to be a perfect person to discuss this with if anyone.

Whether here or in PM, I'd be very interested to know what aspects of this idea are viable, which aren't.. and maybe brainstorm some compromises and game plans about how achievable it is. Most important of all, if the game can measure any certain points of strong reaction, whether positive or negative, it helps guide DE's decisions to create a better player experience.. so what are the tools through which we can achieve this.

Thanks for responding, by the way. Having a programmer around to fact check is certainly handy.

Link to comment
Share on other sites

1 hour ago, Pr1A said:

I usually do, but sometimes there's only endless missions (or f*cking Mobile Defense) available.

They should change the fissure generation to ensure that there's always at least one endless and one non-endless mission available. And get rid of MD fissures because no one wants those.

True stuff. Often I just want to speed run Ext or Caps, but the specific relic type may not have anything desirable available. Like right at this moment. I want to burn through my Liths, but there's only a Lith Survival lol.

Link to comment
Share on other sites

5 minutes ago, kapn655321 said:

let's marginally adjust priority to reflect that

All your suggestions are very smart, but the issue is that, as exemplified by this line, with each new thing you say you're adding new variables for the algorithm to account for. Increasing the variables increases the complexity of the algorithm's internal architecture. And that in turn increases costs in terms of resources consumed, loading times, and actual money spent for the research. As all these costs keep increasing while you attempt to overcome what boils down to the central conundrum of AI (that unlike humans an AI is not capable of imagining, and thus cannot comprehend things such as a "reasonable margin". AI can only act based on discrete data input), eventually the time comes to ask yourself: is It really worth it?

This is a commercial videogame, after all. We're not trying to change the world here, we're just helping a bunch of people play pretend. And we're also trying to make money off it. So after a while the innovations required to implement an idea become so extreme and so costly, it becomes objectively more feasible to just settle for something less effective, but also less expensive.

Or to put it in layman's terms: don't spend a billion to bring about a minor change that will earn you back a million at most.

  • Like 1
Link to comment
Share on other sites

1 hour ago, TearsOfTomorrow said:

Or to put it in layman's terms: don't spend a billion to bring about a minor change that will earn you back a million at most.


If you could. So let's scale it down. Setting an arbitrary goal of ~6 complex, but finite factors.

  1. Player loadout.
  2. Similarities in drop tables/mission types.
  3. (1 Through 10) rating system regarding match satisfaction, or like upvote/downvote/maybe-later
  4. 2 emote regarding specific players that you only tell the game itself.
  5. Player's average completion time/success rate in mission types.
  6. It detects good/bad correlation, it can shift priority of suggestions.


Each match is an iteration (or however it works), 'till it aggregates methods resulting in higher overall match satisfaction.
Suggest and prioritize games by potential positive matches.
"People who use this gear, in this mission type, finishing in these ranges of time frames and success margins, tend to enjoy these factors more." 

You could find your own matches just the same as now, or select from a short list of favorable picks.. 2 strongs and 1 maybe (whichever way,) or pick your own.
The more you play, the more it gets an idea of what you're happy with.. and you can tell it, "No," or, "not now," or, "nothing specific," at any time.
Otherwise it's throwing you at the most enjoyable game it can think to give you.
No matches on that node? Provide suggestions by relevant game type and suitable preferences.

We would give feedback in game, convert frustration into both solutions, and some nebulous sense of validation..
and have it merely suggest, or offer alternatives that it thinks we'll like.
If it remembers you matching well with others under different circumstances, it could suggest friends or clan alliances some day.

Spoiler

DE can never fully seem like, "they do not listen to us," if their game is literally always, "listening and responding," to our input, providing ever better results.


Here's another thought: Bad reactions having a half-life.
Never fully disregarded; less influential over time.
Considerable differences from last encounter, count less against the match up.
Both players have different loadouts, the mission type is different than last time, both seem to enjoy games involving their factors.. That goes in the Maybe pile.

Edited by kapn655321
Link to comment
Share on other sites

 

5 hours ago, (XB1)Erudite Prime said:

What do you mean by this? I am saying that I don't like the way the something works, and that I'd like it to change. Replying with "Well that's just how it works" isn't an argument because I know how it works, and the point of this thread is to make it different. 

Solo the content, hotshot!  You seem to know what is best not only for yourself, but for any other player that may be graced by your presence, as well as the game itself.  Surely one so enlightened and better than the community they query can take matters into their own hands and do 20 rounds of defense, right?  If YOU want to go for an hour and a half in a 5-min mode its up to YOU to make it happen unless you've arranged a party to complete the objective.  

There are plenty of problems with this game, but having the autonomy to extract at specified intervals is not one of them.

  • Like 2
Link to comment
Share on other sites

7 hours ago, (XB1)Erudite Prime said:

How can we solve this problem?

Its not a problem that needs fixing, its a issue you may have but beyond you choosing to play with a preformed group that has agreed to do X for Y then you will have to accept that in a PUG people will do what they want, not what you expect them to (or to match your narrow expectations) Solo extraction actually fixed the major problem of people being held hostage in missions that they wanted to extract from but the rest of the party didn't without losing what they had earned to that point.

 

Link to comment
Share on other sites

Recruit or play with friends just to get that outta the way.  In my experience if you're with randoms and ask at the beginning how long everyone wants to stay it kinda works.  If someone doesn't answer then majority of the time they're leaving after first rotation.  If it's the host that won't answer I just leave and queue up again.

Link to comment
Share on other sites

3 hours ago, (XB1)Erudite Prime said:

Soloing Fissures is a complete and total waste of time.

I've got every Prime frame in the game (don't have Vauban built because no point, but still have the parts) and every P weapon I have any desire to own.  Aside from a few that were vaulted at the time that I traded for (maybe 5%) the rest was obtained through solo relic cracking.  I save up until I've got 10-12 of a relic I want a part from, forma any gear I want to level and equip those items, and then go grind a (preferably) endless fissure until I get the part while leveling gear.  Worst case scenario I get everything but the rare out of the relic w/ spares to sell for plat and have a melee, secondary, primary, and frame leveled.  Takes an hour or so and I nearly always get everything from that relic, and now I don't have to do a boring ESO run to level gear.  If this sounds like a boring waste of time, frankly, you must hate this game, lol.  That's the majority of the game: grinding.

If that's an unfathomable scenario then type into recruit chat, for example: "hosting Axi B2 rad/flaw+/etc 10+ rds" or "LF axi b2 farm" if you don't want to host.  In the 25min you've waited to get a recruit party together, I've already cracked 5 relics.  To each their own...

Link to comment
Share on other sites

4 hours ago, (XB1)Erudite Prime said:

Soloing Fissures is a complete and total waste of time.

I'd guess some of the people quitting your games after one rotation would give that explanation as well.

I wouldn't mind better matchmaking, but I think any system that doesn't fragment the userbase or actively drive players away is also not going to solve the problem for you.

Likewise, better ramp-up of rewards and better communication of those rewards would help.  But again, you're still going to be dealing with people who often have different agendas than you.

  • Like 2
Link to comment
Share on other sites

1. Don't play Limbo*

 

That's it.

 

 

*or any other loadout that makes 5 defense wave last 13 minutes.

I find myself leaving mostly when I encounter other people intentionally playing in a way that makes the mission way longer than expected. Because this kind od douchebaggery is their free choice and also is commonly approved, finishing the 5 and leaving seems like a reasonable compromise.

Edited by saradonin
  • Like 1
Link to comment
Share on other sites

The mission is over when i get what i want. Taranis, I need 1 argon cristal. I get it and i leave. Im no going to stay 25 rounds just cause you want to stay too.

If you were my friend or you asked it as a favour, i would consider it. But you are basically asking me to waste my time for nothing. Most times i do 15 rounds cause i feel like it. But its my choice, when im bored and just want to stay without a clear reason. No cause i feel like im obliged to stay.

This thread was created by a troll or someone incredibly dense. I think is the first case.

  • Like 4
Link to comment
Share on other sites

5 minutes ago, Travis05 said:

someone incredibly dense.

Their goal is to have missions people want to partner up and hang around in. That's a perfectly sensible request. Coming in to bad mouth OP and provide nothing gets us nowhere if not further back than where we started.

Sounds like, If there's something you'd still want, you might want to stay... and THAT would be a solution. Better rewards. Got any in mind?

Link to comment
Share on other sites

hace 8 minutos, kapn655321 dijo:

That's a perfectly sensible request. Coming in to bad mouth OP and provide nothing gets us nowhere if not further back than where we started.
 

Im the first who want a mission well done. But this isn about a specific objective. There are infinite rounds and people hang out as much as they want. I will stay until everything is completed but that doesnt apply here. I say is dense cause come on, how can you miss why these missions exist in the first place?

But no, i said its one of two things. And i dont think OP really made this thread as a genuine concern. This seem like a poor fabricated bait, and i was just pointing that. That is incredible how this thread is still alive.

You want rewards, you stay, you dont want rewards, you leave. Its that simple. Its different if you agree to something or use the recruitment chat. You cant expect to enter, dont say a word, and have people read your mind. "i would really like if my squad would stay until round 20, i need those resources". And even if you ask they could say no, and that would be fine. They arent forced to do it.

Respect people time, stop harrassing them for having life outside a game.

  • Like 2
Link to comment
Share on other sites

All this is solved if you just talk to others in recruit chat and put in the effort to find people that stay for Rot C reward. This sounds more like a social anxiety issue or a low motivation/drive to actually find other players to play with for a online Co-op game then the people staying problem.

you want DE to hand you a solution so you do not have to put in the bare minimum.

At this point OP you pretty much are being incredibly abrasive with how you go asking for solutions to your problem which is subjective since some ppl wont see it as a problem.

Edited by (XB1)Dex Xean
  • Like 2
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...