Issue link: https://digital.copcomm.com/i/1524863
M OT I O N P I CTU R E S O U N D E D I TO R S 111 how different iterations felt in multiplayer situations was essential for the Helldivers processing for example. In particular, for the Helldivers, we wanted to ensure that the processing didn't just sound awesome but enabled players who don't use microphones or keyboards to communicate with their teammates with the ping and emote systems in the game. Speaking of the Helldivers, a fun fact people may be surprised to learn that the main element of that processing is a vacuum tube IR. We found that this, once properly tuned and filtered, had the right slight boxy element we were looking for to sell fantasy of being a perpetually helmeted super soldier! DO: We did have some instances where voice lines are concatenated at runtime. A good example of this is when a Helldiver tags something or a location on the map with the ping system. For example, if you drop a pin on the map, you can get a sequence of lines such as "marking location," "South," "50 meters." In this example, three different voice lines have been played back one after the other at runtime. In all these cases, we had the actors record the voice lines in context and the editors picked out the pieces needed for each prompt. We have found that if you were to only record a part of the sequence completely contextless, it won't gel the tone or intonation leading in or out to the next part which can sound stilted. To elaborate, using the example above, we have a large range of distances that can play at the end of the sequence. The actors recorded an example of the whole sentence, so we get a natural flow ending with the final statement. So "marking location, South, 50 meters," "marking location, South, 100 meters," etc. The editors then cropped off everything but the prompts that we needed. Working like this ensured a more natural flow. For robot/creature sound design— considering movement textures, feet, vocals, battle sounds, and etc.… How many samples do you think needed to be prepared on average per robot/creature? How did you differentiate between the types? ED: I made some quick estimations using the Automaton Standard 'Devastator' enemy type as an example. It has Attack sets, Footstep and Movement sets, Damage and Death sets, and Vocalization sets, with each subset containing roughly 10 assets. Based on this, I calculated that this enemy unit has at least 160 audio assets, with some enemy types having more but probably not fewer than 160. Arrowhead's senior sound designer Kristian Johansson handled the Terminid enemies, whilst I worked on the Automatons. Both the Automatons and the Terminids function as hive mind factional races, presenting a creative challenge in developing unique characters while maintaining uniformity and consistency across each enemy type. For consistency, the Automatons sound angry, rusty, ugly, and heavily industrial, more so than sci-fi or high-tech. Each Automaton is based on different archetypes. For example, the 'Berserker' draws from the historic berserker warrior, exhibiting highly energetic and trance-like mania as it rushes players with dual rusty chainsaws. Its VO design is shrill and screechy, making it particularly more vocal than other enemy types. In contrast, the 'Devastator' has different vocal characteristics: it's dopey, bulkier, tank-like, with a lower pitched, slower cadence. This applies to its movement as well, featuring less high-frequency detail and more low- mid-bulky metal layers. The source of the Automaton's vocalizations was raw recordings from the cyborg enemy faction in the first Helldivers game. The team at Arrowhead Game Studios took turns screeching into the mic, making up words and phrases. As you can imagine, this didn't sound flattering. However, with heavy processing—glitching, dynamic squashing, and distorting using notable plugins like Decapitator, Portal, and various Melda suite plugins, along with manually chopping up syllables—I built a library of Automaton vocals for each enemy type. Which DAW is preferred by the team? Did you use middleware solutions for implementation? If so, which? F-Mod? Wwise? Something else? HS: Our team largely uses Reaper for sound design, although there are those that primarily use Pro Tools, particularly for recording Foley or dialogue. In general, the game audio community has leaned toward using Reaper, but I don't think there is a right and wrong here, every DAW can do the job. It comes down to personal preference and team pipelines. In terms of middleware, we did use Wwise, which provided us with many out-of-the-box tools to implement and mix, as well as utilizing some of our own proprietary implementation Wwise scripts and plugins. Multi-player always provides its own challenges for game audio.