The Introduction To The DBD Death-Efficiency Problem.
What is the Death-Efficiency Problem?
Take a quick look at this graph:
Do you see the 4 dots? Each dot represents a survivor's Generator efficiency, meaning how much % on average they can be on a generator.
X = the amount of survivors who have died.
Y = the amount a the % speed at which 3 survivors would be doing generators. (it's pointless to take it for 4 survivors as we expect 1 to be in a chase).
The dots are a rough estimation from personal experience and the experience of many others.
So as you can see, it doesn't decrease in equal amounts. The moment the first survivor dies, we see that the difficulty of doing Gens has more than doubled.
In theory, when you kill 1 survivor, you've only taken down 25% of their power. But this number would be an inaccurate representation of the amount of Generator-efficiency you've really taken away.
So let's try: you work with 3 people on gens, so when 1 person dies, you only have 2 people on gens and the other gets chased, so we lose 33% of our Generator-efficiency? We're still not quite there...
What causes the Death-Efficiency Problem?
There are a few crucial states that you need to understand:
Stage 1:
-4 survivors are alive > Nobody has been hooked
-3 survivors doing gens, 100% efficiency
Stage 2:
-4 survivors are alive > A survivor is hooked, requiring a rescue.
-2 survivors doing gens, 66% efficiency
-When another survivor has been found > -33%
-When the rescuer is done healing > +66%
Stage 3:
-3 survivors are alive > 1 person is being chased.
-2 survivors doing gens, 66% efficiency
Stage 4:
-3 survivors are alive > A survivor is hooked, requiring a rescue.
-1 survivors doing gens, 33% efficiency
-When another survivor has been found > -33%
-When the rescuer is done healing > +66%
For every survivor alive, there are 2 stages, making a total of 8. But you can already see that halfway, at Stage 4, we've hit rock bottom. When someone is hooked, 1 is healing and the other one is chased and dies before they are done, they they are no longer able to progress in the game.
They could consider to stop healing to more quickly boost up their generator-efficiency in the hope they complete things fast enough before another opportunity of them being hit, but after stage 4, their options have plumbed.
There are 2 other important factors that we need to take in consideration that contribute to what we see in the Graph:
1. Pallets
2. Generator Regression
As you might have noticed, not every stage takes just as long. Sometimes the entire game is Stage 1, meaning that you have this 100% generator efficiency all match long.
This is because of tools like pallets, which allow you to stall whatever stage you're in, but is finite, meaning that the later in the game (typically) most pallets will be gone, meaning that the chases will start to take shorter and shorter. This already adds to the despair and the curve of the Death-Efficiency Problem.
To add to this, our second mechanic to make things worse is Generator Regression. It seems innocent at first, but it mainly targets low-survivor count teams.
"How?" you ask; Because the Generator Regression regresses alongside of a constant De-efficiency, meaning that it doesn't follow the Generator-Efficiency of the survivors. When there are 4 survivors alive, the regression is a low % of the survivors Generator-Efficiency, where as the Gen-Efficiency quickly drops, Regression as a % of Gen-Efficiency increases, and the survivors can no longer keep up.
Now with those 2 added factors, we are where we are; the Graph starts to make sense to us.
End note
This discussion is here to bring to light some of these issues. For a lack of time, I will not go over possible solutions in this discussion, but many of my other posts have been created around the awareness of this issue.
For those of you who feel appointed to solve this issue; you are highly invited to leave your thoughts and approaches in the discussion section down below.
Thanks for taking the time (:
Comments
-
can you sum it up for me, im too lazy to actually read it
5 -
What is the problem with this?0
-
TL'DR:
In a 4 survivor game as this, there is an effect called the "Dead-Efficiency" problem.
This problem makes it so that, upon losing 1 unit (25% of all survivors), you lose disproportionally more in strength, giving survivors a feel of despair when things go wrong, and making it so that in games where no survivor dies early are almost instantly lost on the killer's side.I cover how pallets and Generator Regression both worsen the problem and invite readers so solve it. (I will do so myself when I have more time)
0 -
@AlwaysInAGoodShape said:
@alivebydeadightTL'DR:
In a 4 survivor game as this, there is an effect called the "Dead-Efficiency" problem.
This problem makes it so that, upon losing 1 unit (25% of all survivors), you lose disproportionally more in strength, giving survivors a feel of despair when things go wrong, and making it so that in games where no survivor dies early are almost instantly lost.I cover how pallets and Generator Regression both worsen the problem and invite readers so solve it. (I will do so myself when I have more time)
thanks fam, this would make more sense due to a fear of survivors not wanting to die next, I will find a way okay?
2 -
@JanTheMan said:
What is the problem with this?Because the game must be balanced around a said escape-rate or whatever factor you want to use, you find yourself picking a somewhat average spot in between the good and bad Stages.
However, because of this, early plays/mistakes are rewarded and punished disproportionally, where as later plays have significantly less effect. This gives the feel of "It's already over", or in the opposite case; "this is won for sure".
It also kills most of the end-game potential this game has, as it's all about keeping the first 1/2 stages up as long as possible, and if it flunks, it's already over.
It is why Tunneling is a strategy so vital to killers, where as it ruins the fun for survivors.
It would be healthier if the Death-Efficiency problem would've been solved, making it so that survivors more consistently die, but that is less frequently spirals into the death of all.One can say that the Devs already noticed this problem; it's the sole reason why there is a new hatch mechanic. But that of course is a lousy way to "solve" it)
1 -
@alivebydeadight said:
can you sum it up for me, im too lazy to actually read it4 survivors alive = bad if you want to win as killer
Solution: kille one ASAP (also called tunneling by survivors)
1 -
@Master said:
@alivebydeadight said:
can you sum it up for me, im too lazy to actually read it4 survivors alive = bad if you want to win as killer
Solution: kille one ASAP (also called tunneling by survivors)
That is why your win should not be decided upon pulling this 1 thing off to get the better end of the Death-Efficiency Problem.
Your decisions (on both sides) should be valuable throughout the game and the importance of them should be spread.
Killing the 1st survivor is already infinitely more valuable to killing the second or the third. This is how our problem is defined.4 -
I understand what you mean, and this is why balance in an asymmetrical game is such a fragile thing. It's why the so-called "snowball effect" exists. If either side has a strong start, they can wipe the floor with the other one. It also occurs in any game where the extent to which one side can pull ahead of the other is theoretically infinite (MOBAs, for example).
There are two common ways to fix this.
- Give one side better tools to close the gap as their side begins to lose. In DbD, this could be something like increasing generator repair speeds as more Survivors die (but it would have to be inversely proportional to the number of repaired generators), or giving the Killer a boost as more generators are repaired (if the Survivors are still alive and well).
- Give one side fewer advantages for pulling ahead, so that they still become more powerful as they complete their objective, but not quite as much. I'm not sure what this could be in DbD because the gameplay is extremely basic.
2 -
Oh, okay yeah I see your point now. (the use of vocabulary and math, in your post confused me lol)AlwaysInAGoodShape said:@JanTheMan said:
What is the problem with this?Because the game must be balanced around a said escape-rate or whatever factor you want to use, you find yourself picking a somewhat average spot in between the good and bad Stages.
However, because of this, early plays/mistakes are rewarded and punished disproportionally, where as later plays have significantly less effect. This gives the feel of "It's already over", or in the opposite case; "this is won for sure".
It also kills most of the end-game potential this game has, as it's all about keeping the first 1/2 stages up as long as possible, and if it flunks, it's already over.
It is why Tunneling is a strategy so vital to killers, where as it ruins the fun for survivors.
It would be healthier if the Death-Efficiency problem would've been solved, making it so that survivors more consistently die, but that is less frequently spirals into the death of all.One can say that the Devs already noticed this problem; it's the sole reason why there is a new hatch mechanic. But that of course is a lousy way to "solve" it)
1 -
This reminds me of some stats that were collected from Gears of War tournaments. Up until Gears 4, the competitive mode was 4v4 Execution (a last-team-standing elimination mode for those that don't play). The game is broken into rounds, and you need to win 4 rounds to win the map (and you usually play best of 3 or 5 maps).
Someone did an analysis of "first death" and the outcomes and found that the team that got the first kill had something like a 75% chance to win the round (ie. the game drops from 4v4 to 4v3). The team that got the first 2 kills had a 95% chance to win. And if you got the first 3 kills you have over a 99% chance to win (basically a 4v1, you can't come back from that except in EXTREME circumstances).
It was very interesting because it meant that games actually came down to the opening plays and not end game plays. And thinking about the series of events it makes sense. You get a kill now it's 4v3 and you will likely secure the crucial weapons on the map because you have a numbers advantage. Once you do that, you now have a weapon advantage too, which can snowball.
All of this is part of the reason that the Gears devs created Escalation in Gears 4 as the new competitive mode. However, IMO as someone that has played the game since the original, Escalation isn't very fun. It's very competitive, and has a very robust meta, and is incredibly balanced, and is much more exciting from a spectator standpoint, but I don't find it fun like I did Execution or even KOTH because it's not very intuitive and having a full squad is basically a requirement at anything about Onyx rank.
2 -
@AlwaysInAGoodShape said:
@alivebydeadightTL'DR:
In a 4 survivor game as this, there is an effect called the "Dead-Efficiency" problem.
This problem makes it so that, upon losing 1 unit (25% of all survivors), you lose disproportionally more in strength, giving survivors a feel of despair when things go wrong, and making it so that in games where no survivor dies early are almost instantly lost on the killer's side.I cover how pallets and Generator Regression both worsen the problem and invite readers so solve it. (I will do so myself when I have more time)
That's why its better to get a survivor out of the game as fast as you can.
3 -
@se05239 said:
@AlwaysInAGoodShape said:
@alivebydeadightTL'DR:
In a 4 survivor game as this, there is an effect called the "Dead-Efficiency" problem.
This problem makes it so that, upon losing 1 unit (25% of all survivors), you lose disproportionally more in strength, giving survivors a feel of despair when things go wrong, and making it so that in games where no survivor dies early are almost instantly lost on the killer's side.I cover how pallets and Generator Regression both worsen the problem and invite readers so solve it. (I will do so myself when I have more time)
That's why its better to get a survivor out of the game as fast as you can.
take the pawn first, king last
Oh my god that would be an amazing perk quote
3 -
Oh my god I didn't even think of this. Last month I wrote a fixing dead by daylight post detailing what I thought was wrong and what I thought how the game should be, with specific examples. I'm gonna go write a revision #3 with this information and will be crediting you.
PS: here's the post https://forum.deadbydaylight.com/en/discussion/28411/my-how-to-fix-dead-by-daylight-plan-revision-2/p1
1 -
I found this to be a rather interesting problem to ponder so thanks for that.
However, it occurs to me that surely solving the ‘death-efficiency problem’ is the point of the game. As a killer your aim is to lower the survivor teams efficiency by chasing them away from generators in any way you can, as well as causing regression, and all round applying the pressure of your presence to as much of the map as possible in order to buy you enough time to kill all of them.
As survivor your aim is to keep the efficiency as optimal as you can for as long as you can, prolonging chase to buy your team some time, getting people of the hook to keep your numbers optimal, etc. I don’t think that a loss of efficiency is a ‘problem’ so much as it is a ‘result’ of early game plays.
Now obviously you mention the sharpness of the drop being disproportionate to the percentage of players left in the game, and while yes in the example of stages you gave those numbers appear to add up, what about the time it takes for the killer to even find someone to chase in the first place. Before the first survivor is found then have an opportunity to operate at 133% efficiency (using 3 survivors working on a generator as 100%).
There’s also a lot of chance to consider too. The selection of perks, items, add ons, and some offerings can impact the survivors efficiency at the start of the game. Hex: Ruin can lower it, Leader can improve it. Perks like Sprint Burst and Decisive Strike are aimed at prolonging a chase, therefore giving other players more time on a generator.
In conclusion (in case I haven’t explained myself well) my stance on the matter is that the reason that the game can become super unbalanced at times is because it is the job of either side to take the balance at the start and tip it into their own favour. Sometimes one side does a better job than the other of taking control, and sometimes the fickle finger of Lady Luck just wont point your way.
2 -
Is it me or does “Stage 4” seem to be the one that goes on most of the game?1
-
aaaaand this is why games shouldn't start if 4 people aren't present.
It's also why a lot disconnect after someone else has disconnected 30 seconds into the game
4 -
This is why a killer is smart to tunnel. It's the most efficient move.
3 -
this is also why camping is such a pain to deal with... it's an automatic tipping point in the killers favor assuming they have hex ruin (or some other perk to slow genrush) or the survivors are too altruistic.
I honestly don't care if they tunnel or not, that's usually just target prioritization.0 -
It is not much information. His point is that the game gets harder with less players on the survivor side.alivebydeadight said:can you sum it up for me, im too lazy to actually read it
He does not like it, it is too severe for him.
Now, to give it seemingly some meaning he incorporates some mathematics to calculate percentages. That way, he does not have to phrase it, preventing it from appearing as either obvious or whiny.1 -
@DrVonKrumm said:
I found this to be a rather interesting problem to ponder so thanks for that.However, it occurs to me that surely solving the ‘death-efficiency problem’ is the point of the game. As a killer your aim is to lower the survivor teams efficiency by chasing them away from generators in any way you can, as well as causing regression, and all round applying the pressure of your presence to as much of the map as possible in order to buy you enough time to kill all of them.
As survivor your aim is to keep the efficiency as optimal as you can for as long as you can, prolonging chase to buy your team some time, getting people of the hook to keep your numbers optimal, etc. I don’t think that a loss of efficiency is a ‘problem’ so much as it is a ‘result’ of early game plays.
Now obviously you mention the sharpness of the drop being disproportionate to the percentage of players left in the game, and while yes in the example of stages you gave those numbers appear to add up, what about the time it takes for the killer to even find someone to chase in the first place. Before the first survivor is found then have an opportunity to operate at 133% efficiency (using 3 survivors working on a generator as 100%).
There’s also a lot of chance to consider too. The selection of perks, items, add ons, and some offerings can impact the survivors efficiency at the start of the game. Hex: Ruin can lower it, Leader can improve it. Perks like Sprint Burst and Decisive Strike are aimed at prolonging a chase, therefore giving other players more time on a generator.
In conclusion (in case I haven’t explained myself well) my stance on the matter is that the reason that the game can become super unbalanced at times is because it is the job of either side to take the balance at the start and tip it into their own favour. Sometimes one side does a better job than the other of taking control, and sometimes the fickle finger of Lady Luck just wont point your way.
To respond to:
However, it occurs to me that surely solving the ‘death-efficiency problem’ is the point of the game.
The Death-efficiency problem is not that survivor efficiency decreases upon the death of a survivor. This is obviously part of the game.
It is, as you recognise, the problem that it does so disproportionally, due to certain mechanics of the game.No kills and the 1st kill is the difference between night and day. The difference between 100% efficiency and 0.
A place where everything gets done and where 0 does.This is why a killer is smart to tunnel. It's the most efficient move.
It mostly only requires them to tunnel the first person. After they have gotten rid of them without the survivors having had the opportunity to push many gens, their game is basically over.
A killer can often stop tunneling after his first kill and still consistently win the game.
That shows just how strongly this problem is manifested.This reminds me of some stats that were collected from Gears of War tournaments. Up until Gears 4, the competitive mode was 4v4 Execution (a last-team-standing elimination mode for those that don't play). The game is broken into rounds, and you need to win 4 rounds to win the map (and you usually play best of 3 or 5 maps).
>
Someone did an analysis of "first death" and the outcomes and found that the team that got the first kill had something like a 75% chance to win the round (ie. the game drops from 4v4 to 4v3). The team that got the first 2 kills had a 95% chance to win. And if you got the first 3 kills you have over a 99% chance to win (basically a 4v1, you can't come back from that except in EXTREME circumstances).
Correct. In games that take their competitive side very serious see this as a great issue. League of Legends dealt with this snowball effect by making a killed champion worth less gold upon being killed frequently. This forces the winning player to have to find other players of the enemy team to snowball further, thus taking into consideration everyone's decision making.
I don't think you understand the problem. The problem isn't:
that the game gets harder with less players on the survivor side.
The problem is that there isn't much of a game after the first survivor dies.
This means that:
A. The killer gets completely beaten up, by failing to end the first 1/2 stages. (unfun/bully experience)
B. A first kill spirals into a 4 kill. (Killer OP survivor claims)This is why you, in case of an early death, rely so much on the Hatch mechanic.
The premise for solving the Death-Efficiency Problem
This is the ideal world: (taking 4 survivors working on a Gen as 100% this time)
From:
4 survivors alive: 75% working efficiency
3 survivors alive: 15% avrg working efficiency
2 survivors alive: 0% working efficiency
1 survivor alive: 0% working efficiencyTo
4 survivors alive: 100% working efficiency
3 survivors alive: 75% working efficiency
2 survivors alive: 50% working efficiency
1 survivor alive: 25% working efficiency
(reduced by 25% when in a chase)So what does this mean?
It means that the last survivor still has the ability to progress the game at a 25% of the total survivor power."But survivors already have this speed, what do you mean?", you ask. Well, they don't, and for the simple reason: A killer can patrol quickly enough to regress the Generator you're working on, checking the area and patrol further, before you could've switched to another one, let alone get any progress done.
The point is to make survival for lower survivor counts last longer, meaning that they have to endure longer under dangerous and deadly conditions.
It means that 1 survivor should be able to win against a killer, if he manages to play 4 times better (thus 4 times longer) under conditions of being haunted by a killers presence.
It means that there are no mechanics that unfairly cripple low survivor counts through any dynamics of the game.1v1ing a killer in a 5 Gen game? It would be possible, but very unlikely. Unlikely, not because it demands the killer to make a severe mistake or mis-patrol, but unlike as in the duration of you playing flawless is extended by 4 times the time.
In addition.
If we managed to achieve the previously mentioned, then follows an interrelated balance change:
The base conditions for escape are increased in difficulty. Meaning that it'll be less likely (along fair lines) that all 4 escape, but also with the lack of spiralling into a 4k.If 1 or 2 team-members died and there are plenty gens to do, don't worry; but this requires twice your patience.
quick mention:
I mention, taking the solving of the Death-efficiency as in its perfect state, that 1 survivor would take 4 times the time to open the exit games. This is of course untrue. Even with the disproportionality of Gen-Efficiency solved, it would still require low survivor counts to play extra carefully, thus adding a disproportional amount of time needed, but this time they at least have a chance.
Then again, I will cover a plethora of approaches and solutions later when I have them all prepared. The forum is typically a bad place to communicate interrelated ideas, especially when they come in separate posts.
2 -
I’ve given it some more thought but I’m afraid I still don’t see a problem here. Yes you could argue that the drop in efficiency is disproprtionate after the first kill, however I would argue that it’s not.
Let’s say for example that it takes each survivor at the start of the game 10 seconds to find a generator and start working on it, 10 seconds later one is found by the killer and chased away, as long as that survivor can keep the killer busy for another 70 seconds (a feat not all that difficult if we account for the number of available pallets, choice of perks, and the time it could take to pick up, transport, and hook the survivor), then that’s three generators done. One survivor goes for a rescue and the other two find and start the two final generators.
The problem with my example though is that it is only taking into account a very specific set of assumed values, a flaw that, I hope you won’t mind me saying, I also find in your example. There are far more factors that impact the game than just who’s working on a generator at any given time. Aura reading on either side can impact how long a survivor can stay on a generator for example.
I’d also like to add that having a survivor hooked will only drop the efficiency of other survivors if they choose to leave the generator immediately to attempt a rescue, but the first hook state is 60 seconds, that’s ample time to finish your generator then go in for a rescue.
In regards to your example of a lone survivor having little to no chance of winning with 5 generators left, well, I’m not so sure that they should have a fair chance to survive. I don’t want to sound harsh (because I’ve personally been in this situation) but if someone ends up in that situation then it is most likely that they are part of a team that has failed to act as such and are punished severely for it. When playing survivor I recognise that I could play my game perfectly but I’m still at the mercy of team mates decisions throughout.
0 -
Then look at it from this way:
We remove the Hatch mechanic. (Which is the quick fix or free out of jail card for dealing with the the rock bottom of the Death-Efficiency Problem)
(The hatch mechanic is after all a way of saying: "We know our low-survivor game malfunctions. Here have a mini-game when entering this part of the game; play find the hatch instead.")Maybe you start seeing it now?
1 -
I do see the point you’re making, I’m afraid I just don’t agree that it’s a problem that needs solving. Obviously though I’m not the one you’d need to convince, that’d be the developers, so I wish you all the best in that endeavour.
Just to add though that while I don’t see the situation as ‘problematic’ I’m also not opposed to seeing interesting new mechanics brought to the game that might make the it more fun while addressing the situation for players like yourself who do see a problem.
1 -
-
@AlwaysInAGoodShape - This also illustrates quite well why Killers can be struggling to keep up as more generators are done and less survivors being dead.
1 -
@Incarnate said:
@AlwaysInAGoodShape - This also illustrates quite well why Killers can be struggling to keep up as more generators are done and less survivors being dead.Exactly! It creates problems on both sides!
Either then killer gets victimised and steamrolled or the survivors become hopeless and hatch dependant!
2 -
@AlwaysInAGoodShape said:
@Incarnate said:
@AlwaysInAGoodShape - This also illustrates quite well why Killers can be struggling to keep up as more generators are done and less survivors being dead.Exactly! It creates problems on both sides!
Either then killer gets victimised and steamrolled or the survivors become hopeless and hatch dependant!
Yep - it's a fundamental design and balance flaw we're all suffering because of.
1 -
I think giving survivors an increased speed for repairing and healing for each dead survivor would be the best way to solve this problem. Like a 40% increased repairing speed for every dead or disconnected survivor, and a 20% increased healing speed for every dead or disconnected survivor. I do feel like this should only be applied once gen rush has been nerfed a bit though.
0 -
@ad19970 said:
I think giving survivors an increased speed for repairing and healing for each dead survivor would be the best way to solve this problem. Like a 40% increased repairing speed for every dead or disconnected survivor, and a 20% increased healing speed for every dead or disconnected survivor. I do feel like this should only be applied once gen rush has been nerfed a bit though.The problem isn't just for survivors/victims, but also for the killers - it's doublesided issue, as I just mentioned before - "This also illustrates quite well why Killers can be struggling to keep up as more generators are done and less survivors being dead." So if you were giving remaining survivors/victims an increased repairing speed you would just be making it much harder for the killer.
1 -
@Incarnate said:
@ad19970 said:
I think giving survivors an increased speed for repairing and healing for each dead survivor would be the best way to solve this problem. Like a 40% increased repairing speed for every dead or disconnected survivor, and a 20% increased healing speed for every dead or disconnected survivor. I do feel like this should only be applied once gen rush has been nerfed a bit though.The problem isn't just for survivors/victims, but also for the killers - it's doublesided issue, as I just mentioned before - "This also illustrates quite well why Killers can be struggling to keep up as more generators are done and less survivors being dead." So if you were giving remaining survivors/victims an increased repairing speed you would just be making it much harder for the killer.
True. Which is why I think a gen rush nerf would be needed in order for something like this to be added.
1
https://www.youtube.com/watch?v=s7vG49IiNQ8