I have worked with sound engineering for a few years now and I would like to suggest some modifications to your vent procedure, The current settings are for a
Hard limiter. These devices tend to make prolonged speech, effects or music sound really irritating, and harsh. Some simple modifications can soften the sound somewhat from the present settings. I've explained a few of the things below the previous example given.
6. Change the following to the these settings:
a. Gain: Adjust for how loud you want people to be (set to like 10-15, your preference).
b. Attack: 0.01
c. Release: Approximately 500
d. Threshhold: Approximately -30
e. Ratio: 100
f. Pre Delay: 4.0
a.
Gain: 6-9 should be adequate for most users, ever 6db is effectively an amplitude modification factor of 2. +6db is twice as loud and -6db is 1/2 as loud as the
present sample. Too much Gain on the pre processed sample will result in some bad distortion making it sound like crap
b.
Attack: The attack time is fine, since this is effectively how quickly the compressor will affect a sound that Rises above the
Threshold (Explained later) so a smaller number is a faster and more dramatic response
c
Release: The release time just the opposite of the attack time, its how quickly the effect ceases after the sample no longer exceeds the
Threshold. 500 is uneccesarily high, as I believe the Ventrilo timing is expressed in miliseconds, though I couldn't locate the exact data in a very brief search. 50-75 miliseconds is plenty of time for release, since any sub threshold sounds will still experience the gain reduction effect beyond the inititial, again, distorting the sound.
d.
Threshold: This the Maximum volume an effect can have before it starts to experience the effect of the Compression Ratio (explained after this) Typically, comfortable volumes are between -20 and -15 db. -30 tends to be too low for most people .
e.
Ratio: This is the ratio that equals how many db it takes to go 1 db above the threshhold set by the user. typical musical compression rates are between 4:1 and 8:1. anything more than 10:1 and you are considered to be in the realm of Hard Limiting. 20:1 is what I would suggest to most users since more than this again begins to distort amplitude peaks and can screw up your sound. quality
f.
Pre Delay: This is simply how long the processor takes to analyze the signal before it plays it to the listener. This is expressed in miliseconds. 99% of the audio processors out there can process a compression effect in under a milisecond, but 2, 4, whatever, it doesn't reaaly effect anything other than how quickly you hear the processed effect.
Sample Rate isn't mentioned anywhere in discussions on compression, but its an important factor. Sample rate is simply how many samples per second of digital audio. For example 44,100 is the standard wave sampling rate used, that means 44,100 samples per second are taken. This is considered a benchmarks for clear digital audio. This becomes important because programs like ventrilo use a substantially lower sample rate. Using any sort of digital manipulation, such as compression, causes these samples to decay, and the lower the number of starting samples the smaller the amount of sample decay you can tolerate before the the listener begins to notice.
by Rhodar Tue May 20, 2008 2:58 pm
But then off to Stygia chat baby!
