Why stealth sucks and how to make it stop

[Norfleet]

much less able to know they should go check out that cardboard box that wasn't there the last time they turned their back.

You'd be surprised how well that actually works in real life. Most people do not, in fact, have an exact memory of the position of every cardboard box in their vicinity. They will only notice a change if it violates their "relevant topology" map of the environment. If the anomalous cardboard box does not interfere with what they are considering relevant, they will not notice it. If you put a cardboard box down in a place where there should be no such cardboard box, people will notice. If you put a cardboard box down in the middle of their path suddenly, they will notice. If there are 30 boxes in an out of the way corner which they are not paying particularly close attention to, and you add or remove another one of these boxes without doing anything to draw attention to those boxes, they will probably not. Especially if they are told to pay close attention to something else. At that point, they wouldn't notice even if a man in a gorilla suit walked through their immediate field of view.

 

[Skunkpew]

Other minor additions I would make to a Thief game AI:
- the sounds you make are more likely to be heard when the room is silent, less likely when people are talking/guards are walking and hear their own footsteps/there is noisy machinery running
- footsteps might not raise guard alertness if you walk along a well-known patrol route (they'd just assume it's another guard or a servant walking there)

These are fairly easy to implement in Thief if someone wanted to. Every area is already wrapped in a sound bubble (called a roombrush), and as the player enters a specific roombrush you can add a property to the AI in a mission (all or some of them) to reduce their hearing. When you leave the roombrush, their hearing is put back to normal. You could do the same with eyesight as well, aggression factor, etc. You could even rig up a sense of smell system if you wanted the burricks, apebeasts or other creatures to react if they smell you.

I'd like to see these changes, as well as the others you had, as then you can add various puzzles where machinery has to be turned on or off in order to creep through an area, or a situation where the player has to mimic a guard's movement and walk along certain routes at certain times, while also sneaking elsewhere in the few free minutes they have (almost like a KQ3 type scenario).

Importing some of the Commandos behaviors into Thief would be quite outstanding actually, such as when a guard reaches the end of a patrol route, they expect to see another guard also reach the end of theirs, and if not, then they sound the alarm, get defensive, turn more lights on or call the off-duty guards to come and search and get into position. So if you knock a guard out, you need to first observe the patrol routes of other guards that might notice the missing guard. Then even these tricky situations could be resolved with an item Garrett carries which can be thrown down to 'mirror' a guard's movement or something. This could be what the Scouting Orb could do, it mimics a guard's patrol route with an 'illusionary' guard, but only for so long and you only have so many uses.

 

[DraQ]

Importing some of the Commandos behaviors into Thief would be quite outstanding actually, such as when a guard reaches the end of a patrol route, they expect to see another guard also reach the end of theirs, and if not, then they sound the alarm, get defensive, turn more lights on or call the off-duty guards to come and search and get into position.

Shouldn't be that hard to implement - either, if the patrols are highly synchronized, have guard reaching defined meeting point wait for a short time (to accommodate irregularities, pathfinding, etc.), and then go into alerted state if the other guard isn't met; or set up much longer timer that's ticking down all the time and gets reset when meeting that other guard (separate timers for individual guards met), if it ticks to 0, alert.

 

[belowmecoldhands]

Creating an RPG is time consuming.
Creating a good combat system is time consuming.
Creating a stealth sub-game for a subsection of characters is time consuming.

Stealth is often represented as an in-game cheat option because most studios would feel like they had won the lottery if given appropriate time and money to build the first two.

I agree with this. Just having a compelling and deep system of exploration and combat is demanding. And yet I think some stealth can be a part of making a deep system of exploration and combat! I'm not so sure it matters how deep stealth is on its own. I think how all of hte parts in a game interact is what's more important. Even a shallow stealth system can work if it's incorporated well with the rest of the game.

I think you're right in that game makers have to make compromises. I think a game can't do everything DEEPLY, but it can still do everything. You ask how? If it's going to do everything then all of hte individual parts will be lightweight and the focus will instead be on how they interact to create a fun engaging experience. If they want to focus on something particularly, like stealth, then they have to make that the central experience around which everything else revolves.

There're some games which center on Stealth, like Thief 1/2 or the latest Thief (2014). In these games the whole point of the game is to use Stealth, although some felt the first two games had deeper stealth mechanics. It's oftne better to avoid a fight than ti's to fight your opponents, although this might be more true in the first couple Thief games. (I played some Thief 2)

Ok, if games can successfully employ shallow stealth systems then how does Stealth sometimes fail? I made some posts in the Morrowind<->Skyrim thread about how Stealth failed in my experience playing it. While Morrowind did have lots of work put into Stealth, my feeling is/was it wasn't tuned appropriately. It failed too often and so instead of flowing into what I was doing it was often rendered useless. It's like the sum of my stealth experience was just pressing a stealth key over and over and over until it was high enough to be useful. Keep in mind it failed repeatedly even as I was behind enemies or dressed in light armour or had a comparatively high Stealth versus my other skills. I came to a realization I did not enjoy pressing the stealth key now and then to raise it very small amounts when I wasn't getting much out of it. That's when I started grinding it near NPC guards. To make a long story short, even at high ranks it wasn't very effective. The game must have been tuned in such a way that you needed to combine it with spells or potions? Was there a trick to it? Regardless, I never was happy with hte results.

 

[belowmecoldhands]

Creating an RPG is time consuming.
Creating a good combat system is time consuming.
Creating a stealth sub-game for a subsection of characters is time consuming.

Stealth is often represented as an in-game cheat option because most studios would feel like they had won the lottery if given appropriate time and money to build the first two.

That would be a point if existing combat systems didn't look like designed from ground up by a blind simian (or more likely a committee - it's hard to tell).
Solid base design gives you a batter result and less work trying to work around issues that stem from inadequate design - yeah, I know - months of coding can sometimes save you even hours of thinking and planning.
(...)

And yet making those first two things so they don't look like they were designed by a blind simian requires extra time and know-how. I don't know how you can not understand his point. While I agreed to some extent, I added the caveat (without explicitly saying it) if the first two are lightweight then a third is easy to afford. How htey come together is what's important.

The point - and I know you do get it somehow - is game makers can't do everytihng. They only have X resources to spend on creating the game. This is true for everyone. Some have more aptitude than others, true, but even the most able must similarly obey the universe and hold themselves accountable for what they're doing with their time.

What I see is some RPGs add these extra elements/skills - like Stealth - and don't always do a good job. I think many RPGs are guilty. Should they add extra elements if they're not going to do a good job? It's debatable. If the game is moddable then it might be nice having some undeveloped portions which the modders can finish, preferably if it's disabled in vanilla.

 

[DraQ]

And yet making those first two things so they don't look like they were designed by a blind simian requires extra time and know-how.

Actually, careful examination of modern titles indicate that it's a question of demand.

Average gamer can't be said to be as dumb as a brick only because such statement would be insulting to masonry.

This basically means that trying to introduce any demanding system such as good AI or stealth that wouldn't be based on exploiting dumb vulnerabilities is a bad move when you're trying to sell your game.

Take DX:HR with its patrols actively avoiding peeking behind obvious cover, Dishonored where the AI never looks up, or Skyrim with actually decent AI that has been set to forget it was searching for you before it got halfway there.
Those are purposeful design choices to make games playable by utter morons. Those same devs could probably make much better stealth expending the same amount of time and effort if not less.

 

[Dreaad]

The point - and I know you do get it somehow - is game makers can't do everytihng. They only have X resources to spend on creating the game. This is true for everyone. Some have more aptitude than others, true, but even the most able must similarly obey the universe and hold themselves accountable for what they're doing with their time.

Lamest and oldest excuse ever. It just doesn't hold up on any level. There were games created 20+ years ago with more complicated coding languages, shittier tech resources and far less people and money than what is poured endlessly into high profile developers today. The problem is a lack of incentive for your average employee on these things, who couldn't care less about anything since management changes leaders and overall direction monthly based on the current hip trends.
It's also why you see indie developers produce consistently better mechanics, because they care about what they are making, it isn't just a pay check.

 

[belowmecoldhands]

The point - and I know you do get it somehow - is game makers can't do everytihng. They only have X resources to spend on creating the game. This is true for everyone. Some have more aptitude than others, true, but even the most able must similarly obey the universe and hold themselves accountable for what they're doing with their time.

Lamest and oldest excuse ever. It just doesn't hold up on any level. There were games created 20+ years ago with more complicated coding languages, shittier tech resources and far less people and money than what is poured endlessly into high profile developers today. The problem is a lack of incentive for your average employee on these things, who couldn't care less about anything since management changes leaders and overall direction monthly based on the current hip trends.
It's also why you see indie developers produce consistently better mechanics, because they care about what they are making, it isn't just a pay check.

I disagree. I guess we must part ways. Only reason you think indie developers produce better mechanics is because you like their games. That's your own opinion. The designers and programmers for the big companies have just as many concerns when they do their work. In fact, they might have more. How is that? Because not only do they have to have a convincing, engaging game for the most prolific gamers, but they must also make it playable by the least attentive and least consistent gamers. Making a game for EVERYONE is just as hard - if not harder - as making a game for the few.

I guess I just have to bow out of the discussion. All my point was - and I thought it was simple - is designers and programmers have X resources to spend on a project and have to set limits on what they do.

 

[belowmecoldhands]

And yet making those first two things so they don't look like they were designed by a blind simian requires extra time and know-how.

Actually, careful examination of modern titles indicate that it's a question of demand.

Average gamer can't be said to be as dumb as a brick only because such statement would be insulting to masonry.

This basically means that trying to introduce any demanding system such as good AI or stealth that wouldn't be based on exploiting dumb vulnerabilities is a bad move when you're trying to sell your game.

Take DX:HR with its patrols actively avoiding peeking behind obvious cover, Dishonored where the AI never looks up, or Skyrim with actually decent AI that has been set to forget it was searching for you before it got halfway there.

Those are purposeful design choices to make games playable by utter morons. Those same devs could probably make much better stealth expending the same amount of time and effort if not less.

That's not lack of effort, that's a game design choice. Some players will be thankful. Not all of them care whether the NPCs will look up at them just as long as it's fun for them. And you don't think those designers and programmers are told to work on other things? You think they're busy smoking in the lounge room as they're told not to work further on the stealth system? They're being told to do other things. They're not paid to do nothing.

Furthermore, enabling some of these features you mention also requires additional work so it's playable. You say they could instantly enable these things without any additional labor but I say BS. All of that has to be added on top of everything else in the game, so the point still stands: only so much can be featured in the game.

I recall reading the EQ developers experimented at first with healer NPCs whom actively healed the NPC fighters. Their aim was to have more intelligent NPCs. They produce all variety of other behaviors to compliment this - like NPCs whom don't fight you without calling for help. After some time of doing these tests, they just couldn't come to a point where it became playable. So instead of pursuing the issue further - in the interest of finishing the game - they simplified the AI and focused on other areas. The end result is some NPCs did heal other NPCs and some did "call for help" in a rough manner of speaking, but it was toned down.

I also recall reading the UO designers and programmers experimented with an actual ecology. Predators would hunt prey and vegetation would grow/regrow to attract prey. They'd planned on them reproducing too. Raph, one of those who worked on it, stated it was abandoned at a certain point phase in development. He regretted it because he apparently thought it had potential. Now, just because they stopped developing this feature and/or abandoned it doesn't mean the rest of hte game is terrible or that they had permission to binge drink. Instead they took what they had and they focused on other things.

You can find a reference ot that here:
http://www.raphkoster.com/2006/06/03/uos-resource-system/
http://www.raphkoster.com/2006/06/04/uos-resource-system-part-2/
http://www.raphkoster.com/2006/06/05/uos-resource-system-part-3/

How develoepd a feature is doenstn' matter because it's all in context with the game. If the feature, light or heavy, interacts well with the rest of the game to produce a fun game experience then ti's a success.

 

[Dreaad]

I guess I just have to bow out of the discussion. All my point was - and I thought it was simple - is designers and programmers have X resources to spend on a project and have to set limits on what they do.

And my point is that you don't understand the industry. There is almost infinite resources available for large publishers, the problem is not with the resources, if a small polish company with 20 or so people can make the Witcher 1 why is it Ubisoft can't make stealth an interesting element in a game where you play as a fucking assassin? The resources are squandered on poor scripted coding, voice overs and pushing graphics to the absolute retarded limit every time (with almost no noticeable effect except system slowdown for the user).

You are right in saying that this happens because of pandering to the masses, which for the most part are dumb as fucking rocks and have the attention span of ADD 8 year old children.

 

[Ninjerk]

I guess I just have to bow out of the discussion. All my point was - and I thought it was simple - is designers and programmers have X resources to spend on a project and have to set limits on what they do.

And my point is that you don't understand the industry. There is almost infinite resources available for large publishers, the problem is not with the resources, if a small polish company with 20 or so people can make the Witcher 1 why is it Ubisoft can't make stealth an interesting element in a game where you play as a fucking assassin? The resources are squandered on poor scripted coding, voice overs and pushing graphics to the absolute retarded limit every time (with almost no noticeable effect except system slowdown for the user).

You are right in saying that this happens because of pandering to the masses, which for the most part are dumb as fucking rocks and have the attention span of ADD 8 year old children.

Probably also too many cooks. I'd imagine in companies and developoment studios that big you have a lot of man-hours wasted on needless bureaucratic tasks and tampering from the management.