How can we discourage botting?

[Adposatnr]

Hello, its been a while since ive been in this thread but i'd like to say something(that very likely goes against what i might have said in this thread in the past).

Botting is like the drug war, especially in the US(a few years back before legalizing weed in some states).
Instead of fighting it and spending countless of resources and time to make it go away, make it a business of some kind.

How could it be done to make it so that bots are a healthy part of society?
lets think outside the box for a second.

-Maybe give all players one bot for free that works for your server, but make it so that you can only bot 1 hour a day per character, account, or per IP.
-Make it so that bots give benefits to those that dont bot like the current online population is baked in to how much bonus exp monsters give.
-Players that bots can be honest about it and not be shamed of doing it, but they agree to that botting will give the character some negative buffs like lower exp/h, give them lower cap so more time is spent running back and forth with loot making the world more alive.
-You can sell "bot tickets" in your shop, and those that bot without these gets their accounts deleted/suspended.

What about people using custom made bots without any debuffs? It will be even worse than current situation

 

[Flameria]

You cannot win against bot, there will always be people who are interested to break what you create so in that regard make one official client where people can use bot and one which they cannot if you are trying to make a higher client verison of server.

I've tried to make some mechanisms myself with banning if any dll or unknown packages are sent but the bot developers are quick to pick this up and make the bot send falsepositive packages the server accepts.

It's a very hard fight but I recon the best thing is that you yourself as a GM shows activity in lookin people up & actually deleting their account if anything was suspicious, in that regard you will create fear which will most likely remove 80% of the "botters" but hardcores will always give it a try.

 

[Silba]

I seen quite a few good ideas here, and they inspired me. They way I seen it, there was four main points to minimize or reduce botting. Each idea on its own merit could help with lowering the amount of botters, but definitely not eliminate it all together. To eliminate botting shouldn't be the task, but rather minimizing the botting to the point that the staff of the server has a good chance to weed out the ones who continue to attempt botting anyways. So I decided to make a mind map ! That mind map is quite unfinished , however I had more than one good idea emerge from that process, so I decided to share here the four points that were great ideas, and the mind map itself incase anyone else wanted to continue it or do the same thing themselves.

To be clear this mind map focused towards the cavebot and runebots, not so much for magebot.

The four main points were

1. Make advancement more fun / engaging.
2. Make botting more difficult to do
3. Reward Non-Botters
4. Create software or script systems to monitor botting, things that could indicate botting

Anyways here is the unfinished mind map, there are many many nodes that could easily be extended with more ideas, and any / all ideas added to this mind map would just add to the overall diminishing of bots on ones server. If anyone wants a link to have access to this mind map, just message me directly and I'll share the link, but this image along with the link above is all anyone needs to get started with expanding and coming up with their own ideas

That's really cool, thanks for that.

Hello, its been a while since ive been in this thread but i'd like to say something(that very likely goes against what i might have said in this thread in the past).

Botting is like the drug war, especially in the US(a few years back before legalizing weed in some states).
Instead of fighting it and spending countless of resources and time to make it go away, make it a business of some kind.

How could it be done to make it so that bots are a healthy part of society?
lets think outside the box for a second.

-Maybe give all players one bot for free that works for your server, but make it so that you can only bot 1 hour a day per character, account, or per IP.
-Make it so that bots give benefits to those that dont bot like the current online population is baked in to how much bonus exp monsters give.
-Players that bots can be honest about it and not be shamed of doing it, but they agree to that botting will give the character some negative buffs like lower exp/h, give them lower cap so more time is spent running back and forth with loot making the world more alive.
-You can sell "bot tickets" in your shop, and those that bot without these gets their accounts deleted/suspended.

This is the direction i am going on my current project, i will allow botting and the client even ships with it (otcv8). However everything is being designed to not limit botting at all. Currently using a bot has little to no advantage except for cavebotting. However i expect cavebotters to be pked on release due to the pvp incentives i will be implementing.

that may be good for servers that are post 7.72 version, but allowing everyone to bot on those 7.1-7.72 versions would be crazy, no different than playing 8.0/8.6, especially in pvp

when cip introduced hotkeys they made it so they can never miss a player, it is up to the server owners to decide if they want that or not on 7.x. My current project hotkey aiming and bot aiming is not perfect and can miss. It's 7.6. As for autohealing, that's just meh. A good pvper is as fast as any bot assuming a good ping.

What about people using custom made bots without any debuffs? It will be even worse than current situation

Exactly.

You cannot win against bot, there will always be people who are interested to break what you create so in that regard make one official client where people can use bot and one which they cannot if you are trying to make a higher client verison of server.

I've tried to make some mechanisms myself with banning if any dll or unknown packages are sent but the bot developers are quick to pick this up and make the bot send falsepositive packages the server accepts.

It's a very hard fight but I recon the best thing is that you yourself as a GM shows activity in lookin people up & actually deleting their account if anything was suspicious, in that regard you will create fear which will most likely remove 80% of the "botters" but hardcores will always give it a try.

Winning against botters is not the goal, existing with them is. They are inevitable and unstoppable. I would rather my game not encourage botting so i don't need to worry about it and then if they do bot it won't be so destructive.

 

[Ochman]

A good pvper is as fast as any bot assuming a good ping.

I could go into details and try to explain why you are wrong on so many layers, but I do really think it's pointless, so please let me just leave those two letters:
X D

 

[Silba]

I could go into details and try to explain why you are wrong on so many layers, but I do really think it's pointless, so please let me just leave those two letters:
X D

Most brain dead comment of the year so far on otland? Well done I guess.
Context matters, grow up.

If you do decide to grow up feel free to pm me and I'll 'ELI5' my statement for you including a brief lesson on why context is important to avoid such embarrassment in the future.

X D

 

[Ochman]

Yes, please eli5 this lesson here at public, we’ll avoid confusion of someone saying that person is as fast as bot.

 

[iNux]

Allowing botters with ingame bots but giving them much fewer resources would probably be an option. I've never seen such a thing, though some games may have this kind of function?

This is what ChatGPT Suggests:


1. Interactive Anti-Bot Measures: Implement random captchas or other interactive activities that bots can't handle but are fairly easy for a human. This can be during login or at random intervals during gameplay. However, keep in mind that too frequent interactions can become bothersome for genuine players.

2. Behavioral Analysis: Bots often display characteristic patterns of behavior which distinguish them from human players. You can use machine learning algorithms to identify such patterns and flag suspicious accounts. This could include patterns like repetitive movements, abnormal play hours, no chat or interaction with other players, and quick, inhuman response times to events.

3. Reward System for Reporting Bots: Encourage players to report suspected bot activity with a well-established reward system. This incentivizes players to actively participate in maintaining the integrity of the game environment.

4. Alternative Gameplay Mechanics: Consider incorporating features that require strategy, decision-making, or dynamic player interaction, which are more challenging for bots. This could involve complex puzzles, variable quests, or random events.

5. Account Verification: Implement measures like email or phone verification for account creation. This won't stop dedicated botters but can be another barrier to entry.

6. Economy Monitoring: Since one of the major reasons for botting is economic gain, consider setting up system to monitor the in-game economy and detect anomalies which might indicate botting, such as rapid accumulation of wealth.

 

[Evil Puncker]

Allowing botters with ingame bots but giving them much fewer resources would probably be an option.

\/

What about people using custom made bots without any debuffs? It will be even worse than current situation

@topic
AI to detect possible botters+manual review and acount deletion

 

[alejandro762]

Something stupid that can be done , we see that in a server , kicking players everyone , if you dont accept a modal Windows and didnt answer correctly , its reported for 30 min and if you Forget you get kicked again , using something that you cannot Live in game while modal is opened. I dont know how that can be reproduced but seems working

 

[Silba]

AI to detect possible botters+manual review and acount deletion

I think the manual review is the important part here, but that's expensive as hell. I'm interested to see where ai takes us in this regard

Would also be funny to unleash a legion of ai controlled characters that will play and talk to others and learn who bots.

Something stupid that can be done , we see that in a server , kicking players everyone , if you dont accept a modal Windows and didnt answer correctly , its reported for 30 min and if you Forget you get kicked again , using something that you cannot Live in game while modal is opened. I dont know how that can be reproduced but seems working

This is very intrusive and it's easy to make a bot to either solve the problem or alert the human that a problem needs solving. Not acceptable or effective. It probably only works on the average botter, but the average botter does not harm the game.

 

[Night Wolf]

1. Interactive Anti-Bot Measures: Implement random captchas or other interactive activities that bots can't handle but are fairly easy for a human. This can be during login or at random intervals during gameplay. However, keep in mind that too frequent interactions can become bothersome for genuine players.

2. Behavioral Analysis: Bots often display characteristic patterns of behavior which distinguish them from human players. You can use machine learning algorithms to identify such patterns and flag suspicious accounts. This could include patterns like repetitive movements, abnormal play hours, no chat or interaction with other players, and quick, inhuman response times to events.

3. Reward System for Reporting Bots: Encourage players to report suspected bot activity with a well-established reward system. This incentivizes players to actively participate in maintaining the integrity of the game environment.

4. Alternative Gameplay Mechanics: Consider incorporating features that require strategy, decision-making, or dynamic player interaction, which are more challenging for bots. This could involve complex puzzles, variable quests, or random events.

5. Account Verification: Implement measures like email or phone verification for account creation. This won't stop dedicated botters but can be another barrier to entry.

6. Economy Monitoring: Since one of the major reasons for botting is economic gain, consider setting up system to monitor the in-game economy and detect anomalies which might indicate botting, such as rapid accumulation of wealth.

1 and 2: It's weird to see it written like this. This text seems to be taken out of a work I've done in university years ago, I'm curious about the reference hahaha.
3. Personally I don't think this would work, you'll end up taking more time to analyse and confirm reports than facilitating your life.
Useful if you have a quick way to validate all players data massively to confirm if the account is not legit, however, you might get a lot of fake or troll reports.
4 is the same thing I've discussed here.

this argument is in fact what many people inconsciently think when considering a bot.

As I told in page 2, people bot simply because the grinding part of the game isn't fun. You can close the thread, your answer is already there.
You can circle around 500x times and give other reasons but in fact this is all directly related to this.

You have a game in which vast majority of players are only looking to level up fast and go to PVP with their friends to establish dominance in the server. Make them grinding is a way to hold back their progress and make the game last longer. But in counterpart, this is all repetitive work and makes people want to skip/automatize this boring part.

How to discourage botting? Make the game fun. If you need to grind, make it random, turn it all in a discovery. Make people play together and interact. Put more exp for people playing together, invest in a task system.
Make good maps, with surprises and traps. Put monsters with random tiers, chances to spawn bosses upon XXXX monsters killed.
Make the client have a better layout, make changes in the way you play so you don't constantly need to take your hand out of the mouse and press F1-F8. There's plenty "small details" on the interface and the way we interact with the game that could be improved, but everyone simply overlook them.

If your game is just like 42368528532 other servers out there, people will simply have one goal: grow as fast as they can, establish dominance and them quit. You can't blame them for automatizing this.

5. This doesn't make any sense unless you're trying to restrict multi-accounts.

Now, about [2] and other comments like @Evil Puncker rightly said about using IA for detection. This is a gray area and I'll try to explain why.
To begin with, there's little to none literature or practical examples of this being applied in a real life scenario. When I've built the antibot project for my private server years ago I could only find 3 references and neither of them were close to what I was trying to do.
You can check the article (in PT-BR) here: antibot/Article.pdf at master · andersonfaaria/antibot (https://github.com/andersonfaaria/antibot/blob/master/Article.pdf)

You need to understand that this is not an easy task first of all, it involves identifying all types of bots including assisted ones and trainning algorithms to try and find a specific pattern to how they behave to separate them from real players. Apart from this, hundreds of millions of dollars were spent in this direction to try and make games bot-safe with the biggest examples being Valve and recently Riot.
Meaning, any person that do this work wouldn't give hints on how they did it, but instead would built a company to sell this as a service (like other bot-detection companies). This relies heavily on types of bots people use, their functions and which data you can collect to distinguish it.
Now, for my personal proof-of-concept project I have noticed most bots didn't used "noise" in the intervals they apply between functions. The interval might make the bot behave more human-like in what comes to speed (time between actions) but it would still be showing as a bot if you compare other layers of the speed, such as mean and closeness to mean (standard deviation).

Even like this, giving enough randomness could be enough to mask the actions less predictable.
One might argue that in the future AI will be able to be used in a better way to instantly spot bots and group them by functions they are utilizing in a given interval, however, when the technology to detects evolve so does the one to improve/create bots. You can expand that until you actually have 2 artificial brains fighting to see who will win (spoofer vs. detector), even so the spoofer will have always have a reward period until detector reaches a point to caught it.

 

[iNux]

1 and 2: It's weird to see it written like this. This text seems to be taken out of a work I've done in university years ago, I'm curious about the reference hahaha.
3. Personally I don't think this would work, you'll end up taking more time to analyse and confirm reports than facilitating your life.
Useful if you have a quick way to validate all players data massively to confirm if the account is not legit, however, you might get a lot of fake or troll reports.
4 is the same thing I've discussed here.


5. This doesn't make any sense unless you're trying to restrict multi-accounts.

Now, about [2] and other comments like @Evil Puncker rightly said about using IA for detection. This is a gray area and I'll try to explain why.
To begin with, there's little to none literature or practical examples of this being applied in a real life scenario. When I've built the antibot project for my private server years ago I could only find 3 references and neither of them were close to what I was trying to do.
You can check the article (in PT-BR) here: antibot/Article.pdf at master · andersonfaaria/antibot (https://github.com/andersonfaaria/antibot/blob/master/Article.pdf)

You need to understand that this is not an easy task first of all, it involves identifying all types of bots including assisted ones and trainning algorithms to try and find a specific pattern to how they behave to separate them from real players. Apart from this, hundreds of millions of dollars were spent in this direction to try and make games bot-safe with the biggest examples being Valve and recently Riot.
Meaning, any person that do this work wouldn't give hints on how they did it, but instead would built a company to sell this as a service (like other bot-detection companies). This relies heavily on types of bots people use, their functions and which data you can collect to distinguish it.
Now, for my personal proof-of-concept project I have noticed most bots didn't used "noise" in the intervals they apply between functions. The interval might make the bot behave more human-like in what comes to speed (time between actions) but it would still be showing as a bot if you compare other layers of the speed, such as mean and closeness to mean (standard deviation).
View attachment 76441
Even like this, giving enough randomness could be enough to mask the actions less predictable.
One might argue that in the future AI will be able to be used in a better way to instantly spot bots and group them by functions they are utilizing in a given interval, however, when the technology to detects evolve so does the one to improve/create bots. You can expand that until you actually have 2 artificial brains fighting to see who will win (spoofer vs. detector), even so the spoofer will have always have a reward period until detector reaches a point to caught it.

Great read and information. I like your shart, explains a lot

 

[oen432]

Damn, that's a lot of effort to get rid of 60% of potential players.

 

[Ftilt]

Damn, that's a lot of effort to get rid of 60% of potential players.

Idc about bots, but designing a game around botters is really annoying. If i was to create some rpg server and a handful of players destroyed the economy and ranking/clearing every quest for everyone since their just abusing the fact they hunt med spawns for 24/7 then yeah, i wouldnt care about losing 60% potential 'players' but have a healthy environment in the type of game i actually want to develop.

 

[Mateus Robeerto]

I understand that if I add anti-bot measures, players may stop playing in my OT for at least three days. I understand that people have commitments like work, studies (college) and other types of responsibilities that consume their time, leading them to use bots. However, I would like them to play manually, without the use of bots. It's becoming more and more complicated to deal with this


For whoever invented this OTClient bot, it would be good not to have created this type of bot.

 

[Sir Mirror]

I play on a long time running 7.4 based open tibia server and from personal experience and what I see in other people this game allows you to receive a lot of passive gains from custom foods they've added into the game as well as increasing the afk timer to 30 minutes, along with custom equipment pieces that you can get later in game from quests that are 2.5hr rechargeable ring of healings - and when its time to come back and make runes or whatever there is no exhaust on rune spells, so you can burn your mana fast. I think when the game design is engaging and progressive people will feel more desire to do things manually, this is as well is just going to attract a non-botting player base that outlives any pro-botting player base. Good staff too

 

[Silba]

I play on a long time running 7.4 based open tibia server and from personal experience and what I see in other people this game allows you to receive a lot of passive gains from custom foods they've added into the game as well as increasing the afk timer to 30 minutes, along with custom equipment pieces that you can get later in game from quests that are 2.5hr rechargeable ring of healings - and when its time to come back and make runes or whatever there is no exhaust on rune spells, so you can burn your mana fast. I think when the game design is engaging and progressive people will feel more desire to do things manually, this is as well is just going to attract a non-botting player base that outlives any pro-botting player base. Good staff too

Seems like some quite healthy changes to discourage botting, people will still bot it and benefit from these changes though, however the impact is much less - this is exactly the type of changes i'm talking about, they do not harm casual players but they do harm botters. Damn even that 30 minute afk timer is a great leap forward in reducing the effectiveness of training using a bot.

 

[Mateus Robeerto]

I think it would be a good idea to implement a 'modal window' type system when a player uses the bot for more than 6 hours. Automatically, a message would be displayed in a modal window, for example: 'If you wish to continue playing without using a bot, press the 'Yes' button.' If the player clicks 'Yes', he can play normally. Another message example would be: 'If you want to play without using a bot, press the 'No' button.' If the player clicks 'No', they will be logged out for a period of time. I'm not aware of how bots work, but it might be interesting to reward players who report or eliminate other players who use bots for a long period of time. They can receive rewards like extra experience or something interesting for fighting players using bots.

 

[BahamutxD]

I think it would be a good idea to implement a 'modal window' type system when a player uses the bot for more than 6 hours. Automatically, a message would be displayed in a modal window, for example: 'If you wish to continue playing without using a bot, press the 'Yes' button.' If the player clicks 'Yes', he can play normally. Another message example would be: 'If you want to play without using a bot, press the 'No' button.' If the player clicks 'No', they will be logged out for a period of time. I'm not aware of how bots work, but it might be interesting to reward players who report or eliminate other players who use bots for a long period of time. They can receive rewards like extra experience or something interesting for fighting players using bots.

Sorry to say but thats a terrible idea.

First you are assuming the player can read, that actually understands the language, is willing to read, understands the message and is actually willing to coop.

You also assume no players will missclick or select the wrong answer. You'd be surprised how many people fail a test for missing the date/name or other simple questions

That system is very intrusive and hurts legit players for no reason. What if that modal window makes you die?

The solution is to build your game around being fun and not a grind fest that can be acomplished by a 1 brain cell living being just to reach a goal where "the fun" is.

There is a reason why big MMOs are selling level boosts, faction changes and so. They own their game market and can monetize features that players are willing to pay for. Before this, they'd be forced to either bot or not play at all. The so famous WoW token is an excellent example on how to combat bots and actually profit out of it.

Theres a lot of things you can do to have a healthier game and lower the chances of people botting.

Level caps to move the shift from infinite grind (vertical progression) to other objetives and have some kind of horizontal progression instead (group content, PvP, achievements, side activities...), more quests/story/side missions that rewards higher than mindless monster killing, spell rotations/combos and several different monster behaviours...

I'm not even suggesting to put a cap to how long you can kill monsters (stamina) or play a day - if players enjoy that way of playing let them do it but if your game is fun and manual gameplay is so much faster/better a lot of players will shift into this type of content and while doing that the game will be much more active and healthy. If anything defines MMO players is that they love maximizing everything... so theres that.

 

[Mateus Robeerto]

Sorry to say but thats a terrible idea.

First you are assuming the player can read, that actually understands the language, is willing to read, understands the message and is actually willing to coop.

You also assume no players will missclick or select the wrong answer. You'd be surprised how many people fail a test for missing the date/name or other simple questions

That system is very intrusive and hurts legit players for no reason. What if that modal window makes you die?

The solution is to build your game around being fun and not a grind fest that can be acomplished by a 1 brain cell living being just to reach a goal where "the fun" is.

There is a reason why big MMOs are selling level boosts, faction changes and so. They own their game market and can monetize features that players are willing to pay for. Before this, they'd be forced to either bot or not play at all. The so famous WoW token is an excellent example on how to combat bots and actually profit out of it.

Theres a lot of things you can do to have a healthier game and lower the chances of people botting.

Level caps to move the shift from infinite grind (vertical progression) to other objetives and have some kind of horizontal progression instead (group content, PvP, achievements, side activities...), more quests/story/side missions that rewards higher than mindless monster killing, spell rotations/combos and several different monster behaviours...

I'm not even suggesting to put a cap to how long you can kill monsters (stamina) or play a day - if players enjoy that way of playing let them do it but if your game is fun and manual gameplay is so much faster/better a lot of players will shift into this type of content and while doing that the game will be much more active and healthy. If anything defines MMO players is that they love maximizing everything... so theres that.

I understand your opinion, my idea is really bad... I don't know what I can do with the players of this, but thanks for the tip.