Sound System

Strata Source’s sound system supports the following file formats:

• Microsoft WAV format 8-bit unsigned PCM
• Microsoft WAV format 16-bit signed PCM
• Microsoft WAV format ADPCM
• MP3
• OGG (Opus, Vorbis, Speex & PCM supported)
• FLAC

Support for MP3, OGG and FLAC files is provided by libsndfile.

Additionally, the following sample rates are supported:

• 11 kHz
• 22 kHz
• 44.1 kHz

Sound files which do not conform to these requirements may not play correctly or at all.

Sound Scripts

Sound scripts in Strata are nearly identical to those found in other Source games. These are found in the scripts/ directory, often named in the format game_sounds_xxx.txt. The names of sound scripts to be loaded are defined in the manifest: scripts/game_sounds_manifest.txt, which contains a list of sound script files to precache.

Manifest

As previously described, game_sounds_manifest.txt defines which sound scripts are to be precached. This file is loaded only from the MOD search path.

Sound Script Format

Sound script files contain multiple keyvalues objects describing sound parameters. The name of the keyvalues object determines how the sound can be referenced in-game. For example, MySound.Test will be referenced as MySound.Test in ambient_generic.

A complete description of the sound script format is as follows:

"MySound.Test"
{
    "soundentry_version"    "2" // Always set to 2!

    "channel"   "CHAN_AUTO"
    "volume"    "1.0"
    "pitch"     "100"
    "wave"      "mysound.wav"
    "distvariant"   "240,1320"
    "rndwave"
    {
        "wave"      "mysound.wav"
        // ...
    }
    "attenuation"   "ATTN_NONE"
    "play_to_owner_only"    "0"
    "delay_msec"    "0"
}

channel (string)

This determines how the engine will allocate a channel to mix this sound.

The following are valid values:

• CHAN_AUTO
• CHAN_WEAPON
• CHAN_VOICE
• CHAN_ITEM
• CHAN_BODY
• CHAN_STREAM
• CHAN_STATIC

volume (float or string)

volume determines how loud the sound is. This is either a floating point number in the range 0 to 1, or the string VOL_NORM representing a normal volume level.

pitch (interval or string)

pitch determines if and how the engine should pitch adjust this sound.

Pitch can be an interval representing the min and max pitch of the sound, or a string representing a predefined pitch setting. A pitch of 100 is considered normal and the engine will not modify the sound.

Valid predefined pitches are:

• PITCH_NORM (pitch of 100)
• PITCH_LOW (pitch of 95)
• PITCH_HIGH (pitch of 120)

Examples

Pitched between 65% and 90% of normal:

"pitch"     "65,90"

Pitched at exactly 50% of normal:

"pitch"     "50"

Pitched at 95% (using PITCH_LOW):

"pitch"     "PITCH_LOW"

wave (string)

wave specifies a single sound file to associate with this sound. They may be prefixed with a sound char, which controls how the sound is spatialized.

soundlevel (string)

soundlevel determines how far the sound can be heard from.

Valid values for this key include:

• SNDLVL_NONE
• SNDLVL_20dB
• SNDLVL_25dB
• SNDLVL_30dB
• SNDLVL_35dB
• SNDLVL_40dB
• SNDLVL_45dB
• SNDLVL_50dB
• SNDLVL_55dB
• SNDLVL_IDLE
• SNDLVL_TALKING
• SNDLVL_60dB
• SNDLVL_65dB
• SNDLVL_STATIC
• SNDLVL_70dB
• SNDLVL_NORM
• SNDLVL_75dB
• SNDLVL_80dB
• SNDLVL_85dB
• SNDLVL_90dB
• SNDLVL_95dB
• SNDLVL_100dB
• SNDLVL_105dB
• SNDLVL_110dB
• SNDLVL_120dB
• SNDLVL_130dB
• SNDLVL_GUNFIRE
• SNDLVL_140dB
• SNDLVL_150dB
• SNDLVL_180dB

distvar (interval)

distvar gives more control over distance variant sounds. When a stereo wave file is marked with CHAR_DISTVARIANT, the left channel represents the “close” sound, and the right channel represents the “far” sound. These channels are blended together between the near and far distances.

By default, distvar sounds have a transition start and end controlled with the snd_dvar_dist_min and snd_dvar_dist_max cvar. If desired, this key may be used to specify a distance instead of relying on those cvars.

The value of this key is an interval in the format near, far, where near and far are measured in engine units.

Warning: You must specify soundentry_version 2 for this parameter to work!

Examples

A sound that should start blending with far at 512 units, and transition to the far sound at 1024 units:

"distvar"       "512,1024"

A sound that should abruptly switch to the far channel at 120 units:

"distvar"       "120,120"

rndwave (block)

rndwave contains a list of sound files that the engine will pick randomly to play. This block may contain up to 63 sounds.

Examples

Plays either test1.wav test2.wav or test3.wav

"rndwave"
{
    "wave"  "tests/test1.wav"
    "wave"  "tests/test2.wav"
    "wave"  "tests/test3.wav"
}

attenuation (interval or string)

Defines sound attenuation. This is either an interval, or a string representing one of the predefined attenuation values.

Setting this key will override soundlevel, if any was specified.

Valid predefined attenuation types:

• ATTN_NONE
• ATTN_NORM
• ATTN_IDLE
• ATTN_STATIC
• ATTN_RICOCHET
• ATTN_GUNFIRE

play_to_owner_only (bool)

Determines if the sound should only be played to the owner.

delay_msec (int)

Number of milliseconds the sound should be delayed until it starts playing.

soundentry_version (int)

Determines the sound entry version. This should almost always be set to 2 for custom scripts. When set to 1 or below, certain features may not work.

operator_stacks (block)

A keyvalues block of operator stacks for this sound entry. See sound_operator_stacks.txt for an idea of how these work.