Speedruns are about completing a game in the fastest time possible, but can someone go too fast? According to rules, absolutely. Equipment can change how fast a game runs, leading to results that are so wild that communities come together to ban those settings entirely.
Mathias is a speedrunner from Denmark who streams under the handle LoveBotFF. Five days ago, he achieved a new world record on the first person narrative game Blameless, surpassing the previous record of one minute, fifty four seconds by ten seconds. The run drew criticism from the game’s small community, with some accusing Mathias of cheating.
“Don’t fucking call me a cheater when there is no grounds in your outrageous accusation,” Mathias said on reddit in response to the allegations.
As the wider community searched for an explanation for Mathias’ run, a common suggestion was made: the issue wasn’t necessarily cheating, it was Mathias’ monitor. His monitor’s refresh rate was shortening load times and making his run much faster than normal.
To understand the problem, it’s important to understand what frame rate is. Frame rate is the frequency a piece of digital media is presented in, determining how many consecutive images are shown. If frame rate is high, image quality can appear to be more smooth as there are less gaps between the images output to a screen.
In video games, frame rate determines how much information is relayed to the player from moment to moment. 30 frames is a commonly accepted minimum for frame rate. Some designers prefer this frame rate for artistic reasons, as was the case in 2015’s The Order: 1886. 60 frames per second is growing in popularity as gamers and developers seek smoother gameplay, but as personal computers and games systems are able to player older games with higher framerates, there are unforseen consequences.
Consider a game like Half Life. Changing the frames per second (FPS) causes the game to run scripts faster as well, which can change how quickly the player accelerates speed or how quickly they can move up slopes. Conversely, lower frames per second can cause tossed items to have less friction and slide further along the ground and even allow enemies to fire their weapons faster.
A factor complicating the effect of FPS on games is a monitor’s refresh rate. The refresh rate of a monitor determines when the display shows new data. The standard has shifted over time, but most monitors have a default refresh rate of 60Hz, updating their data sixty times in one second. Newer monitors can go higher than this.
It’s often popular for competitive players of games like Overwatch or Street Fighter to set their refresh rates higher to avoid input lag on their movement and actions. The faster you have up-to-date information, the faster you can react to it. But much like raw frame rate generated by hardware, additional speed afforded by faster monitors can have drastic effects.
The video above shows Super Meat Boy running at an unlocked 800 frames per seconds with VSync turned off. This affects behaviour so much that it breaks the game, allowing players to rush through menus and levels at blistering speeds. It causes various glitches including clipping through walls. Because of this, standard categories of the game require that the FPS be locked to 60.
One option to ensure this is called VSync or Vertical Synchronisation, which limits a game’s framerate to the monitor’s refresh rate. VSync on a monitor with a 60Hz refresh rate would limit FPS to 60, for example.
In the case of Blameless, the Glitchless category Mathias was running does not alter game settings. Vertical Synchronisation is on by default. His monitor was running at 144Hz, and the VSync locked his frame rate to that high value. That higher refresh rate interacted with the game to produce quicker loading times and created a world record ten seconds faster than the second place run.
“I completely agree [refresh rate] shouldn’t be a factor in speedrunning at all,” Mathias told Kotaku. “But I’m not the one who created the categories or rules.”
At the moment, Mathias’ record stands at the top of the leaderboards, while the rest of the community contemplates the possibility of locking runs to a 60 FPS standard to avoid such issues in the future. The small oversight in how Blameless interacted with monitors might lead to a new rule change. Asked by Kotaku is this would be a smart choice of action, Mathias agreed.
“Yes, I think that would be a good thing to do,” Mathias said. “But I think is should also be explained to runners that disabling VSync makes the game run a lot faster. Which I would assume is what everyone wants, right? The fastest possible times.”
Comments
11 responses to “Speedrunner Accused Of Cheating Just Had A Really Good Monitor”
I am just surprised that there weren’t already rules in place for things like this.
It’s common to remove loading times from PC games by setting up an auto-splitter to pause the timer during loading screens. For whatever reason, the people running this game decided to split the leaderboards instead which lead to this.
You have to remember that the speedrunning community is incredibly fragmented. Just because something seems like best practise or common sense doesn’t mean it is going to be universally adopted. Some people establishing the rules for their games don’t know all of the options available to them to unify things, others think that their solution is better and some just don’t care.
Additionally not every game is going to have their times affected by refresh rate, only those who have their game logic tied to their render loop.
It really ends up as a decision for each category for each game what they think is best.
Nothing to do with the original topic, but physics being tied to processing lower is why I can’t play Mechwarrior 2 anymore. Sad.
You’d think after nearly a year the SR community around Blameless would have locked in some ground rules…
Considering speedruns don’t play properly anyway whats the big deal?
They all use hacks so it’s not skill based thing anyway.
i hope this is bait, some glitches have frame perfect timing to pull off, go watch a mario 64 speed run and tell me speedrunning isnt skill based pls.
Uhhhh… unless you’re doing a TAS, ‘hacks’ (external programs that give the player some advantage in the game) aren’t usually allowed in speedrunning.
I’m not sure what you mean by ‘playing properly,’ but if you find speedrunning to be trivially easy, I look forward to your forthcoming world record for single-segment 120 stars in Super Mario 64!
I don’t understand how a faster frame rate would shorten loading screens. Is the game actually leaving the loading screen up for longer than it actually takes to load the level? Because I would have thought the time to load assets off disk would be independent of frame rate (and potentially quite variable if some players are using SSDs and others not).
Some engines space out loading assets over a number of frames, calling for texture one and then waiting for the next frame before calling for texture two (or bundling a few per frame etc.). Commonly this is to show a loading animation in engines that don’t support multithreading. If they didn’t do this, the game would look like it had frozen while all the assets were loaded.
If a game using this method doesn’t disable any frame limiters before its loading screen, then it will load faster on a computer that doesn’t have any active (like Vsync on a 60Hz monitor) than one (with comparable specs) that does.
What engines do that? I’ve worked with a lot of them and I’ve never seen one that binds asset loading to the rendering pipeline. Games that need to load assets and still render a loading screen in a single thread bind frame rate to asset loading (ie. update loading animation when a resource is loaded), not the other way around.