Davide Pensato is founder and CEO of dpstudios, a game audio consulting service. With over 15 years in the industry, he’s worked on numerous titles and with studios worldwide, including Ubisoft, Snapchat, Nike, and Atari. Recently he finished working on a new VR title called Theseus, an immersive VR experience by Forge Reply, offering a new take on the myth of the Minotaur and the Labyrinth. He walks us through his audio workflow.

Can you start with telling us a little more about Theseus and how you first got involved in the project?

We have been working with Forge Reply since 2011 as external outsourcing studio, but in 2013 I started working in-house for Joe Dever’s Lone Wolf project. From there, getting involved in their next production was straightforward, which meant I was able to work on Theseus from the beginning of pre-production. This allowed me to concentrate on project’s vision and choosing the right technical tools for both production and integration.


What were some sound design and music inspirations for your work throughout this project?

I worked closely with Fabio Pagetti (Creative Director) and Alessandro Bragalini (Art Director) to find the right direction and the best references for the different elements of the game. They have a wide videogame and movie culture and helped me a lot with their vision. Dark Souls and Bloodborne were a main framework for mood but great importance was also given to the horror aspect. Ico and Shadow of the Colossus helped us with emotional narrative gameplay and games like Tomb Rider and Alien Isolation were strong references for audio support to gameplay.

This game is laced with suspense, and contains some outstanding sound design. How did you create this suspense in a VR setting? Did you try anything that didn’t work out and needed revision?

For Theseus, we worked on creating two different layers of suspense, the first one being environmental suspense. As the main character wanders around a huge labyrinth, we learn that he is being chased by a giant fiery monster. So to always remind us that we need to sneak and be careful while moving, distant noises and repeating voices were built in. To add to that, falling debris and some structural crumbling added tension-filled moments to help build up even greater suspense so it never remained too constant while you play. The second layer we built into Theseus was emotional suspense, which is completely built into the music. A clever use of stings (short clues) helped us create the horror feeling we defined during the audio design process and underlined, or anticipated, important narrative moments.

There are some sections of the game with an underlying soundtrack, but there are also lots of parts containing pure soundscapes which sounded really effective for VR. Was that your choice and did not having a constant soundtrack complicate things?

We decided to use music as a narrative element. The first half of the game is just soundscapes with some suspense-filled, instrumental stings. Starting from first combat, you learn that music helps you anticipate cues or reinforce gameplay. This use of music was really challenging both in terms of composition and implementation, but also offered an additional design element. In video games, we can’t really link suspense created by music to a linear scene since the player can walk back and forth or stay still for minutes in one place. So we had to compose music to follow interactive music design and then integrate it using FMOD Studio to create a state machine logic. This integration phase required a lot of tuning to better deliver the right amount of suspense in all game-play situations.

Ariadne’s voice/spirit is supposed to be your guiding companion throughout this game, and she leaves quite an impressive mark with her ethereal style. What was the creative process behind that like?

First, we defined the language, or better “non-language”. It was created by us taking English text, translating it into Greek, and then only taking the phonemes to create Ariadne’s sentences. In this way, we were able to make sure they matched the text’s length. I created the ethereal effects using a double pass of reverb: first using an inverse preset to obtain a gated voice. This enhanced the feeling of a voice coming from elsewhere. Then, I processed each sentence with a plate reverb to widen the sound and give the large, labyrinth feeling. Of course, before this reverb processing, I equalized and compressed to reinforce the presence. Finally, after post production we asked for a Greek revision to avoid inappropriate or incorrect sentences for Greek users.

Throughout the game, there are various enemies the player encounters, one of the first being the Beholder, a giant spider-like creature. Since real spiders don’t growl/hiss, what helped you come up with this creature’s voice?

In the original concept, Beholders were generated by Asterion (the Minotaur) and shared with him some physical aspects. Their mouth is similar to the one Asterion has on its stomach so I decided to give a growling taste to their voice to remind their descendancy. They also have some barbell’s, giving the impression of a hissing body part. After doing some research we also found confirmation of our theory watching the Shelob scene in The Lord Of The Rings: The Two Towers, where spiders are hissing. I started gathering source materials and organized a Dehumanizer session for this.

Walk us through the process of creating the audio for the massive minotaur. What was your inspiration for it?

Asterion is a huge monster and we had a lot of film and games references. However, we wanted something unique and very distinctive. First, we concentrated on his body. I spent few days doing research on monster foot stomps and found the right sound for its footstep to be used as a gameplay clue. The player could stand still listening to footsteps pace and volume to understand how to proceed in the level.

Then I elaborated a sound for Asterion fluid body: a constant slimy noise is linked to any movement animation of the monster and this was used as a proximity clue. Body noise is perceived only when it’s close to the player.

Finally the voice. Each reference had an interesting element but none was fully appropriate. So I decided to create the timbre from scratch using Dehumaniser to process different animal sounds. This offered me the advantage to have infinite variations available, just processing animal voices articulations with the plugin.

Were there any interesting recordings or techniques that you used while working on Theseus or with VR games in general?

While both Theseus and Asterion are walking around the labyrinth, most of the time they don’t see each other. So we wanted to give a “sense of presence” using sound occlusion, a technique by which sounds coming from different rooms are muffled and distant and their direction is from room doors. I worked closely with Alberto Barbati (Lead Programmer) who created a system of portals and nodes, generating a sound path for each audio emitter in the game. It was a challenge because I had to put portals on each opening in every room, then connect portals in a network to create sound paths. This controlled a low pass filter associated to each sound, creating occlusion and was invaluable, adding a lot of realism to the overall VR experience.

What advice would you give to anyone who is trying to create a VR sound design/soundscape?

VR Audio is not all about binaural, which is the last link in the chain, and it’s totally useless if the audio design and integration are not executed properly. When you play in VR, you feel actual dimensions like in the real word, so the right 3D positioning, distance model and reverberation is crucial. Audio designers need to pay attention to many details because the 3D World is no longer behind a monitor, but instead completely surrounding the player. If an object sounds too far away, or too close, it can really break the experience. I’d also suggest that you should be curious and experiment. In Theseus, we did exactly this with the music. At first, I was afraid that using 2D music could cause an inappropriate result for the setting, since in real world music is spatialized. However, I soon realized that we could add it as a narrative layer since the gamer’s brain can easily separate what’s real (gameplay sounds) from what’s fictional (music).

Read more about dpstudios on their website.