In Trackmania, one of the oldest and most prestigious speedrun world records was just beaten in 6 precise inputs... What does this mean for the competitive nature in the game?
Find More Trackmania Content On My Second Channel:
@WirtualTV
You should watch me live on Twitch:
http://www.twitch.tv/Wirtual
Edited by zero
https://www.twitter.com/lolzeroedits
Find More Trackmania Content On My Second Channel:
@WirtualTV
You should watch me live on Twitch:
http://www.twitch.tv/Wirtual
Edited by zero
https://www.twitter.com/lolzeroedits
Category
🗞
NewsTranscript
00:00On the 30th of April 2024, the oldest world record in Trackmania Nations Forever was finally beaten.
00:07On the track A12 Speed, the record had long stood at a 10.61 by Hefest, but on this fateful day,
00:14the player PastaGross blew it out of the water with an astonishing 10.40. Such an accomplishment
00:20would usually be met with celebrations in the community, but not this time.
00:24You see, Pasta had just discovered a way to take down the oldest and most legendary
00:30campaign record in just 6 precise button presses. This is the advent of low-input strategy.
00:43Trackmania Nations Forever released 16 years ago, and across over a decade of fierce competition,
00:49the game has gone through many different eras. In the early days,
00:53competition on A12 Speed was just like in every other racing game,
00:57solely about who could drive the best racing lines. But as players got better and better,
01:02they realized there were a lot of quirks in the Trackmania physics engine that they could
01:05take advantage of. And in 2010, a trick called Bug Slide was discovered. Bug Slides immediately
01:12took the game by storm. Since grass is a surface where the car doesn't have much grip,
01:17when players realized it's possible to jump and land 90 degrees sideways,
01:21drifting with a sharp angle and keeping full grip, it had massive implications.
01:27On any track where there was a tight grass turn, players would now bug slide as it saved a ton of
01:32time, and world record leaderboards on these tracks became unrecognizable. Players who used
01:38to have the world record with regular driving were understandably upset. But it also just
01:42seemed like the way the game would evolve. Once a new fast strategy was found, the older one would
01:48become obsolete, similar to how MP3 players died when smartphones were invented. To keep up with
01:54the competition, you had to get with the times, and learn the new way the game was played.
02:00The Bug Slide discovery also completely changed the meta on A12 Speed. It was no longer about
02:06precise racing lines, but about who could get the best Bug Slide flick over the wall,
02:11landing down on the grass and skipping directly to the finish. And a further development on this
02:16strategy was found in 2013, when the Uber Bug was discovered. Players found out that bug sliding
02:23into a wall really messes with the game's physics, and if you do this with enough speed while pitching
02:28the car slightly upwards, the car can sometimes gain a huge boost of speed upon crashing,
02:34launching it into the air. A12 just so happened to be the perfect track to make use of this bug on,
02:41as with enough attempts, players eventually began getting launched almost directly towards the
02:46finish line. The Uber Bug was something the developers completely missed, but since it
02:50existed within the physics engine, the community decided that records with it should be allowed to
02:55go on the leaderboard. But not without controversy, many were complaining that the Uber Bug was
03:00unrealistic, and too far gone from the original racing premise. Others were completely fascinated,
03:07with breaking the game and optimizing tracks far beyond what the developers ever could have
03:11foreseen. In the years following 2013, many players tried the Uber Bug for hundreds of hours,
03:18and the record was brought down to 11.5 seconds. But in the start of 2019,
03:24the community was blown away when Half-Este got this run.
03:37In the 5 years since this run, the world records on all other official campaign tracks in the game
03:42were improved, and it was long overdue when finally, in April of 2024, after around 80 hours
03:50of methodical hunting, Pasta Gross finally beat the record, with a whopping 0.2 second margin.
03:56But as mentioned, this record was achieved in a very unusual way.
04:01Under normal circumstances, a player contesting for the world record on A12
04:06will go out to the left for a start trick, followed by a quick right-left movement to
04:10maintain full grip, making for an optimal start downhill. Then, they will use their
04:16game sense from thousands of hours to precisely set up the back wheel on the black border,
04:21and precisely time a drift so that the car both kicks slightly higher into the air,
04:26and turns enough sideways to land in a bug slide at the checkpoint.
04:30Then, once the car lifts into the air, players have to quickly react and do the best air control
04:35they can. In his world record run, Hefes realized in the middle of the flight he would overshoot
04:41the finish at full speed, so he adapted on the spot and did three short release taps,
04:47totaling 0.58 seconds spent not accelerating in the air. If Hefes could see the future and
04:53know beforehand that this exact uber bug would happen, it's been shown he could have released
04:58at the very start of the flight for 0.35 seconds, making for a lower and faster flight path,
05:05and resulting in a 10.4 finishing time. Despite only being a 10 second speedrun,
05:10there's a lot of Trackmania mastery needed to execute each part precisely enough
05:15for a time as fast as that, and compared to Hefes, Pasta's record was driven very differently.
05:22Pasta went straight forwards in the start, and did the downhill in just two inputs,
05:27one pressing right at 2.61 seconds, and releasing it again at 2.9 seconds,
05:33losing a chunk of time towards Hefes. Then, as he approached the black border,
05:37he spent another input to hold left exactly when the back wheel touched the line,
05:42followed by a precisely timed airbrake, and a mysterious final left tap just before he landed,
05:48after which all he had to do was hold the bug slide for a guaranteed world record.
05:54Pasta recorded his gameplay with an additional camera, and documented all of his hours grinding
05:59to get the eventual run, and everyone knew that it was done legitimately. But that didn't make
06:05it any less controversial. What Pasta had discovered was a sequence of inputs that
06:10guaranteed him the world record with just 6 precise button presses, perfectly taking advantage of the
06:16fact that Trackmania's physics are deterministic. What that means is that the same inputs always
06:22produce the exact same result in-game, and if anyone watching this video play out these exact
06:28inputs on A12, you will get the exact same run as Pasta did, finishing at 10.40.
06:35Now, that might sound trivial, but 6 perfect button presses back to back in Trackmania
06:40is quite difficult. The game runs at a tickrate of 100, meaning it can receive an input every
06:46one hundredth of a second, and for the recipe to work, every button press has to be exact.
06:51If you change any of Pasta's inputs by just one hundredth of a second, the entire run breaks.
06:57The best players in Geometry Dash and other rhythm games can hit many one hundredth of a
07:02second input timings, but that is with both visual and audio cues helping you perfect the
07:07timing. In Trackmania, there is nothing that can assist you, and the game doesn't give you
07:11any feedback about whether you are rushing or dragging. To discover this string of A12 inputs,
07:18Pasta had used a tool called Trackmania Interface, which makes it possible to simulate runs by
07:24manually changing input values and having the tool perfectly play them out for you.
07:29With this tool, it is also possible, and most efficient, to brute force millions of possible
07:34input combinations, until the tool finds one that produces a fast time. After spending many
07:40hours researching such a strategy, what remains is tediously trying to execute that combination
07:45by hand. This means that instead of spending every attempt going for the uberbug, most of
07:50your resets will be in the downhill if you miss either the press or release of a key in the first
07:54three seconds. Instead of using game sense and skill gathered from thousands of hours of
07:59Trackmania to get consistent uberbug setups, you are now staring at the in-game timer and
08:04visual cues to get the exact button press that you need. And once you do uberbug, rather than
08:10needing to quickly react to the car's fight path and make a split second decision to full
08:13speed or release in order to get a perfect direct finish, you don't need to worry at all,
08:19because everything is prearranged and you already know the exact outcome of your uberbug,
08:23which in Pasta's case, was just holding the buttons at full speed and soaring into the
08:28finish at 1040 without breaking a sweat. Low input runs stand out from normal driving not
08:34only because the strategies used will be more primitive, but also because the number of button
08:38presses is so abnormal, and comparing Pasta's record with the other fastest keyboard times on
08:43A12, it's clear to see just how much of an outlier it is. Because of these differences,
08:50many top players wanted low input strategies to be banned from the main Trackmania leaderboards,
08:55which seems like a really harsh move, but it's far from obvious whether allowing it would
09:00actually be a good idea for keeping Trackmania's leaderboards fun and interesting. Instead of
09:05measuring who is the fastest driver, a leaderboard with low input records would just determine who
09:10is the best at timing button presses, which there already are plenty of other games that do,
09:16such as Geometry Dash and Dance Dance Revolution, but it has nothing to do with Trackmania mastery.
09:21Any knowledge of the car's handling and game physics are irrelevant. All that matters is
09:26being precise for a handful of arranged inputs. As a faster, the Trackmania community value the
09:32records atop the leaderboard so highly specifically because they are the crowning
09:36achievements of mastering the game. A technique that removes the need for any Trackmania-specific
09:41skill also makes it impossible to find out through competition who the best player is.
09:47Perhaps the most concerning and boring thing of all is that once an optimal low input sequence
09:52is found, the best strategy for other players is to repeat that combination and duplicate the time
09:57on the leaderboard, again and again and again, which would completely kill competition on those tracks.
10:05On the other side of the coin, banning a strategy that saves time would be unprecedented in
10:10Trackmania. The goal in the game is to get the fastest time possible, no matter which strategy
10:15you use, and at no point in the game's history has a strategy been banned for making it too
10:21trivial to get the best time. It's a crazy situation to be in as a community, with no clear
10:26answer as to what is the correct move. In the midst of many discussions, a vote was held between the
10:32top players, and although opinions differed thereto, most voted in favor of it being considered a
10:38separate category. Shortly after, the leaderboard moderators made a decision to ban low input
10:43strategy runs from being uploaded to the leaderboards, stating in a forum post that their
10:48decision was made both out of practicality and to preserve the integrity and competitive state of
10:53Trackmania Nations forever, and that they're looking at adding a separate category leaderboard
10:58for low input strategy runs at a later time. To someone outside Trackmania, this might seem like
11:04a wild decision, and I'd be interested to hear your opinions in the comments about what you think
11:08the moderators should have done. But there is one more thing to consider, namely cheating. Allowing
11:14low input strategy would open a whole can of worms of potential cheating that would have to be
11:19tackled in a good way. As it stands, suspicious looking inputs are the single most effective way
11:24to catch cheaters in Trackmania, and many have been caught this way in the past. But since all
11:30inputs in a successful low input run look exactly the same, it would be impossible to tell apart
11:35runs that were played by a macro or a script, and those actually played by hand. This could
11:41partially be solved by hand camps being a requirement, but that would also drastically
11:46raise the bar of entry, which is one of the most beautiful things about Trackmania competition,
11:51that all you need to do is install the game and the competitive patch, and then you can drive your
11:55best replays and submit them to the leaderboard without any further ado. And besides, even with
12:00a hand cam, you could still receive help from a visual or audio tool that informed you when
12:05to press buttons. Using such a tool is already banned, but if you placed it on a monitor outside
12:10the field of view, cheating this way would be virtually undetectable. The only hope you could
12:15have of catching someone cheating that way would be due to their own stupidity. And as it turns out,
12:21shortly after Pasta's Record, two new improvements were found to the low input strategy on A12.
12:27The player Axel, who at first staunchly opposed low input runs, got interested in grinding A12
12:33with these new finds, and after some 30 hours, he beat Pasta's Record, also uploading video footage
12:40with proof of him doing it legitimately. Considering how actively Axel had been against LIS,
12:46many found it weird that he suddenly wanted to play this way, and how quickly he was able to
12:51beat Pasta. And because of this, people started to dig. The hand cam footage was extensively
12:57analyzed, and he was asked to provide a lot of additional proof. When asked for an unedited video
13:02of the session, he claimed it was deleted for hard drive space. And when asked for unblurred footage,
13:07he cited privacy as a reason. When asked why he didn't show his second monitor during the run,
13:13he claimed that he doesn't have one. And yet, if you zoom in, reflecting off the side of his
13:18gameplay monitor, you can clearly see a Google Chrome logo from the taskbar of a monitor that
13:24is out of frame. Nothing concrete could be proven yet, but a few weeks later, when Axel grinded and
13:30improved the record on the track again, that was the final straw for one of his friends,
13:35who had had enough of the LIS and wanted to tell the truth to the community.
13:39The friend of Axel shared proof to moderators of him admitting to cheating in DMs,
13:44with a tool that would allow him to visualize exactly when to press the buttons,
13:48which he would have open on his second monitor. This is also how he is able to say
13:53yes I got it before the car even lifts off, and also probably why his face is blurred,
14:00to not show that his face isn't looking at the gameplay screen at all. It sadly only took a few
14:05weeks for the first cheated low input runs to appear in the game, and if it had been done by
14:10someone who was more careful, this form of cheating is almost completely undetectable
14:14and impossible to manage for the moderators. The only way to moderate low input runs would
14:19be to sacrifice the accessibility of Trackmania competition, and the only upside would be slightly
14:25faster but duplicated records filling the leaderboards, which ultimately the moderators
14:30and community members deemed not worth it. In the month since April, many players have
14:36gotten closer to beating A12 without low input. The player Viyun has put in well over 1000 hours
14:43on only this track, and gotten many promising uberbugs.
14:52And Pastagross is still doing low input runs, even though there is no leaderboard yet for him
14:57to submit the times on, and he recently beat the record on D11 Acrobatic, another official track,
15:03with a new 8 input run. There's likely still a few more tracks that he can optimize this way,
15:08and only time will tell where things go next. A few years ago, nobody would have thought it's
15:14possible to beat FSA12 with such a low number of inputs, or even at all. But now, both of those
15:21things seem like they will happen, even if they aren't on the same leaderboard. Speaking as someone
15:26who actively competes in Trackmania, I personally am in favor of the ban, but I can also see some
15:32very strong arguments against the ban. So what do you think should be done? Do you think the
15:36leaderboard moderators made the right call? I'm looking forward to reading everyone's opinions
15:40in the comments, please discuss further there. And thank you so much for watching this video
15:44till the end, and I appreciate the patience as these do take a long time to put together,
15:49but expect another World Record documentary coming very soon. Until next time, have a good one.