4 ms·

The solution is simple: do everything server side. Look at World of Tanks for a great example of this. They pretty much let people run free with mods in multipl

The solution is simple: do everything server side. Look at World of Tanks for a great example of this. They pretty much let people run free with mods in multiplayer and even the worst of those mods can barely provide an advantage in online play. They do this by managing absolutely everything server side. Ammo, direction of aim, position, enemy players appearing and disappearing etc. Basically your server should be built to send only what's absolutely necessary and your client should be built to use absolutely everything so cheats can't provide much advantage rather than vice versa.

bcook 10y ago
World of Tanks is exponentially less reliant on latency than an FPS, right? Honestly, I immediately smell naivete BS when someone says "the solution is simple"...
- chris_wot 10y ago
  It might be naive, but then again I don't think games development has ever been easy. I'm constantly amazed by the skills of games devs, they just seem to surpass anything done for business applications. If a company wants to attract competitive gamers, then they have to develop games you can't cheat in during competitions. Right now, the most reliable way of doing so is server side, with all the challenges that entails. Otherwise, they have to install what essentially amounts to rootkits - which isn't exactly what most gamers want to encourage.
exDM69 10y ago
Doing everything server-side is not enough to prevent cheating. There are plenty of client-side tricks that can give the player an advantage, starting from simple visual cheats. E.g. turn the opponent's textures bright purple so you can see them better. Make the walls transparent. Aimbots that spoof mouse input. The server must be authoritative or all bets are off, but you still do need client side anti-cheat countermeasures (stuff like verifying screenshots). This stuff can't be made 100% fool proof, so competitive games where big bucks are at stake must still take place with the organizers providing the hardware and the players located in the same room.
- ultramancool 10y ago
  You're right - you can never prevent that. Which means your client needs to offer it to keep it from becoming an advantage. If the players are already bright, recognizable colours you're basically already there. Transparent walls don't help when the server doesn't send you the enemies hiding behind them until the very instant they pop out. And if you're only spoofing mouse movement, well, that is getting pretty high risk, you may have a slight advantage, but human players will still be able to kill you without too much to effort. And I think that's the main thing anticheat should be preventing.
  - cataflam 10y ago
    Even all that still doesn't stop a very old type of cheats: aim bots. Even if the server only sends you information about enemies the instant they pop out, a computer will be faster and more precise to aim and shoot at this enemy than any human player could be. Add in some randomization and delays so the server can't easily differentiate the movements from a human player, and you're a mechanically world-class player. Doing everything server-side is not enough to prevent cheating.
    - ultramancool 10y ago
      Sure, but aimbots aren't as much of a problem if they have to aim like humans - which you can enforce. You can buy some advantage like this, yes, but not much if it's implemented correctly. Additionally if you can't detect a player behind you for example, you're still dead, doesn't matter if you aimbot or not. "World class player" perhaps, but only if your bot can really play as well as a human within the confines of the game - knowing which corners to check, knowing when to look behind you, where to be aimed at a given time, etc. Your bot needs to be a hell of a lot smarter than a bot that's just given all the data and has to do basic trigonometry to aim at their head - your bot has to think a lot more like a human and may even underperform many humans. Ultimately no matter what you do, some small percentage of your players will cheat. Every major game on the market has paid cheats which fully bypass the protections available for it. It's better if those cheats can only buy them a little than buy them a lot. It's not worth investing a lot of effort in more than basic cheat protection, it is worth investing it in handling as much as you can server side and exposing as much as you can client side as that is something cheaters can't game.
swombat 10y ago
An even better solution: let the competitive advantages be given only by doing things the computer is bad at and human beings enjoy doing. If you get a competitive advantage from doing something like grinding, which computers can do better and more tirelessly than humans, and humans tend to dislike, then inevitably you will have cheating. (e.g. WoW) If on the other hand the competitive advantage is from working out a strategy of in what order to play your cards to maximise your chances of building up your strength while attacking the opponent and yet keeping something in store for surprises the opponent may spring on you - a really complex strategic task - then that's something people enjoy doing and the computer sucks at. (Hearthstone) People may be addicted to the first kind of games, but is that something that should be protected? It's kind of similar to people addicted to slot machines... sad, a reality we have to accept, but not something we should expend vast amounts of intelligence trying to sustain. Of course, there are people who will do it for the money, and don't stop them, but realise that they are basically just a few steps short of people refining formulae for meth to make it more addictive - the fact there's a business case that exploits a human weakness doesn't mean it's a productive use of intelligent people's time.
- cataflam 10y ago
  There are some things computers are really good at and that human beings also enjoy doing. Computers are insanely fast, yet plenty of reflex-based games exist. A lot of high-level competitive games have a dexterity dimension. FPS games for example are a mix of strategy and dexterity, with a big emphasis on dexterity. RTS games like Starcraft put a bigger emphasis on strategy but still require insane finger dexterity at the highest levels. The dexterity part of the game could be done far better by a simple AI. Aim bots in FPS or inhumanly micro-management in RTS have been demonstrated, yet those games are still played, and interesting. One of the examples you gave, Hearthstone, actually had many bots running on the ladder for a time. Their goal was grinding, but even so, a number of them was better than a significant portion of the human playerbase. With the recent advances in artificial intelligence, computers are not that bad at this kind of task. Overall, I don't disagree with your main point though. I have no qualms automating boring and repetitive tasks in non-competitive games. And if a competitive game has those parts, I'd probably avoid playing it. I'd rather spend my time coding than wasting it grinding because it's part of some game I otherwise enjoy. I feel it is indeed a more productive use of my time. I wouldn't go as far as judging other people who like the grind though.

bcook 10y ago

World of Tanks is exponentially less reliant on latency than an FPS, right? Honestly, I immediately smell naivete BS when someone says "the solution is simple"...

chris_wot 10y ago
It might be naive, but then again I don't think games development has ever been easy. I'm constantly amazed by the skills of games devs, they just seem to surpass anything done for business applications. If a company wants to attract competitive gamers, then they have to develop games you can't cheat in during competitions. Right now, the most reliable way of doing so is server side, with all the challenges that entails. Otherwise, they have to install what essentially amounts to rootkits - which isn't exactly what most gamers want to encourage.

exDM69 10y ago

Doing everything server-side is not enough to prevent cheating. There are plenty of client-side tricks that can give the player an advantage, starting from simple visual cheats. E.g. turn the opponent's textures bright purple so you can see them better. Make the walls transparent. Aimbots that spoof mouse input. The server must be authoritative or all bets are off, but you still do need client side anti-cheat countermeasures (stuff like verifying screenshots). This stuff can't be made 100% fool proof, so competitive games where big bucks are at stake must still take place with the organizers providing the hardware and the players located in the same room.

ultramancool 10y ago
You're right - you can never prevent that. Which means your client needs to offer it to keep it from becoming an advantage. If the players are already bright, recognizable colours you're basically already there. Transparent walls don't help when the server doesn't send you the enemies hiding behind them until the very instant they pop out. And if you're only spoofing mouse movement, well, that is getting pretty high risk, you may have a slight advantage, but human players will still be able to kill you without too much to effort. And I think that's the main thing anticheat should be preventing.
- cataflam 10y ago
  Even all that still doesn't stop a very old type of cheats: aim bots. Even if the server only sends you information about enemies the instant they pop out, a computer will be faster and more precise to aim and shoot at this enemy than any human player could be. Add in some randomization and delays so the server can't easily differentiate the movements from a human player, and you're a mechanically world-class player. Doing everything server-side is not enough to prevent cheating.
  - ultramancool 10y ago
    Sure, but aimbots aren't as much of a problem if they have to aim like humans - which you can enforce. You can buy some advantage like this, yes, but not much if it's implemented correctly. Additionally if you can't detect a player behind you for example, you're still dead, doesn't matter if you aimbot or not. "World class player" perhaps, but only if your bot can really play as well as a human within the confines of the game - knowing which corners to check, knowing when to look behind you, where to be aimed at a given time, etc. Your bot needs to be a hell of a lot smarter than a bot that's just given all the data and has to do basic trigonometry to aim at their head - your bot has to think a lot more like a human and may even underperform many humans. Ultimately no matter what you do, some small percentage of your players will cheat. Every major game on the market has paid cheats which fully bypass the protections available for it. It's better if those cheats can only buy them a little than buy them a lot. It's not worth investing a lot of effort in more than basic cheat protection, it is worth investing it in handling as much as you can server side and exposing as much as you can client side as that is something cheaters can't game.

swombat 10y ago

An even better solution: let the competitive advantages be given only by doing things the computer is bad at and human beings enjoy doing. If you get a competitive advantage from doing something like grinding, which computers can do better and more tirelessly than humans, and humans tend to dislike, then inevitably you will have cheating. (e.g. WoW) If on the other hand the competitive advantage is from working out a strategy of in what order to play your cards to maximise your chances of building up your strength while attacking the opponent and yet keeping something in store for surprises the opponent may spring on you - a really complex strategic task - then that's something people enjoy doing and the computer sucks at. (Hearthstone) People may be addicted to the first kind of games, but is that something that should be protected? It's kind of similar to people addicted to slot machines... sad, a reality we have to accept, but not something we should expend vast amounts of intelligence trying to sustain. Of course, there are people who will do it for the money, and don't stop them, but realise that they are basically just a few steps short of people refining formulae for meth to make it more addictive - the fact there's a business case that exploits a human weakness doesn't mean it's a productive use of intelligent people's time.

cataflam 10y ago
There are some things computers are really good at and that human beings also enjoy doing. Computers are insanely fast, yet plenty of reflex-based games exist. A lot of high-level competitive games have a dexterity dimension. FPS games for example are a mix of strategy and dexterity, with a big emphasis on dexterity. RTS games like Starcraft put a bigger emphasis on strategy but still require insane finger dexterity at the highest levels. The dexterity part of the game could be done far better by a simple AI. Aim bots in FPS or inhumanly micro-management in RTS have been demonstrated, yet those games are still played, and interesting. One of the examples you gave, Hearthstone, actually had many bots running on the ladder for a time. Their goal was grinding, but even so, a number of them was better than a significant portion of the human playerbase. With the recent advances in artificial intelligence, computers are not that bad at this kind of task. Overall, I don't disagree with your main point though. I have no qualms automating boring and repetitive tasks in non-competitive games. And if a competitive game has those parts, I'd probably avoid playing it. I'd rather spend my time coding than wasting it grinding because it's part of some game I otherwise enjoy. I feel it is indeed a more productive use of my time. I wouldn't go as far as judging other people who like the grind though.