A Study of Video Game Loudness 🔊
While the television and radio industries have widely accepted standards for how loud their audio should be, the video games industry isn't as standardised in this regard. A figure of -23 LUFS can show up with a bit of googling, but this isn't universally enforced across the whole industry. As a result, it is possible for some games to be much louder than others, which can lead to players having to change the volume of their devices when changing games. In my view, it would be ideal for players to have to touch their volume control as little as possible. To achieve this I consider it to be important to learn how loud a typical video game is, and for game developers to keep this in mind when implementing audio into their games.
Throughout the last couple of years I have been capturing gameplay of video games I've been playing in my free time. From these recordings, it is possible to attain a LUFS value determining the average loudness of each game.
I eventually built up a spreadsheet which allows me to compare the games and to judge what a typical game's loudness might be. You can view the active spreadsheet here.
Before I present the data to you, it is worth considering a couple of things.
The selection of games is largely based on my own preferences for what I play for fun. This means it is somewhat skewed towards small indie games and nostalgic personal favourites. As a result it could mean that the data may not represent the loudness of modern AAA games as effectively as smaller indie titles. Though, I think this data would be less useful to AAA developers anyway, as sound designers for AAA studios are more likely to have specified loudness requirements to work to than indie teams would. Some games are difficult to categorise between AAA and indie, but out of the games I have measured so far I would categorise 17 as indie and 8 as AAA.
It is also important to consider the way in that audio is used in a game. For instance, some games feature almost no sound other than music. Imagine a chess video game — this could potentially have long pauses between turns and as a result may present a low loudness figure unless it contained overly loud music or ambience sounds. On the other hand, you could have a fighting game with continuous loud impact sounds and upbeat music which could present a much higher figure.
Below is a chart showing the measured loudness in LUFS. Indie games are shown in green and AAA in grey. On this scale, a lower number (or a larger bar) represents a quieter game.
Considering all of the recorded games, the mean average loudness is: -19.9 LUFS.
You may notice that there are some fairly steep jumps in loudness near the quieter and louder extremes of the chart, so we may want reduce the effect of these outliers by taking the median average, which is: -20.3 LUFS.
Considering only indie games, we get a mean of -20.6 and a median of -21.5.
Considering only AAA games, the mean is -18.3 and the median is -19.8.
With this data in mind, I would personally mix a typical game to sit around the -20 LUFS mark. I think this would be suitably loud to make it unlikely that players will have to change their device's volume to experience the game at a comfortable level.