Sneaking

Game Talk The Drawing Board
41

Roguelikes typically solve it with skills and rolls and it works better that way. Brogue, Sil, Crawl, Forays into Norrendrin - all have workable stealth in the form that could be introduced here.

Project Zomboid is a completely different game that relies on real time and delayed insta-kills. That example doesn’t really answer any questions and doesn’t offer any possible course of action to make workable stealth in a turn-based game with no insta-kills (other than headshots).
That turn based part is absolutely vital here. Being able to scan the area with a mouse is totally fine as a requirement in a real time game, in DDA a mouse support can’t even be guaranteed.

42

Ye, but I just hate this systems. They works well but is just stupid.

I said: “how to make these mistakes more deadly - different story”. But if you want here is: you can be grabbed by something from behind and bitten in the neck or head, recive a lot of damage, bleeding and etc. This is different story about how to make this more intresting, but with rolls and skills - this at least will look stupid.

Just front and back sight will be enough. This is not about a lot of microcontrol and mouse required.

Okay. I see. You all think that this is stupid, boring idea which can’t be realised. Fine, I shut up.

43

How about my idea of camouflage based sneaking? It’s completely dice-free.

Also, driving an 8 wheeled military composite armored deathmobile with periscope is the ultimate sneaking.

44

If you somehow managed to bring up anything new to the table that would somehow fix the massive glaring flaws with the whole idea, it could work.
But so far you keep bringing up examples that you don’t describe more than “zombies will somehow sneak up on you”, while not having anything to say about the giant problem of having to manually look around, which makes the whole thing way less sensible than skill rolls - both when it comes to game design and making the whole thing realistic.

It’s similar to my attempted implementation (a year ago or so).
My idea was that everything has a visibility multiplier: creatures around 0.5 (similar to ranged size), “tall” terrain (walls) 1.0, flat terrain something lower (possibly even less than creatures), vehicles something in between.
This visibility multiplier would essentially be a multiplier on range from which the object can be spotted, assuming all other visibility calculation elements remain constant.

45

That’s realistic. How would it display? Would distant wall appear on unseen black space because the ground is less visible? That would look strange but actually quite realistic when we project 3D view to 2D display. Probably we can get used to that.

46

The issue is that this still doubles or triples the number of key presses required to do what are currently trivial actions that a player must perform a large number of times.

To add FoV to players such that a zombie might sneak up on you would increase the tedium of actually playing the game by potentially 100%.

The benefit being…
Inattentive players who don’t want to double their keypresses might get ambushed by something, sometimes?

The change here can be summed up as “Players now have to hit keys twice as often to look around or search or notice things, and failure to do so has a small chance of sudden death”.

A stealth system doesn’t require player FoV and works perfectly fine to solve the issue of ambushes without making every other part of the game more tedious.

If a zombie further away is slightly harder to see, or if a zombie in tall grass is slightly harder to see, or if a zombie just moved it’s easier to see, or if desks and tables and chairs and windows make things slightly harder to see past them, etcetera.
There could also be an awareness value based on things like whether you’re sneaking or running, or if you’re hungry or thirsty or tired or sad which would make you easier to sneak up on.

The only way I could want an FoV being partially implemented is if it simply modified the visibility of creatures opposite of any direction you just moved.
For example if your last action was to walk to the north, things to the south would get a visibility modifier similar to simply being further away.

47
The only way I could want an FoV being partially implemented is if it simply modified the visibility of creatures opposite of any direction you [i]just[/i] moved. For example if your last action was to walk to the north, things to the south would get a visibility modifier similar to simply being further away.

even this I would hesitate to call an “improving” mechanic especially in situations where the player is doing a lot of back and forth. Perhaps if it was a buildup to a still fairly minor penalty the longer the player went in “one” direction, with each movement changing penalties like so:
NW N NE
W . . .E
SW S SE
penalties:
NWP = __ NP =__ NEP = ___
etc…

So the proposed vision system would be similar to the health system (in that the system constantly moves a value towards another value) but with no random on the penalties/bonuses

standing still would be a target bonus value of… lets say 5 in all directions
moving N would move N penalty/bonus value towards 0
NE and NW towards -1
W and E towards -3
SW and SE towards -8
and S towards -10

value movements would be 1 in cases of > 5 and 2… maybe 3 in cases of <5 difference between target and current

Then these penalties would be applied to mob stealth bonuses, player perception(after exausted/focus penalties or perks) etc…

No I think that gets far to complicated on a per-turn caculations level. That seems like about as simple as it can get too. unless you want it to be a 2D direction last moved = 0 opposite = -10 which would lead too the tedium of run backwards every so often, or just sit still every so often or face ambushes from behind… tedium?

I like the idea, but I can’t imagine how it would work with a future CDDA stealth system without bogging down everything a bit.

A flat mobs terrain/movement stealth system and player system would be easier for players to understand through play, and would probably be a required prerequisite anyway to installing a directional movement penalty to perception system

Of coarse the main problem would be in large numbers of mobs such as a zombie horde which could be minimized by just saying if Z = horde member then skip all stealth calculations, and doing a similar thing where visible lit area is minimal, such as when entering a relatively small room.

I would like to point out that I am assuming terrain remains visible for referance, but perhaps “greys out” with tiles the “less aware” of an area it is. However it would be basically impossible to do in ASCII this would quickly ostracize ASCII players and accessibility (blind players) for any fork that went this direction or force them to play in a manor they don’t prefer.

But then Im certainly no SME (Subject Matter Expert) so I’d like to hear what those that know what they’re talking about think about something down this alley.

48

This seems far more complicated to do, I was mostly thinking of a fairly small effect that would simply be amplified by things like exhaustion or traits that reduce your awareness.

No I think that gets far to complicated on a per-turn caculations level. That seems like about as simple as it can get too. unless you want it to be a 2D direction last moved = 0 opposite = -10 which would lead too the tedium of run backwards every so often, or just sit still every so often or face ambushes from behind.... tedium?
This is more or less exactly what I would consider a legitimate gameplay change. My numbers were more like 100% awareness in front and 75% awareness behind, though.

The tedium is minimal to non-existent and unless you have very poor awareness you’d still notice zombies moving around behind you, they could just get closer than normal if you only ever moved in one direction.
Also, I was thinking of applying a ‘maximum’ penalty based on other variables, such as whether you’re running, walking, or sneaking.

If you’re just walking you’d never reach the point of getting hit from behind unless you were basically falling asleep already.
Running would decrease general awareness in every direction except ahead of you, but you’d be moving faster anyways so it wouldn’t matter a lot.
Sneaking would lower your visibility and increase your awareness at the cost of speed and maybe height (so you could see over fewer obstacles but you’d notice more things that are still within your sight).

I would like to point out that I am assuming terrain remains visible for referance, but perhaps "greys out" with tiles the "less aware" of an area it is. However it would be basically impossible to do in ASCII this would quickly ostracize ASCII players and accessibility (blind players) for any fork that went this direction or force them to play in a manor they don't prefer.
I was going to assume the player has perfect memory regarding terrain, furniture, and traps. A FoV indicator seems a bit overkill though, honestly.

My suggestion was mostly just to add a slight tweak to an actual stealth system towards a given direction, by far the overwhelming factors should be distractions like tiredness, fatigue, focus, traits, obstacles, enemy visibility, sound, etc.

49

Even if FOV is implemented, players would probably request an auto-look-around feature.

50

Yeah, my suggestion was more along the lines of adding more depth to an actual stealth system in a fairly intuitive way that doesn’t add too many keypresses and still allows for the turn abstraction.

Sneaking around would be the ‘auto-look-around’ feature that prioritizes maximum stealth+awareness.
If you’re walking one direction you can still look over your shoulder and hear stuff behind you, but if you’re running you don’t really have time for that.

Though I did start punching in actual numbers to see how it would work out.
For default and usual values the change would be primarily thematic, the average player might barely or not even notice that they have different visibility distances ahead/behind them when walking around (average morale/focus, not sleepy or fatigued, etc).

It would only really come into play when you were already suffering from lack of awareness such as being exhausted, in which case it’s simply adding a bit of urgency.
You’d already have trouble detecting nearby enemies that are obscured by the environment when in poor condition and it’s a perfectly reasonable limitation on the player’s local omniscience that you won’t notice a crawling zombie several tiles behind you when you’re having trouble staying awake to begin with.

Naturally if there were traits like “Oblivious” that lowered your overall awareness you could get snuck up on in a more regular situation.
But at that point you should have to deal with that kind of thing.

51

Is it just me or did the thread go from ‘the player can sneak’ to ‘monsters and environment will hide’?

52

It went from “simple hack to make player sneak” to “consequences of having (specific) visibility functions in the game”.
What we certainly don’t want is another hackjob like zombie grabs.

53

Conversation focuses on the controversial parts, having monsters not spot the player is totally uncontroversial and has no dramatic complications, so there’s not much to discuss.

54

besides one player or entities having anything aproaching a stealth mechanic means it makes things more tied together and whole if the other does too. Other wise parts of the game would begin to feel to disjointed. So impimenting one, means serious consideration should be given to how it should, or at least could be applied to other parts of the game.

[spoiler= @ Weyrling]

This seems far more complicated to do, I was mostly thinking of a fairly small effect that would simply be amplified by things like exhaustion or traits that reduce your awareness.[/quote]
yea probably the better way to go, I was trying to get too realistic, but your post made me think of this system, so I wanted to get it down in case it could be usefull.

No I think that gets far to complicated on a per-turn caculations level. That seems like about as simple as it can get too. unless you want it to be a 2D direction last moved = 0 opposite = -10 which would lead too the tedium of run backwards every so often, or just sit still every so often or face ambushes from behind.... tedium?
This is more or less exactly what I would consider a legitimate gameplay change. My numbers were more like 100% awareness in front and 75% awareness behind, though.

The tedium is minimal to non-existent and unless you have very poor awareness you’d still notice zombies moving around behind you, they could just get closer than normal if you only ever moved in one direction.
Also, I was thinking of applying a ‘maximum’ penalty based on other variables, such as whether you’re running, walking, or sneaking.

If you’re just walking you’d never reach the point of getting hit from behind unless you were basically falling asleep already.
Running would decrease general awareness in every direction except ahead of you, but you’d be moving faster anyways so it wouldn’t matter a lot.
Sneaking would lower your visibility and increase your awareness at the cost of speed and maybe height (so you could see over fewer obstacles but you’d notice more things that are still within your sight).


Agreed. The entire point of making it so complex was to
1: simulate that moving will make you less able to pick up SUBTLE things like a slow moving triffid in the forest that is minding its own.
2: keep a balanced directional awareness when moving so that doubling back for one space every so often doesnt get you magically seeing things you should not be able to take in so quickly is such a fast changed observational setting. but as you said, if your moving in the opposite direction you usually don’t have a lot of concern with what is behind you so usually people won’t be doubling back, so its unnecessary. again, I was getting to realistic.

I would like to point out that I am assuming terrain remains visible for referance, but perhaps "greys out" with tiles the "less aware" of an area it is. However it would be basically impossible to do in ASCII this would quickly ostracize ASCII players and accessibility (blind players) for any fork that went this direction or force them to play in a manor they don't prefer.
I was going to assume the player has perfect memory regarding terrain, furniture, and traps. A FoV indicator seems a bit overkill though, honestly.

My suggestion was mostly just to add a slight tweak to an actual stealth system towards a given direction, by far the overwhelming factors should be distractions like tiredness, fatigue, focus, traits, obstacles, enemy visibility, sound, etc.


again, I agree simpler is better here. My system was only to be a layer on top of such variables to give it more depth focus and perception would be the deciding factor that all these would act as “multipliers” too, but I think it would give too much depth that would have too little-to-no effect to be worth implementing, especially considering the added caculations it would me on a by-turn basis increasing the turn length. I’d love to see it, but it’s just too overly complex to be worthwhile despite how “simple” I kept it.[/spoiler]

55

Personally I think a “full” sneaking system would be, hands down, the best option providing the most fun to the game.

What I mean:

  • Restricted monster cone of vision indicated by a flashing direction arrow when player not sneaking and a cone when player is sneaking.

  • Restricted player cone of vision with disjointed controls for movement and looking, as in and because;
    - ability to move backwards (much slower with risk of tripping, which will solve gamey-ness of kiting)
    - while kiting, sniping or otherwise being preoccupied the player may decide it’s a good time to retreat only to be surprised by a horde of zombies right behind him, which sounds super fun and will actually increase the importance of player hearing. All the traits, systems and hard work that went into hearing will now pay off.
    - The need to constantly look around and scan for threats will add greatly to the thrill and tension of playing. It will also open the grounds for new traits and mechanics like paranoia in addition to schizophrenia, which would make the player hear and see things that are not there, making them look around. That’s the mechanic that initially made me fall in love with Project Zomboid back when it had that feature.

  • Back-attack vulnerability both for monsters and player. Maybe as simple as no dodge-skill.

  • Special monsters who specialize in utilizing back-attacks.

  • CBMs and mutations that give back-of-the-head vision or something like life/movement signature indicators. Alien roleplay/monster anyone? Maybe even somethig as simple as glasses with small mirrors, or a hand-mirror rifle attachment that gives limited vision behind the player. ACTUALLY!! Imagine playing as a BLIND character, relying solely on ‘super-long-whiskers’ mutation or CMB and movement signature/hearing extension CMBs/mutation that show enemies as red dots swimming in the sea of black. That’s SO awesome.

It offers so much awesome, the only question is; is anyone willing to put in the effort to make it.

Edit; By disjointed-movement controls I mean something as simple as making the player face the opposite direction when moving while pressing control. Slow-down and tripping is a bonus.

56

It’s not a matter of effort, a number of the features you outlined are fundamentally incompatible with the way the game works (vision cones) and others are extremely questionable from a “will it be fun” point of view (disjoint sight and movement).
This isn’t a new feature, this is overhauling how sight and vision work in the game.

57

Incompatible? If you want to hack it just place 5 invisible walls ‘behind’ the character and keep track of last movement to know where to put them. Just reverse that last part for disjointed movement. Not really sure what do you mean by saying it’s not compatible. Or do you mean the cone-indicators? Doesn’t driving give a direction indicator? How is that different?

And as I am just saying for myself - It would be extremely fun for me. Hell of a lot more fun than current shallow rng hack’n slashing hundreds of zombies with kiting which makes me drop the game after a couple of days every year I pick it up again.

Fewer but more varied, complex and involved fights definitely sounds like a lot more fun. Even disjointed movement actually makes kiting fun by introducing more uncertainty to the outcome, and adds incentive to actually do more in the game than just throwing rocks and kiting. After hundreds of hours in the game I never used any guns - why would I if throwing rocks is so much better and works every single time?

And if someone just loves omni-vision? Make it optional or let them use the CBMs and mutations I spoke of above. It sounds like a well tied together idea to me.

58

Imagine the screen gore from a horde of zombies all facing in random directions when they have flashing direction indicators…

59

I’m imagining and “gore” is not the word I would describe it as. It actually looks very decent.

60

I know nothing about the design of the systems, so this suggestion is probably way off base.

Assuming a zombie tile and the terrain that zombie is standing on is known, would it be possible to assign a ‘visibility’ probability to a zombie tile? Some chance the player will see a zombie in said tile vs. the terrain.

Things that multiply into that factor have already been stated: proximity to player front or backside, zombie stats, player stats, etc.

The mechanic could be very cool. Sometimes you see the zombie, then the next turn it disappears. Perhaps when it’s far away this effect is obvious, and maybe even when the zed is closeup, imagine that it’s hard to keep track of your closest angles. And focusing one zombie lowers your perception of others.

Biggest problem I see here is performance. But then again I’m not familiar with the setup so I have no idea if there are fast methods for updating the tiles each turn.

This idea sounded great in my head, and I’m typing this on my phone and likely not explaining my idea clearly. If there’s any Merit to this so I can explain in more detail.

Edit: I’m also not averse to trying to implement this myself, but the startup cost for me will be high without some direction :grin:.