Jump to content
The Dark Mod Forums

Chloroform And Blackjacks


bob_arctor

Recommended Posts

  • Replies 120
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

That was just one of the problems. Pathfinding was another (where do they go upon waking up? how do they get back to their route if they've been carried away or dropped down a ladder?). There were also a lot of possible exploits--what does an AI do if it wakes up after being thrown behind some heavy barrels or other barrier? Just stand there? What happens if the AI wakes up but his weapon has been thrown somewhere

else? etc... It's like ladder climbing. Sounds good in principle, but it's not so easy when you start thinking of the details.

 

After considering all the factors, we decided we probably couldn't do it without damaging immersion more than just having guards stay unconscious.

Link to comment
Share on other sites

BTW, what were the exploits about ladder climbing, apart from waiting until the AI's helpless then filling him full of arrows?

 

I agree that waky up unconscious dudes would be a great big can o' worms, but I might enjoy leading an AI up a ladder, then somehow making him fall off... On a related subject, how about guard mantling?

--

Somethin' fishy's goin' on here... Come on out, you taffer!

 

~The Fishy Taffer

Link to comment
Share on other sites

The only ladder exploit is the one you mention. I don't see it as an exploit.

If a player wants to cheat and lure AI to ladders in order to kill them, them that's fine by me. IF they want to ruin ther own game with that sort of crap, good luck to them.

I don't see how that should stop the rest of us having the extra thrill of knowing that we can't escape by simply climbing a ladder, like in all of the Thief games.

You could do the same sort of exploits in previosu thief games, like lure AI into water and drown them, or under elevators and crush them etc

Like I say, if some dumbass players, the sort who obviously like killing the AI, want to play the game like that, I don't give a fuck, but I don't want features not to be implemented on the basis that people like that will exploit them.

I'd like to see them mantling as well, but thats a lot more technically tricky to do.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

There were a few different exploits, but those weren't the only considerations. There would be a tremendous amount of work involved. I know we've discussed this already in the public forums, but to briefly summarize:

 

There are lots of pathfinding exploits. AI is halfway down ladder, thief blocks ladder at bottom. What does AI do? Or the AI is just at the top, but the thief stands in the way. AI is still on the ladder so cannot draw a weapon, so what do they do?

 

AI catches thief while both are on ladder. What does AI do? Combat on ladders would require an entirely different set of coding and animations (two sets, actually one for going up and one for down).

 

There's the kill/ko guard while on ladder exploit, as you mentioned, which can be used against multiple guards.

 

Many types of ladders require the player to turn around and jump off at the top to reach the next level. How does the AI handle this?

 

Anyway, that's several of the problems. Solving them seemed to require either a large amount of work or an immersion-breaking 'fix'.

 

As for mantling, I don't know how difficult it would be to program.

Link to comment
Share on other sites

All of those things amount to saying 'We're not going to incliude this much wanted feature becasue a minority or players might exploit it'

Why shoud we care if they exploit it?

This isn't an MMO, we don't have to be airtight from exploits. If players what to cheat themselves (becasue that's all they'd be doing) then let them.

Someone wants to lure all th AI on the map to a ladder and kill them?

Let him.

I prefer to have the AI climbing ladders in my game, rather than not include ladder climbing at all, just so that person can't exploit it.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

Regarding bodies & pathfinding waking up, the whole point would be the fun of hiding the bodies so the AI wakes up and is stuck. Heh heh heh, can't tell your friends, you're in the well! But yeah realistically I get you, would be more immersion breaking in than out.

 

TDS simple ladders (I quite liked them,better than falling to your death like in TII) would be simpler for the AI. But overall I see the problems there. Maybe some wall climbing enemies? Spiders/undead etc? Or flying?

If you put in lifts that could aleviate the ladder problem. I mean ladders only work for the player because they are the player, so cheats like walking standing up and shooting on them don't matter, can't see them, and don't look too bad in MP games (HL1 based games for example). But SP demands good climbing, and realistically climbing a ladder up to a thief is dumb. You would be knocked off probably. Lots of archers and lifts and alternative ways would be good, so few places were solely ladder accessible.

Link to comment
Share on other sites

Odd, it's not just a matter of exploits. It's immersion problems. The only problem with guards not climbing ladders is that it reminds you that you're playing a game. Well, if the guards don't act in a believable manner while on a ladder, you've just done load of work and have the same problem as a result.

 

If players what to cheat themselves (becasue that's all they'd be doing) then let them.

 

Aren't you usually on the other side of this argument? :)

 

Besides, there's no reason guards can't use elevators. I think we'd also talked about a possible compromise, where guards could only climb ladders when the player wasn't directly nearby, but I don't know whether that went anywhere.

Link to comment
Share on other sites

I would like that one, but... Well it would certainly catch the player by surprise the first time it happened, but players would soon work out what was happening. (Like the shell not hurting enemies off screen in Mario)

We would end up with players deliberately hanging around and trying to pick the guards off, rather than keep running and face the result of being chased sucessfully.

Link to comment
Share on other sites

Aren't you usually on the other side of this argument?

No, I'm always on the side of the argument which advocates ignoring players who kill and cheat the AI in stealth games, as well as leaving them completely out of the loop and not even considering them as human, but some kind of hairy palmed neanderthals who consider both the xbox and the FPS to be appropriate forms of entertainment.

We should never make a decision based on what the worst players might get up to, and should only consider real stealth fans.

In short, we should make an elitist game and be proud of it.

I'm sure we can get AI behaving realistically on ladders. Yes, they'll be open to some exploit for sure, but that alone is not enough reason to forget the whole idea.

Will the real stealth fans lure AI to ladders and kill them?

No, they won't.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

But making a game with known exploits is just ... a sin. It defeats the whole purpose of a video game.

In a real life board game or sport, you have to keep in mind the rules and not break them. I feel that a lot of the fun of a video game is that you don't need to do this - it's a virtual world where you are free to acheive things any way you can. Realisng that guards are helpless on ladders would take me out of the immersion, and I coudn't help feeling "I could take you out, but I won't cause its not fair" that's the attitude I would rather keep for playing against kids.

 

There's a difference between this and what I'm talking about in the violence thread. In the violence thread, I'm talking about making a moral desicion on something that can seem plausable in real life. It's immersion breaking to know of an exploit and have to avoid it. I never really had to avoid serious exploits in T3 or T2 cause most of the things people said were exploits never really worked for me. Eg. there is rarely any slope leading to water to lure AI into - the only one I could find was in Thief 1 in the first mission and I gladly lured those stupid evil zombies in when they chased me that far out.

 

Actually I remember some exploits that did work - llike waiting for them to search close to you in the darkness and just blackjack them in teh face, in T2. Or potentially being able to round evryone up, drop a flashbobm, then blackjack them one by one. It took a lot of effort to forget these tthings, and I was never happy about having to do that. It was immersion breaking for me.

 

For me, avoiding exploits sucked in T2 and it would suck in darkmod too.

Link to comment
Share on other sites

We're not deliberately creating the game with exploits, we'd be introducing a much wanted feature for the benefit of most decent players, which may be open to exploit by some seedy players.

That's a fair trade in my book.

If we're going to have a 'if it's not 100% perfect than don't include it rule' the game will never see completion.

I think we could get AI ladder climbing to a decent state, even if it's not completely airtight. It has been done in other games.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

Ultimately I suspect it comes down to how feasible it is to program and rig the AI models to climb, a compromise would be to have a custom flag set in the ladder texture property and the AI script - that way savy level designers could path and mark apropriate ladders. It'd keep the players guessing because they wouldn't know if an AI would chase them up or not (at least until they had played a level a couple of times) and would also hopefully allow level designers to tailor the situations to avoid brokeness.

Link to comment
Share on other sites

OK, We know oDD mixes up "we should" with "I want to" so disregarding the fact that anyone who blackjacks guards is a hairy palmed neanderthal...

 

I think the other exploits are perfectly valid reasons. Obstructions are easy to deal with IRL and difficult in game for the AI... The problem is, as Dom said - the exploits are easy to accomplish and not using them is a choice, not forced. I can't see how you can claim to always be on this side of the argument when you want all weapons to be removed from the game to prevent the sinners from sullying it with death.

 

"We're not deliberately creating the game with exploits"

 

If you release a game, knowing it has exploits, then you deliberately released it with them. You may not want them there, but that is completely irrelevant. The thing is, you could just build a Thief game where you have no guards, and just trust people to sneak around. It's pointless and, really, quite lazy. That's why it's bad to release it - and now I've heard the other exploits, would be worse than without ladder climbing. Unless you can work out decent vertical pathfinding, fixing those exploits, then it would plainly suck to "imagine" that you weren't able to just put a bottle at the top of the ladder and have them get stuck or force them to return.

--

Somethin' fishy's goin' on here... Come on out, you taffer!

 

~The Fishy Taffer

Link to comment
Share on other sites

This is where we left off in our internal discussion, some time ago.

 

REASONS TO IMPLEMENT LADDER CLIMBING BY AI

 

1. It forces players to suspend their disbelief when guards do not climb ladders (ie, realism).

 

2. Ladders can be exploited as an easy getaway, since the guards cannot follow.

 

3. Ladders can be exploited as a 'safe zone' since guards cannot follow (solved by missile AI, so this is no longer an issue).

 

These are the only reasons I can see for making AI climb ladders, to increase immersion and limit exploitation. But if we implement them in a way that decreases immersion and creates new things for players to exploit, then we haven't really gained anything.

 

These are the issues that need to be resolved (without creating new exploits or immersion issues)

 

 

ISSUES TO RESOLVE

 

1. Attacking guards on ladders exploitation. While the AI is on a ladder, he is virtually helpless. He cannot dodge or parry. A player can exploit this by shooting arrows at an AI while it is climbing, or by getting a free hit or two with a sword or blackjack while the AI is getting on/off the ladder.

 

2. Blocking the ladder exploitation. The AI cannot pass a character on a ladder. If the player stops, what does the AI do? Attacking on ladders creates a whole new subset of combat rules and animations, creating their own issues. It is also possible for a player to block a guard who is trying to get off the top or bottom of a ladder by standing in front of it. The guard is still climbing, so if it cannot attack, what does it do? The player, no longer on the ladder, is of course free to attack.

 

3. What about ladders that do not have neat on/off points? I am imaging ladders like those near the Horn of Quintus, or ladders that end several feet above the ground, or ladders that face the wall, or ladders where you can get on and off in the middle. There are many types of ladders against the wall, where the player has to climb to the top, and then basically turn around and jump to get to the next floor. Each case like this would need a separate animation from the standard 'get on/off ladder'. I don't know how much separate Ai coding would be necessary to keep guards from falling OFF the ladder when they tried to do this.

 

 

These are three of the major issues I see revolving around ladder climbing. In my view, any solution that strains immersion (like guards that can't be hurt on ladders, or go incorporeal, or suddenly get way faster than the player without cause) means we are not gaining anything. And we shouldn't do something like this (that will take significant work) if we aren't really making a big improvement to the game.

 

So, can we solve these issues without resorting to solutions that strain immersion or create new exploits?

 

I've been wracking my brain and can't come up with anything, other than perhaps doing what they did for the wall-climbing imps in D3...allow ladder climbing AI to be scripted, for instances where the Thief can see them but is too far away to interact with them. Essentially creating the illusion that they *can* climb ladders, but just choose not to do so most of the time. Of course, this only addresses Reason #1, not #2. It's obviously far from a perfect solution.

Link to comment
Share on other sites

Obviously a lot of those niggles would be down to the FM author not placing ladders properly. We could build prefab ladders with limits already in place as to how close to the ground/wall they should be etc to help alleviate some of those.

Shooting a guard while he's on a ladder is not an exploit. THey can equally shoot at you while you're on it. Same goes for blocking. They can block you. Two guards could potentially block the top and the bottom and leave you stuck.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

Obviously a lot of those niggles would be down to the FM author not placing ladders properly. We could build prefab ladders with limits already in place as to how close to the ground/wall they should be etc to help alleviate some of those.

 

True, that would help some of #3, but there are a lot of valid ladder types that authors wouldn't be able to use (or perhaps AI could climb 'proper' ladders only?)

 

Shooting a guard while he's on a ladder is not an exploit. THey can equally shoot at you while you're on it. Same goes for blocking. They can block you. Two guards could potentially block the top and the bottom and leave you stuck.

 

But the player has options while on a ladder. They can jump off, for starters. They can also use inventory items if I'm not mistaken. It's also unlikely the guards could intelligently block the player, while the reverse is not true.

 

Anything guards do on a ladder requires seperate coding and separate animations. The above doesn't even take programming issues into account. Look at how much work Ish had to do to get AI to notice when they collided with the player. How many problems like that are we likely to run into, given that current D3 coding is probably optimized for x-y axis interaction?

 

I'm not saying it's impossible, but would be a LOT of work. Probably more than designing the entire combat system.

Link to comment
Share on other sites

I guess if you could detect a blockage ahead and jump off, or a blockage below, and climb back up, then it would be better... Still, nonstandard ladders would be a problem if the guard lept to his death. Actually, he could slide down, I suppose (yes! like Wallace!) but anyway... For one, I expect a decision's a decision, and it's all down to whether it's feasible to get all this stuff in place. Most of it sounds possible but... :/

--

Somethin' fishy's goin' on here... Come on out, you taffer!

 

~The Fishy Taffer

Link to comment
Share on other sites

I really don't know what the fuss is with AI climbing ladders. If a guard gets caught out on a ladder in RL, they have naturally left themselves exposed. So has the thief climbing the same ladder. It is realistic and natural to exploit somones vulnerability for your own advantage, so to call it an exploit as though it is such a bad thing per se is just silly. What would you do if someone was climbing up a ladder after you? You might decide to lurk at the top and push them off when they get to the top, or you might keep on running, but if you push them off and they fall to their death, you no longer have to worry about them running after you (if it is a lone guard). Conversely, if you were trying to pursue a thief and he nips up a ladder, you will either make the decision to follow him and try to grab his leg and pull him off or not based on your judgement as to whether they will be able to take advantage of you while you are exposed on the ladder.

 

The ONLY question regarding AI climbing ladders is whether or not you can give the AI a reasonably realistic way of making the judgement call as to whether they should follow you up or not..

 

I have seen very decent AI botpathing with ladders and elevators in a few MP deathmatch games, and Splinter Cell does ladder climbing AI quite nicely, so I know it is possible to code, but if it is too difficult or too much work to code WELL then I can understand it not being included.

 

Personally, I would rather have an imperfect and slightly buggy implementation of ladder climbing than none at all, as long as it works properly most (80%) of the time.

 

Are you planning to make ladders just by having a texture flagged as a ladder? If so that will create more problems for your ladder. A better way to do it is to create a ladder volume, with an invisible entity at the top and bottom with a directional property, and to define ladder exit/entry points on the geometry where the player AI gets off - this is how most Unreal engine games have done it, and making ladders work in Unreal is quite easy by comparison with Quake based engines...

Edited by obscurus
Link to comment
Share on other sites

obscurus, you've missed the crux of the matter - there are more exploits than guards being vulnerable while climbing ladders. What does a guard do if you are sat at the bottom of the ladder and he's climbing down? Or at the top when he's going up? He can't attack you, and he can't just stop or you'll smack him. You have to program lots more AI rules to cope with these, which will be much effort for just this one effect. There are also problems with obstacles, and remember that if you're having non-standard ladders, such as those requiring you to jump on and off, you need to do animations and AI rules to cope with all these various situations.

 

It's not pathfinding that's the problem.

--

Somethin' fishy's goin' on here... Come on out, you taffer!

 

~The Fishy Taffer

Link to comment
Share on other sites

As long as it's equal for AI and player there's no probelm.

The guards can attack the player while he's helpless on a ladder, and vice versa.

The guards can block the player from getting on/off a ladder and vice versa.

THE guards shoiuld be able to use their swords while on ladders. It's not a problem, it only requires 2 animations - one where the guards is attacking above him, and one where he's attacking below him. THat's only three total animations for the ladder.

We can make prefabricated ladders from level designers to use which are the correct width and state that they have to be a certain didtance from the wall etc to avoid bugs.

It's not as if ladders are extremely common anyway. THey are actually overused in game levels.

I'd be more interested in AI being able to mantle after the player, which of course, has it's own set of similar problems.

A better and easier way to solve the probem of the player abusing legdes and high places where the AI can't go, is to give all of the AI access to ranged weapons.

THere's nothing more ridiculous in Thief than when you can escape a group of crazed sword wielding guards by climbing into a 3 foot ledge.

Civillisation will not attain perfection until the last stone, from the last church, falls on the last priest.

- Emil Zola

 

character models site

Link to comment
Share on other sites

As long as it's equal for AI and player there's no probelm.

 

That's the point - it wouldn't be.

 

Attacking while on a ladder: Are you suggesting that the AI team produce rules for tricking the player into climbing ladders, lying in wait then ambushing them while helpless? Because that's what would be necessary for an actual equal footing.

 

Blocking: again, the guards have to be programmed to do this, and the player can jump off, or climb back up. This is useless for guards if the ladder is high off the ground, but the player has still blocked them somehow, as they'd jump to their deaths. Climbing up is useless if the guard had to jump onto the ladder, as a player would have to in certain situations. Also, if you somehow managed to allow the guards to pick up a large barrel and position it beneath the player, they'd know whether to jump off, to carry on, to pick it up and move it, or whatever. Programming AI contingencies for ALL of these situations, and probably more is nigh on impossible.

 

It's not a problem, it only requires 2 animations - one where the guards is attacking above him, and one where he's attacking below him. THat's only three total animations for the ladder.

 

What about where the player is behind him, or in front at the top? Guards fighting on the level have several different animations for sword fighting, enabling them to do complex things such as parry and so on.

 

We can make prefabricated ladders from level designers to use which are the correct width and state that they have to be a certain didtance from the wall etc to avoid bugs.

 

So you're proposing to have climbable ladders, but they only work if it's a ladder reaching from the floor (or a distance that you needn't jump up to) and then you go over the top when you reach the next level? What about those where you need to jump on from the top, and jump off at the top? What about ladders that allow you to traverse multiple levels? That's a great restriction of level designer's freedom; barely worth it in my opinion. Often ladders are positioned in nonstandard ways. Perhaps ladders are overused, but in some situations they are necessary. In tight spaces such as watch towers they're appropriate, in attics they're useful, anywhere that's run down would benefit from the atmosphere they lend to the place. While using only standard ladders would solve some of the AI problems, it wouldn't solve them all, and would be less than desirable.

 

I'd be interested in mantling capabilities, but you have to give guards some element of common sense in order for them to judge whether to attempt a jump. If there are obstacles, or the player, or something else in the way, then it would futz their attempts, possibly leading to other exploits (e.g. place a crate on a ledge, jump across with an AI chasing you and when they attempt the jump, they fall off because of the crate) I think it's probably just as difficult, although the AI chaps might know better.

 

A better and easier way to solve the probem of the player abusing legdes and high places where the AI can't go, is to give all of the AI access to ranged weapons.

 

But totally unrealistic. It would ruin the variety, and part of tactics is knowing that _some_ places you should be able to access without being followed (e.g. via rope arrows) and ranged weapons for everyone would defeat this.

 

THere's nothing more ridiculous in Thief than when you can escape a group of crazed sword wielding guards by climbing into a 3 foot ledge.

 

You are a sword wielding guard, and an intruder has managed to climb onto a small wall. What do you do? If you climb up, he can simply kick you off the wall, or club you while you're vulnerable. You can't reach him from the ground - are you supposed to magically defeat him? In some situations, yes, guards not mantling is ridiculous, but it's less so than you imply.

--

Somethin' fishy's goin' on here... Come on out, you taffer!

 

~The Fishy Taffer

Link to comment
Share on other sites

If a guard gets caught out on a ladder in RL, they have naturally left themselves exposed. So has the thief climbing the same ladder. It is realistic and natural to exploit somones vulnerability for your own advantage, so to call it an exploit as though it is such a bad thing per se is just silly.

 

A real life guard on a ladder has many choices. If they are attacked from below they might hang by one hand and draw a weapon. They might jump off the ladder. They might kick their pursuer in the head or step on his hands. They might swing around to the other side of the ladder, or slide down the rails quickly. Can you imagine the amount of work to code and animate these options for AI?

 

As long as it's equal for AI and player there's no probelm.

The guards can attack the player while he's helpless on a ladder, and vice versa.

The guards can block the player from getting on/off a ladder and vice versa.

 

It's not equal because AI will never be anywhere near as smart as the player. The player can lure several AI to a ladder, block the top and then attack the guards one at a time as they try to get off. AI are not going to be able to do the same. AI are not going to push a barrel in front of the ladder while the player is on it. Etc.

 

Personally, I would rather have an imperfect and slightly buggy implementation of ladder climbing than none at all, as long as it works properly most (80%) of the time.

 

Obviously nothing is stopping people from trying to implement ladder climbing once the mod is released, but I wouldn't be happy working for half a year or more to release a feature that doesn't work properly.

 

I'd be more interested in AI being able to mantle after the player, which of course, has it's own set of similar problems.

 

Actually the problems with mantling are more about pathfinding. The mantling process is only a few seconds long, so exploiting it for attacks is much less of a problem. Blocking it is also not as much of an issue, since the guard is only a second or two away from the bottom, unlike ladders. Getting the animations right might be an issue though, since surfaces would be at irregular heights (although they did it in TDS presumably...).

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recent Status Updates

    • Ansome

      Finally got my PC back from the shop after my SSD got corrupted a week ago and damaged my motherboard. Scary stuff, but thank goodness it happened right after two months of FM development instead of wiping all my work before I could release it. New SSD, repaired Motherboard and BIOS, and we're ready to start working on my second FM with some added version control in the cloud just to be safe!
      · 1 reply
    • Petike the Taffer  »  DeTeEff

      I've updated the articles for your FMs and your author category at the wiki. Your newer nickname (DeTeEff) now comes first, and the one in parentheses is your older nickname (Fieldmedic). Just to avoid confusing people who played your FMs years ago and remember your older nickname. I've added a wiki article for your latest FM, Who Watches the Watcher?, as part of my current updating efforts. Unless I overlooked something, you have five different FMs so far.
      · 0 replies
    • Petike the Taffer

      I've finally managed to log in to The Dark Mod Wiki. I'm back in the saddle and before the holidays start in full, I'll be adding a few new FM articles and doing other updates. Written in Stone is already done.
      · 4 replies
    • nbohr1more

      TDM 15th Anniversary Contest is now active! Please declare your participation: https://forums.thedarkmod.com/index.php?/topic/22413-the-dark-mod-15th-anniversary-contest-entry-thread/
       
      · 0 replies
    • JackFarmer

      @TheUnbeholden
      You cannot receive PMs. Could you please be so kind and check your mailbox if it is full (or maybe you switched off the function)?
      · 1 reply
×
×
  • Create New...