PC Stealth, monster closets and clear quests

As someone who opened TWO threads about this (although different issues) I can say that I like the idea of hidden sleepers.

The fix is twofold:

1: reduce hidden sleepers. By about 90% (maybe even more. I am serious. More than a hand full and you expect them, learn where they could be and check every cupboard kinda like how behind every picture was loot). Anything more than that feels cheap and you start expecting it.

2: auto wake up Z's when you LEAVE the volume at the exit

This is far more dramatic since your escaperoute is cut off and you solve the backtracking. (Dont autofocus them. Set them to wander and let stealth do the rest.)

3. NO RNG WAKE UPS!!!

Dont.

Increase sleeper hearing by 200%, but don't do it via checks/dicerolls

 
Last edited by a moderator:
I have a tendency to overlook the ability to throw a rock as a distraction in stealth games in general, and that's on me.  I tried it out since you mention it, and it's curious.  If you're loaded up on stealth items so the zombies don't detect you, and you throw a rock, then any already-alert zombies will chase after it.  But you can't wake up a zombie in the first place by throwing rocks.  I tried everything up to and including throwing it right into a sleeper's face, and it stubbornly wouldn't wake up.

Throwing a rock to wake up nearby sleepers, without alerting them to you exactly, does seem to me like a good way to break the deadlock, without betraying your stealth bona fides.  And all the parts are there already, we just need sleepers to hear the rocks.  It's a good idea.
Thanks. Using decoys is good stealth gameplay in general imo, and it gives more to do than just headshot sleeping zeds.

I also think we need more patrol zombies, or if thats too hard just zombies that wake up and wander around the house, it would be more fun having to hide from a zombie wandering around than zombies that are hiding from you. Even if the zombies in the end of a POI that they want you to engage in wake up but dont sense you, so you need to lure them away and take them out one by one. 

I think the current stealth is pretty much just missing those two elements, patrols and distractions, put those in the game and the stealth gameplay would be much more entertaining. patrols would increase the challenge while distractions would allow an element of control. the distraction element is already there, they just need to make distractions wake up zombies without alerting them to you. patrol paths might be beyond the scope of their interests but awake unalert zombies already exist that could just wake up as soon as you enter a poi and wander randomly. that should be fine too.

 
auto wake up Z's when you LEAVE the volume at the exit


Was thinking about this earlier. If you fail to discover the hidden zombies, wake them up when you leave the volume. Keep the number of potential hiding spots, but reduce the number of hidden zombies so their positions are randomised each time a poi is reset.

but don't do it via checks/dicerolls


A check that is based on your skills is fine, I think. Pure RNG that ignores your skills is not.

If you're maxed out on stealth then having a minimal hard fail rate is ok. Consider the critical fail on a roll of 1 in DDO (based loosely on 3.5 ruleset). That equates to 5% guaranteed fail rate, which I think is fine. You could go days or weeks and never encounter a hard fail (or conversely have a real string of bad luck :devilish: ). Each point in stealth skills reduces the chance of a fail until it hits that minimum threshold. Might be worth requiring stealth gear (military stealth boots) and mods (advanced muffled connectors) to actually reach that minimum 5% so sneaking around in tap shoes, eh, steel boots isn't viable at any skill level. Perhaps even incurring a stealth penalty if you are over encumbered - ninjas were renowned for not carrying around pockets full of rocks and tin cans, if I recall my history lessons correctly :D  

 
Last edited by a moderator:
DnD and 7D2D are fundamentally different though.
D&D is storybased. So the game IS the story. So failure is part of the story.
7D2D is a survival game. So if we extrapolate this to the extreme, how would you feel if there was a dice roll for you to die or not?

The game should only punish you, if you do something wrong. So if you get hit, you have a chance to get debuffed.
If you are too loud/walked over trash, the Z's have a chance to wake up.

But if you did everything right and are still getting punished, it doesnt feel great.
The point at which it doesnt feel unfair anymore is the one where the chance is so small as to be nearly nonexistent (talking about <1%)

I see NO reason why you shouldn't just increase hearing from the sleepers by a large amount. I mean they don't wake up if I go over 30 for 1 sec, but if I continue to stay over 30 they wake up.
So lower this number to 20 for normals, 15 for ferals 10 for irradiated 5 for irradiated ferals (do those still exist?)

 
D&D is storybased. So the game IS the story. So failure is part of the story.


DnD is setting based. The story is derived from the actions the players take within the scope provided by the setting and the DM. In 7d2d, TFP are the DMs and provide the setting, we continue to be the players.

No need to extrapolate - we're not talking about some magic death spell. We're talking about investing skill points, clothing slots and mod slots in one specific skill (stealth) to maximise your chances that you can reasonably safely sneak without waking up inert zombies (don't forget, the current system wakes them up regardless of what our skills and equipment suggest). It isn't just our actions that influence the outcome of an event...

But if you did everything right and are still getting punished, it doesnt feel great.


The 5% failure rate is just an arbitrary number, selected because of d20. It would obviously need to be balanced to 7d2d - 1% might prove to be fine. <1% might be fine if the skill check is run on each zombie individually, rather than a volumes worth of zombies. Think of it as simulating a minor issue of, for example, getting your clothing snagged on a door handle or something - a small mishap irl but could have dramatic repercussions depending on the situation. The failure rate could decrease with distance - sound obeys the inverse square law after all.

Plus, we're both in agreement that sleepers should be reduced in number anyway - what I am trying to avoid is ending up where one is completely safe - there must always be a chance of failure; as in a lot of scenarios, the participant can do everything right but things outside of their control will end up dictating whether they have a bad day or not.

 
As someone who opened TWO threads about this (although different issues) I can say that I like the idea of hidden sleepers.

The fix is twofold:

1: reduce hidden sleepers. By about 90% (maybe even more. I am serious. More than a hand full and you expect them, learn where they could be and check every cupboard kinda like how behind every picture was loot). Anything more than that feels cheap and you start expecting it.

2: auto wake up Z's when you LEAVE the volume at the exit

This is far more dramatic since your escaperoute is cut off and you solve the backtracking. (Dont autofocus them. Set them to wander and let stealth do the rest.)

3. NO RNG WAKE UPS!!!

Dont.

Increase sleeper hearing by 200%, but don't do it via checks/dicerolls


Too bad, my pointer was already poised over the "like" button, but then I read point 3. 😁

DnD and 7D2D are fundamentally different though.
D&D is storybased. So the game IS the story. So failure is part of the story.
7D2D is a survival game. So if we extrapolate this to the extreme, how would you feel if there was a dice roll for you to die or not?

The game should only punish you, if you do something wrong. So if you get hit, you have a chance to get debuffed.
If you are too loud/walked over trash, the Z's have a chance to wake up.

But if you did everything right and are still getting punished, it doesnt feel great.
The point at which it doesnt feel unfair anymore is the one where the chance is so small as to be nearly nonexistent (talking about <1%)

I see NO reason why you shouldn't just increase hearing from the sleepers by a large amount. I mean they don't wake up if I go over 30 for 1 sec, but if I continue to stay over 30 they wake up.
So lower this number to 20 for normals, 15 for ferals 10 for irradiated 5 for irradiated ferals (do those still exist?)


Chess is a deterministic game. A survival game/open world game on the other hand is not.

 
Last edited by a moderator:
Too bad, my pointer was already poised over the "like" button, but then I read point 3. 😁
Pfff! I can live without your like! 😑

...

😕

...

☹️

...

🥺 okay, okay I will admit that point three was just a rant and not too serious (although I still feel it is valid :D) that is why I said I have 2 points :D

now please give me the like!  😫

Chess is a deterministic game. A survival game/open world game on the other hand is not.
Ch-chess??? o_o Who talked about chess? xD Did I miss something?
 

But seriously EVERY game that is about survival and/or competition needs to be fair. And even if some games use rng to simulate drama, once the player notices, it is ONLY a detriment. And TFPs have not filled me with confidence when it comes to subtlety.
*sneaks through Hotel POI where EVERY GOD DAMN ZOMBIE is in a room that has auto-aggro behind walls*

Just imagine in darksouls, if attackpatters were recognizable 99% of the time and then suddenly, without a warning, you are dead because "you can't predict every hit in real life" WTF? :D Not fair, not fun, no reason.
Even worse: this doesn't happen unless you haven't been hit for 2 minutes, because you went the dodge build. And you need some drama or the game will get boring!
You don't lose anything, so why do you care?

                         NO!

 
I'd have to agree with Vik here, deterministic is better, but it could well have a "realistic" random element to it.

The way I read "there's a random fail-check at entry to a volume" is basically just a 5% chance of "ok, they're all coming at me for no reason again." That feels bugged whenever it happens.

If you were to actually have a per-sleeper random level of detection, and no mechanic to aggro the full volume once you wake up one, you would at least see that not everyone wakes up and it would feel more natural. Plus it could even be useful for luring out some and not all at once. I think these elements are in the game already, watching some tubers occasionally wake up only some; but for my personal experience, it feels really binary whether a volume wakes up or not. Maybe it's just really out of balance at the moment.

 
I say it again:
If you absolutely HAVE to have sleeper volumes (they have only downsides, when compared to the normal sneaksystem) use them in the whole room, not on a single block.

But the problems of sleeper volumes are so vast, I can't see any justification for them.

They feel unnatural
-why does the Z' 15m away wake up?
-I moved at snails pace with 1 visibility, this is just unfair!
-why did EVERY zombie wake up at the same time?
-so I can knock out the wall and it doesn't wake them up, but stepping on this magical block does?
-why do they only hear me on entry and not while I am in the room?

and a lot more that won't come to my mind right now.
Just use the basic system and lower the requirements for sleepers to hear you.
Hell you can still use the volume system by waking up every zombie once one zombie breaks a "hidden" wall (and makes a sound).

 
My problem with the monster closets is you don't know the monster is there. It doesn't show up on your radar.

So you can't properly clear the area unless you get out of stealth. 

 
Clears should not be 100% of the zombies anyway if hidden zombies will be in 10% of full poi and you need to clear 90%+final zombies on a loot of poi should it be better?

 
Most mission POIs have a generator in them right?

Seems like starting up the generator and having all the lights come on in the POI should be enough to wake up any remaining zombies in closets so they can be killed to finish the mission.

 
Sleeper volumes get alot of hate but without them, performance would be horrible and there would be no pacing whats so ever in dungeon POIs.

Back in the day, there was only one big volume and all of the zeds would spawn in.  Not only does this strain the max alive zombie count but also made clearing POIs pretty one dimensional.

One reason ood movies and TV shows are entertaining is because they have good pacing and build up to a climax.

My problem with the monster closets is you don't know the monster is there. It doesn't show up on your radar.

So you can't properly clear the area unless you get out of stealth. 


A well designed quest POI will lead the player the right direction without having to relay on the red dots on the compass.

Additionally, there is the orange dot for volumes that were missed by the player. 

 
Last edited by a moderator:
performance would be horrible


I'm assuming that due to the fully destructible environments, pre-calculated navmeshes (or whatever they'd be called in 7d2d - my familiarity only goes as far as Skyrim/FO3-4 modding) aren't possible, even for prebuilt POIs? Is this the main performance issue? Assuming so, and spitballing here: could groups of zombies close together share the same path until an individual zombie separates far enough away from the group to warrant generating its own path? 

edit: Acksherly... I've noticed quite a few zombies spawn outside a POI and converge when the player enters a POI. Is this really necessary? Doesn't this just add to the overhead? Perhaps those cpu cycles are better spent inside the POI, or at least have these external spawns randomised (10%/25% chance for example) and/or reserved for less demanding POIs so they come as more of a surprise and can't be so readily anticipated - see point below re over-curating.

no pacing


At times I wonder if TFP tries to control too much - build the POI, set the ai loose and what will be will be. Perhaps these curated experiences could be reserved for larger POIs that demand the optimisation? Houses and other similarly sized POIs could be left to their own devices? (Again, just spitballing - I don't have the expertise to determine feasibility)

 
Last edited by a moderator:
I'm assuming that due to the fully destructible environments, pre-calculated navmeshes (or whatever they'd be called in 7d2d - my familiarity only goes as far as Skyrim/FO3-4 modding) aren't possible, even for prebuilt POIs? Is this the main performance issue? Assuming so, and spitballing here: could groups of zombies close together share the same path until an individual zombie separates far enough away from the group to warrant generating its own path? 

edit: Acksherly... I've noticed quite a few zombies spawn outside a POI and converge when the player enters a POI. Is this really necessary? Doesn't this just add to the overhead? Perhaps those cpu cycles are better spent inside the POI, or at least have these external spawns randomised (10%/25% chance for example) and/or reserved for less demanding POIs so they come as more of a surprise and can't be so readily anticipated - see point below re over-curating.

At times I wonder if TFP tries to control too much - build the POI, set the ai loose and what will be will be. Perhaps these curated experiences could be reserved for larger POIs that demand the optimisation? Houses and other similarly sized POIs could be left to their own devices? (Again, just spitballing - I don't have the expertise to determine feasibility)


No clue about nav meshes but probably to late in development to switch to a completely different system.

Those who experienced earlier alphas can attest how broken spawning was before sleepr volumes. 

A player could run into the middle of the old hub city at coordinates 0,0 and attract so many POI spawns then drag them out into the street kiting them for blocks until the spawn limit was reached causing  POIs all over the map to have no or delayed spawns.

After being around since Alpha 6, my personal feel is that a balance of non questable POIs and dungeon POIs is the gameplay sweet spot.

It allows some POIs (non questables)  to be free-form multiplayer levels with many ways in/out and others (questable) to have a strong character / story.

 
Last edited by a moderator:
Pfff! I can live without your like! 😑

...

😕

...

☹️

...

🥺 okay, okay I will admit that point three was just a rant and not too serious (although I still feel it is valid :D) that is why I said I have 2 points :D

now please give me the like!  😫

Ch-chess??? o_o Who talked about chess? xD Did I miss something?
 

But seriously EVERY game that is about survival and/or competition needs to be fair. And even if some games use rng to simulate drama, once the player notices, it is ONLY a detriment. And TFPs have not filled me with confidence when it comes to subtlety.
*sneaks through Hotel POI where EVERY GOD DAMN ZOMBIE is in a room that has auto-aggro behind walls*

Just imagine in darksouls, if attackpatters were recognizable 99% of the time and then suddenly, without a warning, you are dead because "you can't predict every hit in real life" WTF? :D Not fair, not fun, no reason.


Just imagine Darksouls were not a game where any small error means almost automatically your end. Just imagine Darksouls were not a game where you need to memorize what moves the enemy makes like in a jump-and-run. Then it wouldn't be Darksouls anymore but a game where you could add randomness without problem.

My point: When a room or part of a room wakes up because of a random roll you are not automatically dead, it is just a dynamic situation you have to adapt to. If YOU are automatically dead (or even most of the time this happens) then you play on a difficulty level too high.

Now since stealth was changed recently even I say that balance is probably not quite there yet. So if random wake now means almost all zombies wake up every time even with excellent stealth then that is not an intrinsic feature of randomness but just the balance of such randomness being out of whack.

7D2D on the whole should be fair (for some definition of fair). A good balance means any random element (before the die is thrown) is fair. But the random result is usually not fair. If two players each open a box with the same loot table, that is fair between them. But one finds the OP weapon, the other a stack of leather. That one box is not fair anymore. But after opening hundreds and hundreds of boxes we can again say it was fair.

Example war games: Each single game often turns somewhat "unfair" because one of the players seems to have much more luck with his dice. Still war games are played competitively. Because usually all that evens out and better players still win more often on average. For some definition of fair war games are fair and the random element provides variation and a chance for worse players to sometimes win.

So what is your exact definition of fair and at what level of the game you want to apply it? A competitive game and definitely a survival game only need to be fair overall. I see no reason for it to be fair on the micro-level of any situation.

You enter a room, ideally 0 to all zombies wake up, the more stealth you have and the more precautions you followed it should be more likely 0 or a very low number. This is how it easily can work in 7D2D, while in some other stealth games where detection means 90% sure death it would be clearly wrong.

Even worse: this doesn't happen unless you haven't been hit for 2 minutes, because you went the dodge build. And you need some drama or the game will get boring!
You don't lose anything, so why do you care?

                         NO!

 
Last edited by a moderator:
Has anyone in here actually tried stealth (perked) with attack volumes (feral sense off)?

Light and shadows is very important now.  The zombie will wake up but they may not see you if you manage your light, shadows and noise accordingly.

 
Has anyone in here actually tried stealth (perked) with attack volumes (feral sense off)?

Light and shadows is very important now.  The zombie will wake up but they may not see you if you manage your light, shadows and noise accordingly.


In my somewhat limited experience (haven't played with feral sense on, yet) stealth is doable, and becomes easier the more you invest in it. As with the majority of changes made over the course of the development of this game, you just have to re-learn how it works and adapt.

 
Has anyone in here actually tried stealth (perked) with attack volumes (feral sense off)?

Light and shadows is very important now.  The zombie will wake up but they may not see you if you manage your light, shadows and noise accordingly.
Yea I did. While I am not sure where these volumes are, I did extensive testing for my 2nd thread. And I actually found out that zombies lose you extremely quickly if oyu hide. Which is actually a lot of fun. (this wasn'T always the case and like most veterans, learning something new is hard :D so I just always acted as if they will know where I am once spotted. Now I hide in a closet and stab them from behind. Very rewarding)

But it does not feel fair, since the stealth indicator is actually not an indicator. I can have 8 stealth and everything wakes up 90% of the time and sometimes I just sneaksprint my way through half the poi with 30 visibility.
And "shadows are important" is the reason why this is so stupid. Because there is not a way to see where you are illuminated AND lightsources are either nearly indestructible (from afar) or obstructed. So it is like "hey this is the path!" but then there is a flashlight pointing right at it and you just think "oh god why??

It needs a LO'T more balancing before this feels anywhere near fair and good.

 
Sleeper volumes get alot of hate but without them, performance would be horrible and there would be no pacing whats so ever in dungeon POIs.

Back in the day, there was only one big volume and all of the zeds would spawn in.  Not only does this strain the max alive zombie count but also made clearing POIs pretty one dimensional.

One reason ood movies and TV shows are entertaining is because they have good pacing and build up to a climax.

A well designed quest POI will lead the player the right direction without having to relay on the red dots on the compass.

Additionally, there is the orange dot for volumes that were missed by the player. 
 I will sneak right by them. I got to the end of a mission and only encountered three Z's. It was a fetch/clear and I had the package. There were no red or yellow dots showing up on my radar because they're all hiding in the walls.

I went back to the start, unstealthed, and went full Rambo on them with my SMG. That was fun too.

 
Last edited by a moderator:
Back
Top