Counter-Strike 2

Counter-Strike 2

평점이 부족합니다.
Counter-Strike – Hidden Matchmaking and Hitscan Director Revealed
Tighty-Whitey 님이 작성
Description of Counter-Strike's hidden game mechanics.
   
어워드
즐겨찾기
즐겨찾기됨
즐겨찾기 해제
Introduction

Counter-Strike: Global Offensive is one of the most popular first-person shooter games of all time. Developed by Hidden Path Entertainment and later, after it’s initial limited-access beta-testing, published by Valve Software as a full retail title in 2012, the game was from that point maintained and organized by the same company leading the cutting-edge gaming distribution and anonymous social network platform: Steam. Having a successful track record in game development field, primarily focusing on single player experiences, Valve had only a glimpse what could be a real, true revenue maker, with their only multiplayer experiences, at the time, ranging from poor to mediocre, them being an online counterpart of a nonsensical corridor shooter: Half-Life 1 and their another unsuccessful and laughable attempt at revolutionizing the online gaming space, the worst multiplayer experience of the 20-th century: Ricochet.

After dropping the ball so badly, Valve had another project cooking up, this time by someone more competent than a bunch of university drop-outs, Counter-Strike, developed by some Asian gooseman no one remembers the name of, the project of whom was carefully and strategically stolen by Valve, leaving him at the broken trough that is the Tactical Intervention. He still beats his head against a pillow to this very day while getting kicked by North Koreans from the competitive matchmaking on Silver ranks. As sad as this story is, it all worked out for Valve at the end, as they finally licked the balls of success. Before biting into it’s crotch, they only wiggled their tongues playfully for a little bit, releasing yet another terrible game called Deathmatch Classic, one of the worst Quake parodies of all time, not counting Quake Champions. Realizing they can make a big buck on something that requires as little effort as possible, they advanced their business strategies to a whole new level, eventually making it to Counter-Strike: Global Offensive, a revamp of Counter-Strike: Source.

With the Counter-Strike’s original developer in the trash bin, they could finally explore dark deeds they have planned for a whole century – a CS:GO monetization and gambling system, exploiting every twist and turn of the law, eventually making a name for themselves as the world’s leading virtual hat and weapon skin manufacturer. Hidden Path Entertainment, the offices of which were located conveniently right across the road, have been employed and eventually enslaved by Valve to make Counter-Strike: Global Offensive, built on an outdated stub of Left 4 Dead 2’s engine branch. Making something that many didn’t seem possible, another Counter-Strike game that’s better than Counter-Strike: Source and once again, unsurprisingly, not by Valve, they’ve generously passed the whole development of the game under the Valve’s wing, leaving themselves doing forgotten strategy games that barely anyone is interested in, proving that fools never learn. Valve’s cunning plan at taking all the credit with no effort had worked once again. Making the most contradictory changes to the game’s gameplay and art-style, they’ve once again managed to ruin another Counter-Strike game, making it less enjoyable for a normal human being who isn’t interested in discussing dog testicles or seeking for blood and ego of the game’s competitive mode. Turns out normal people are a minority.

Pressuring the majority of the human race playing videogames, the evil organization known as Valve Software have gotten their hands into exploiting the sick minds of fat YouTubers and their guilty pleasures by making them and their audiences gamble away money in a rigged, not entirely RNG based case system. Little do people know, that behind the dark curtains of the game’s and Steam’s code, there is a heavy gambling element in every CS:GO’s aspect, including matchmaking, shooting, bullet behavior and even server stability. In order to figure out what exactly is going on and how blindfolded the community truly is, a research must be conducted in greater detail. Let’s touch the basics of Valve’s gamedesign to figure out the logic they are working with.
Valve's Game Design - 1

It isn’t just about placing an enemy and making him shoot the player. It’s about his play style, neurology and psychology as a construct in a theory. The game design was always a heavy focus of Valve’s games, be it singleplayer or multiplayer. It’s about giving the player a chance, regardless of how bad he is at the game, impressing him by pulling his attention in many complex techniques, while focusing on delivering a solid by predictable human standards experience and making a brain feel good. That is achieved through simplicity and relativity from the game to personality of a player playing the game. Many methods were used and explored and all has to feel "just right" or in other words, satisfying.

Half-Life 2 was briefly exploring different concepts that one would make a whole game out of. Instead, the time spent in different areas of the game is very limited to avoid repetitiveness, which is partially a reason as to why the game is perceived as a "tech-demo" by those who got bored of Half-Life. What those people do seem to forget is that it’s about the experience that you get first-hand. Not all games were meant for replayability and some that do, have to actually offer that replayability. Half-Life 2 is not one of those games, it’s the game that wasn’t meant specifically for that and it doesn’t make it any worse, just like it wouldn’t make any movie as an image worse if it’s meant to be watched once or more. It’s what about the game is aiming for. Half-Life 2 is a solid experience the flaws of which "pop-up" more obvious each playthrough because the impression factor is now gone. Most of these "flaws" are, however, simply a result of an evaporated novelty second time into the game. None of it makes the game lose it’s quality, as quality is not measured by replayability if it’s not the game’s purpose. That alone makes Half-Life 2 the best game in the series.

Setting it aside, Half-Life 2 had a system at which the enemy A.I gets dampened and more inaccurate if the player’s health is near zero. That is to "help" the player find cover or health kit to stay alive. The player does not see that and it had been secretly concluded by Valve that taking the controls away from the player if he was to fail doesn’t make the game fun and instead it assists the weak player, strategically making him feel good about the "challenge" that was never there to begin with. Is it wrong? It might be. But if the player never realizes that, he’ll never be able to judge. Is it right? If it felt like it, they think it is. This is the pillar of the Valve’s difficulty design in games.

Is CS:GO any different?
Valve's Game Design - 2

Moving to another example, Left 4 Dead, a game where they have fully implemented many assisting functions, ultimately building an entire game out of it. The game’s A.I Director is an algorithm that measures live how many resources should there be on the map or near an area player is about to visit. It "artificially" helps the player if he fails in many other methods to not only make the game feel cinematic, but pleasant for human brain. Avoidance of large hordes, limiting the infected amount only for fewer to spawn at an area that hasn’t been seen by the player, to not only optimize the game, but also to not overwhelm the player. Game is played at a fast pace with a small amount of enemies to minimize tedious gameplay for an unexperienced at videogames individual. Valve strategically conserves human’s arm and brain muscles leaving him craving for more so that he can play more and more campaigns without getting tired. "Blood should feel like ketchup and clicking at zombies with an auto-shotgun should feel like biting through tiny sandwiches."

If the player fails, he will reach the limit of an intensity factor gained through killing enemies or taking non-friendly damage. The game is just going to remove all the zombies from the map for a certain amount of flow distance if someone gets incapacitated pushing players to the intensity limit and switching to a "Relax" mode. When the invisible slider goes down to a certain extent from it’s peak at 1.0 (Maximum) to 0.8 on Expert difficulty or lower on lower difficulties meant for disabled people, that is when an algorithm will consider spawning common infected yet again. Timers for hordes or special infected respawn intervals are frozen depending on current speed of survivors through the map or their intensity level and are sped-up if the player moves fast. Suddenly removing the infected from the map to bring them back in the next horde if it’s timer is reaching it’s final number is also one of it’s functions. That makes the game more "fair" as hordes can be unpredictable which gives the player a chance to react and focus specifically on the horde only. When the "Tank" spawns, the game’s main boss, this is when potentially, player’s focus is directed to the Tank and nothing else. That is why by default, the A.I Director is scripted to not spawn any more special infected, but does not remove remaining spawned on the map before the Tank was triggered. There are many more functions and after delving deeper into the game’s variables, experimenting and testing each and every single one of them and coming up with an even more ultimate and improved balance over the years, it’s clear that the game has a very "cunning" and "tricky" system that by some can be qualified as invisible handholding. The game never gives the player any real challenges, but the option to fail is still there, it is, however, tied to a human error or server instability. Considering an extremely tiny tickrate by the modern standards on Valve’s servers, with some still believing into their server’s superiority (Left 4 Dead community that even CS:GO community puts to shame knowing that it isn’t true and preferring to play on FACEIT servers pre Counter-Strike 2) when it’s clearly not the case. Tickrate on L4D Valve’s official servers is so low that the game feels sluggish or even unstable as soon as you come back from the higher tickrate ones. Even if higher tickrate in Left 4 Dead generates it’s own share of problems, it is still a more comfortable experience to play with a higher, balance numbered tickrate. Even if, fundamentally, by increasing it’s tickrate, the game now becomes smoother and has no lagging enemies, the game’s stability starts truly showing Left 4 Dead 2’s limitations and artificial difficulty based on server instability, such as rocks hitting the player from a corner, rubber banding infected or other desynchronization problems making the player’s experience worse. While getting hit by a rock from a corner is not specifically tickrate related, but primarily world-model related where both the world-model of the player and his camera are desynchronized, it does add up as an object that gets thrown at you is simply too laggy, especially with the movement delay.

















As you can see, there is a lot of invisible handholding involved, however, the majority of the game’s playerbase is not even aware of such algorithm’s existence. The reason for that being, the A.I Director in vanilla Left 4 Dead is simply not effective enough to either stop a rushing player who’s going to be running through an empty map due to a low common limit and many contradictory factors that were not fully developed or thought-out, considering Valve never had a time or will to revisit the game’s balance in it’s substantial form in Left 4 Dead 2, as it was developed and released in one year. A low skill of the game’s developers at the game also did not help, as they simply could not fully test the system working under certain conditions when the player is trying to trick the A.I Director algorithm in different ways. Many issues were not fixed even under the glass-eyed schoolboy interns the TLS development team consisted of. (A Fan-made, community update "by the community, for the community") Yikes.

A lot of cunning ways to devalue the skill of a player have been made and players, themselves, don’t even know it. Player sees a small amount, perceives a limited amount and there is way more happening behind the scenes to try and pull you into their typical, invasive "in your face" gamedesign. The less the player’s brain processes, the more he wants if he enjoys it. This more works as a "trick" to keep the player in the game, rather than something that is meant to challenge the skill of the player regardless of the difficulty. Whether it is the right approach to gamedesign or not, nobody can come up with a conclusive answer, as the position on that is different. Even if theoretically the game design would have been less invasive and would have been improved way beyond of what is currently known in Left 4 Dead 2 or Valve’s older games, many developers still have no idea how to keep the player interested even at a basic level without even dirtier methods such as "overpowered weapons, many, many enemies and super random, convoluted skill trees or card systems."


What does Left 4 Dead and Valve’s typical gamedesign approaches have to do with CS:GO? There’s one more thing to add.
The Deception
Valve’s path is that of deception, rather than actually refined experiences.

Steam is not only a gaming distribution, but is also a social network platform. There are many tricks involved to make players hunt for numbers or higher Steam levels by spending a fortune on something that seemingly serves no purpose other than that of ego. People are being massively exploited to make Valve as rich as possible, while they are using working business methods to attract people with bright colors. There are movies and cartoons that do the exact same thing despite having no narrative or animation depth to them. These methods, at this time are still working, as they are new and people haven’t yet figured them out. You spend money not because there is something of value, but because you are being encouraged to. Numbers are a prime example of Valve’s deception.

Because player numbers are displayed by Valve for all popular games, also some that are less popular on websites such as SteamDB, they give an incentive for older or newer players to get older Valve’s games and play them again. High player counts displayed on Steam for Team Fortress 2, Left 4 Dead 2 and other games are forming a picture in player’s heads that these games are still relevant. In fact, this is not true and Team Fortress 2, as well as Left 4 Dead 2 have less players than what Steam displays. As funny as it is, high player count numbers are playing into people’s wishful thinking to self-program a motivation to play the game based on number, especially when other hidden systems in place make people stay in these games even longer. Even when there is no game session available in Left 4 Dead 2’s server browser, people will still fall for an old trick, while the game is, in fact, not very popular. Here are a bunch of examples of such Steam threads taking place.






























There are many more such threads posted over time with people having no problem with their server browser. The reason for that is their games aren’t truly being played by that many people. Many players are "boosting hours" as usual, but some numbers appear "out of thin air". There has even been an occurrence in history, few years ago, when Team Fortress 2’s player count glitched and was dropped by about 80%. Fake numbers are at play and that is nothing new.

Here is a detailed statistics page for Team Fortress 2 that displays an actual number of players currently playing the dead game.

https://teamwork.tf/community/statistics

While 100,000 players are displayed on Steam page, this is how many players are in the game right now actually:


If you are not convinced then there is no helping you.
CS:GO Matchmaking and Hitscan Director - 1
Here I’ll list two types of hit registration. One is hitscan, the other is projectile-based.

Projectile-based system calculates the bullet’s physical position in-game in live time, meaning the bullet that is being fired is physical. Bullet or a rocket in that instance is a projectile and it exists in a game world.

Hitscan is a hit registration that does not calculate any real bullet, instead the game fakes a bullet by VFX while the actual bullet you’re firing does not exist in the game. It depends on the server whether the hitscan hit counts or does not count and depends on a variety of factors, including hitboxes and server tickrate.

















This ties in with the ultimate practice that Valve have been using. Both CS:GO and Dota 2 came out after their advanced breakthroughs with the A.I Director, meaning dumping such algorithms and systems will not be in their interest for their next games. Their A.I Director had progressed beyond of what has been known in Left 4 Dead 2 and fuses the matchmaking and gameplay of both games together. A dirty secret of Valve is that it’s not that they can’t afford or don’t want to spend on high tickrate servers, but instead, 64-tickrate servers, compared to 128-tick servers common for FACEIT servers is chosen as a middle-ground to blur the actual issue of the matchmaking and hit registration on official Valve servers, which is not tickrate or anything else they normally bring up. Over the years many people have wondered, why is it that their "skill" does not translate from FACEIT to Valve’s servers and are normally blaming a lower tickrate for that. That is however, half the issue, as tickrate 64 is not even that low. The reason for that, is that along with many fused aspects of Valve’s typical gamedesign, shooting in the game feels like a "gamble" because it is one, it is not as random as you might think. Here are the images of people complaining about the tickrate.


























Shooting in Counter-Strike: Global Offensive is being manipulated from server’s side to control the player’s score, giving certain players a higher chance of victory or loss, while the matchmaking matches you with selected players depending on the same factors and that is: your score in a previous match, time before starting the next match, your previous match result being a win or loss, as well as a call for many unknown variables depending on a hidden algorithm built-in both to Steam and the CS:GO’s code. This is the reason why Valve’s matchmaking is harder than FACEIT, people who think differently are referring to a higher dedication and focus on the match, as well as increased toxicity from players on other league platforms. Anyone who was complaining about Valve’s matchmaking in CS:GO after coming from FACEIT misses this key element as to why they are not succeeding. On Valve’s official matchmaking servers, the game, due to hidden adjustments and parameters tries to encourage "teamplay" or adapts to certain player’s playstyles while also, by choice, tries to control their hit registration to give the chance to a selected by algorithm’s system player. Killing one or two enemies may be successful, but recoil for that round, after that, may be altered to give a chance to the enemy team and encourage teamplay, rather than solo-play. That does not however happen if you’re trying to kill multiple players next to each other in a short interval of time to give the chance of an ace. The majority of player’s failures at CS:GO are not due to the low tickrate, bad shooting mechanics or skill necessarily, but the game’s hidden hitscan and matchmaking director.

https://www.youtube.com/watch?v=e8J_RxqhY-s

Many people were complaining about how they can’t hit enemies, because shots "do not land" or it lands, but too suspiciously for the enemy team. The majority of those situations are completely out of the will of players, they, for the most part, have nothing to do with cheaters and these elements are adjusted from the server-side to give everyone the chance to win, move to higher ranks, trying to force multiple losses or do everything for a win. It’s about making the player feel "rewarded" rather than solely relying on his performance at the game.
CS:GO Matchmaking and Hitscan Director - 2
Here are a bunch of examples of this system at work.

1. The game will increase an accuracy of a player when he is "low hp" and make it harder for the opponents to hit him, which is an equivalent to the old Half-Life 2 system described previously. This will increase the player’s chance of victory when it is seemingly impossible. The game may also artificially choose to bring player’s health to 1 hp instead of taking all 100 hp if the system considers that a certain player hasn’t taken enough shots to die, to once again give the player a chance if he wasn’t stressed by bullets significantly enough. Instead of taking his chance away completely, it will give the player a chance to strike back. That’s why 1 hp players are a common occurrence.

2. As soon as one player dies, the game in rare instances will "help" the team with a lower amount of players to try and even out the number of remaining living players, bringing more "dynamic" rounds with cinematic 1v1s and battles where no one can hit each other. (This, along with many other interconnected factors is dependent on other factors and if the right conditions are active, this outcome may happen.)

3. The accuracy of players and resistance to damage in the team is higher the closer they are to each other. As soon as one player "detaches" from a circle of team area anomalies in the opposite team accuracy may happen, such as bullet magnetization. This is a forced regulation of "teamwork". When the team members are extremely close to each other in proximity, there is an invisible "team circle" that makes all the members of that circle more accurate and resistant. If two or more players detach and they are still close to each other, the game will create separate "team circles", but they will be smaller and players inside of them will be less accurate and resistant. Being completely solo will not apply a "team circle" and make a solo player extremely vulnerable. "Team circles" depend on proximity of players to one another and position’s vertical height, meaning if the player is standing on a surface higher than that of the other players, he will be outside of the "team circle". Where the players are looking is also important. If an active smoke is near a "team circle" and if no action is taken by the team in it, the "team circle" may be temporarily suppressed. Flashes will also temporarily "paralyze" a "team circle". Remember that the opposite team may have a "superior team circle" or other assisting factors to counter their opponents.

4. Losing as many rounds as possible will increase a chance of "round comebacks". This is exactly why "suspicious comebacks" happen. This alone is a good way to make a match longer and give everyone a chance so that the game will not constantly end at 16-0. Eventually the game will help with "aim assist" if losing players are desperately trying to win. Player’s positions, health and other factors are always monitored by the system. As soon as one team wins 15 rounds, a "draw" will be offered to their opponents through many manipulations, but such system is low-frequency. Rank-wise, losing as many matches as possible will make it harder to lose more ranks, as the ranking system is not punishing.

5. Organic round advancement results are pre-defined. If possible, the game will try it’s best to "edit" the outcomes of the round to bring them to a "healthy" passage. That means, there is an estimated by the system result of rounds won and lost by each team depending on player’s combined statistics before the match in question and during. The game will then decide how likely it is for one or the other team to win. If the game "expects" you to lose by matching you with underperforming at the game players, it will then adjust many director parameters to reach an expected result. The reason why you’re seeing repeated patterns such as round comebacks at 0-3 to 1-3 and constant team switches ending at 5-10 and 7-8 is because the system is limited and tries it’s best to end rounds at an estimated result, but if there is a "drawdown" in how players play, meaning that the behavior of certain players defies the system’s expectations, the team switch will happen at other numbers, independent of that system.

If those systems would not have existed, all maps in the game would have been CT-sided in Bomb Defusal and T-sided in Hostage Rescue mode. The game’s gunplay is adjusted depending on player’s "luck factor". There is a counter to every system’s factor, but not all are fair.
There are many more factors and these are just some of them. Do keep in mind that these Director systems only work on official Valve’s ranked competitive matchmaking servers, which is why testing it with a stack of teammates and enemies you know and don’t get randomly through the matchmaking player selection system is not possible. Only Valve’s competitive servers are plagued by that system, which is also a reason why people had a hard time figuring it out. Even if the game looks "simple", it is much more complex as there are many hidden factors in play.

CS:GO Matchmaking and Hitscan Director - 3
In case somebody did not know, there is a "forced winrate" system in Blizzard’s Overwatch that tries to make the game either balanced or unbalanced at a selected match:

https://www.reddit.com/r/Overwatch/comments/ylvvmk/forced_5050_win_rate/

The same system exists in CS:GO, it is, however, way more advanced and complex rather than just being statistics based. The reason why you are topfragging one game and bottomfragging the next one is directly connected to how the matchmaking "artificially" changes the outcome of your shooting or other player’s shooting or simply matches you with poor players.

The algorithm is that if you’ve won and started another match, then either you are going to blame your next loss and bottomfragging at your teammates, your mood, think that you’re tired after a previous match, when in reality, the game tries to push you down for other players to have fun this time.

Another classic example is a drop system. While you have a chance to get two case drops per week, getting second case is much harder and if you received an expensive case, then a chance for the second case is automatically lowered to the point you almost can’t get it. This mostly happens if the first drop is considered to be expensive. And if you haven’t played the game for a while, the drop system has a higher change of dropping a more expensive item than the one you get normally if you play frequently. The reason for that is that Valve are trying to pull you in to play the game longer by paying out their fake Steam currency with a rare drop so that you are going to have a much higher incentive to continue. This has been proven countless times even in Team Fortress 2.

While the game’s Matchmaking and Hitscan Director is extremely unfair, it’s not necessarily just punishing. The CS:GO’s matchmaking will try it’s best to soften on these factors if you’re playing very frequently and struggling on one rank both, but you have to expect such practices to occasionally come in play. The game, occasionally, will try to squeeze out everything it can from your skill and will force you to focus and "earn" a rank if the game detects that you’re trying your best or that you are a candidate to replace someone in their rank. If you fail then there are other people who’ve been working to get ranked-up more than you did or the game is only testing you. It comes down to how much you try and the game does, indeed, occasionally become generous so that it won’t purely try to drag you down. This system is so complex that people only "jokingly" rarely talk about it, but normally they don’t even suspect that something like that exists.

Here is the footage of a Hitscan Director in it’s experimental form. Over the years they’ve been working on making it less blatant and use other factors rather than shooting solely.

https://www.youtube.com/watch?v=xDrgfOtr4Nk

Not only that, but the game also fakes stutters and lags for certain intervals so that not only do you experience altered shooting mechanics, but jumps in VAR, chokes and other engine inconsistencies.









By default the game’s bandwidth is limited to 1.5 mbps to make the game even jankier. It doesn’t automatically detect the connection and you have to manually change it to "Unrestricted" if your internet speed allows for that, as well as make use of console commands the game itself never tells you about.


Remember, all of this refers to Counter-Strike: Global Offensive specifically, not Counter-Strike 2. Their next title may have such system disabled on launch and later brought back, or the new sub-tick system that Valve developed will be the new blame for something even much darker.
The Conclusion and Gabe Newell’s potential addiction
Counter-Strike: Global Offensive had many issues. Many of these issues are hidden and were never uncovered or talked about in greater detail. Over the years people have been manipulated by the game’s Matchmaking and Hitscan Director and they haven’t even suspected a thing. It was never the cheaters, it is a complex and unfair system that Valve developed. From Prime, to Trust Factor complexities where the Steam itself observes whether you should be trusted to not be using third-party software as well as many other unknown, hidden in the code parameters, something is certain.


Gabe Newell had a potential drug addiction, which is partially a reason as to why Half-Life 3 hasn’t come out yet and he was depressed in some of his earlier interviews. After going through a rehab he had become even more confused.


Understanding the game’s matchmaking and Director systems may let you into the higher ranks.