The Audio Freak

Unfortunately due to an NDA with my current company I cannot provide Wwise audio engine captures so I will try to explain briefly my work on rainbow 6 Siege. To make things easier to understand, I am going to divide things by sound categories used on the navigation system.

The base Navigation System is divided into 3 different body archetypes and uses weapon customizations that will impact how the characters sound when they move.

• Armor Types :

In the game we have 3 types of armor depending on the operators (=characters) : Light, Medium and Heavy.

These armors are switches inside Wwise that will trigger the correct clothing/gear sounds depending on the armor. This armor is set via our game editor and it’s a standard database table that’s calling back our switches. 

We also have an additional gear type called ‘tool belt’ that will play exclusively when doing rappel (set via an RTPC) and Hazmat suits for special missions. 

The challenge I found the most difficult was to hear the difference between Light and Medium armor so that the player have a feel of the weight, you can hear the final result in my video example.

• Weapons Layers :

Depending what gun the character will use, we choose to add extra layers of weapons on top of the armor, in Wwise so we regrouped the weapons into categories such as : Handgun, LMG, MG, Shotgun and SMG.

Same principle as the gear a sounds : a switch tagged on the weapons is triggering the sound additive layers when moving. This layer will play mechanical sounds when moving depending on which gun the player has in its hands.

• Shoe types :
  

We have 2 types of shoes : sneakers for Ai enemies and Boots for Playable Characters, the boots have an EQ that we modify depending on the type of armor (using an RTPC) so that we can put more weight when the character has a medium or heavy armor. We also have a barefoot type that was used exclusively for the hostage that you need to rescue.

• Surfaces and Debris :
  

This game contained a lot of different surfaces as every map was located in a different country (snow, country with grass and dirt, muddy training camps, etc) so it was a lot of teamwork to make this functional in game. We need every mesh to be correctly assigned by the level designers to detect and play the right surfaces and sounds. It was a classic implementation using switches for surfaces although we added extra layers for debris on the ground (glass, branches, mixed debris) that was detected via ray casting around the player and playing the appropriate debris sounds.

• Integration :

All of these sounds were contained in one play event per movement e.g ‘play_nav_walk’ and it was triggering footstep, gear and the additive layers at the same time so that we didn’t loose time creating too many sounds events for each part of the body. The drawback of this type integration is the cost of the amount of voices being played at the same time so we had to limit the polyphony of the foley and play the ones that were the most important e.g. in the closest range of the player.

We had a basic animation tool to integrate the sound events on the NPCs in our game editor and I also pushed to develop a tracking tool with the help of our friendly coders to notify the sound team if any changes (timing and keyframes) were made to the animations.

To conclude, managing the Foley pipeline was a great challenge on this production and I am glad we managed to raise the ‘foley’ quality of the rainbow games, in addition to the propagation system of this game, sound was taken seriously on the production side and the players seem to enjoy it.

If you have questions and would like to have more info on the subject, just drop me a line.