Truth is players just simply don't want to lose when playing video games, there is a tons of data, tons of articles, tons of gdc material on the matter. People just want to feel smart when beating games, but they do not want to lose. Making a good AI for a video game is complete waste of REAL HUMAN TIME and money.
Counter-Strike is one of the most popular online, first-person shooter games in the world today. The game's Counter-Strike: Global Offensive title is immensely successful in India, with multiple e-sportsteams participating in CS:GO tournaments down the years.
But there's also a seedy underbelly of cheating that is rampant in the world of video games. In Counter-Strike, players can play as terrorists or counter-terrorist operatives at different locations. The game comes with multiple hacks, or cheat codes as they are called, which help players gain an advantage over others. Some of these codes or console commands unlock extraweapons, while others generate special in-game conditions that make it easier for a player to win. These cheats not only make the gaming experience unfair, especially in competitions and tournaments, but can also affect gamers in the long-run.
Now, computer scientists from the University of Texas at Dallas claim to havefound an interesting antidote to the problem. Researchers at the university have created a new cheat-detection system that can be used for any massively multiplayer online, or MMO, game that sends data traffic to a central server. According to the UT Dallas website, the researchers devised their approach to detect cheaters using Counter-Strike, but the same system can be applied to MMOs.
ADVERTISEMENT
Detecting cheating in MMO games is tricky, simply because the data that goes from a player's computer to a central game server is encrypted. Previously, the only way to detect any cheating anomalies would be to go through decrypted game logs. But this new mechanism bypasses the decrypted data problem by analyzing the encrypted data traffic to and from the server in real time, according to an official news release.
'Players who cheat send traffic in a different way,' says Dr Latifur Khan, an author of the study on the research, and professor of computer science and director of the Big Data Analytics and Management Lab at UT Dallas. 'We're trying to capture those characteristics,' says Khan. The findings of the study were published in the IEEE Transactions on Dependable and Secure Computing journal in August.
ADVERTISEMENT
For the study, 20 students from the university downloaded Counter-Strike and three software cheats: an aimbot, which automatically targets an opponent, a speed hack, which allows the player to move faster, and a wallhack, which makes walls transparent so that players can easily see their opponents. The researchers also set up a server dedicated for the project so the students' activity would not disrupt other online players, according to a press release.
While studying the game data, which travels in small packets or bundles of information, researchers looked at the different sizes of data packets. These vary depending on the contents. They also analyzed features like the number of incoming and outgoing packets, their size, the time they were transmitted, their direction and the number of packets in a burst (i.e. a group of consecutive packets). This monitoring of data traffic was used to identify patterns that indicate cheating, the release explains. After that, this information on the patterns was used to train a machine-learning model to predict cheating based on patterns and features in the game data.
According to the UT Dallas website, the researchers also adjusted their statistical model, based on a small set of gamers, to work for larger numbers of players. The researchers believe video gaming companies could use this technique with their own data to train gaming softwares to detect cheating. 'Our aim is to ensure that games like Counter-Strike remain fun and fair for all players,' Khan says in the release.
ADVERTISEMENT
- 17.11.2020 | 05:35 PM IST
Share your views
MOST POPULAR
SUBSCRIBE to the Lounge Newsletter
A.I. is absolutely needed for most modern day video games. If you aren't playing against an opponent, then you are probably playing against A.I. bots of some kind. Outside of computer-controlled enemies, games love to saddle you with computer-controlled allies as well. At best, dumb A.I. is a minor annoyance. At worst, it completely ruins the game you are playing. In this list, we don't rate the games based on how bad the A.I. is per se, but rather how much amusement we got out of seeing the A.I. fail. Hopefully you enjoy them as much as we did.
Doom 3 wasn't the easiest game out there, but that certainly wasn't because of its brilliant A.I. In fact, every single enemy without a projectile attack had one important weakness: tables. If you jumped on a table outside of a melee enemy's reach, there was absolutely nothing they could do. They would run around in circles around the table moaning and wailing, easily within your grasp, without ever attacking. Projectile enemies could always shoot you, but melee enemies never thought to just reach above their heads once in a while.
Video Games With The Worst Cheating Aids
Jump Ultimate Stars was a fun Smash Bros.-style DS game that brings together popular anime characters to slug it out. It was great to play against other people, but the A.I. was so stupid that people would actually recommend that you avoid buying the game if you have no one to play with. The A.I. would repeatedly jump to its death in any stage where they were able to, many times right after a respawn. On top of this, the A.I. actually cheated at high difficulties so that you could barely ever land an attack, but you could still win by waiting at the top of the screen and doing nothing while the A.I. killed itself.
The Uncanny X-Men was made notable by an episode of The Angry Video Game Nerd that focused on its incredibly crappy A.I. Since there was no way to play the game with one character only, single-player mode gave you an A.I. teammate. Unfortunately, this teammate did not necessarily know to follow you when you tried to move forward in a stage, and so you would frequently get stuck as he endlessly ran into walls. It was actually far better to purposefully let you're A.I. teammate get killed so you could play the game freely.
The biggest threat in Fallout was not your enemies, but your own party members. They absolutely loved to shoot you in the back every chance they got. They also loved to trap you in walls and corners to make it impossible to continue on with the game. Strangely enough, this pattern has stayed true even in more recent Fallout releases. Try running in front of your A.I. allies in Fallout 3 and they will shoot you in the back all the same.
Video Games With The Worst Cheating Aim
Video Games With The Worst Cheating Air
Tales of Destiny was one of the first Tales games to ever make it to American shores, but unfortunately, in the early days of Tales, the combat was a little broken. Your A.I. teammates would very frequently do nothing in battle. In fact, to get them to do anything at all, you frequently had to stand behind them so they'd wake up. Since you normally controlled a melee character in the game, this made your A.I. partners pretty much useless.