Audio compression in Vegas?

[David Delaney]

I was just watching DJTV and they were talking about using Audio Compression - is this a good idea? What exactly does it do?

[James Binder]

It compresses the dynamic range of an audio source. In simple terms, it makes the quietest parts louder and loudest parts quieter. It’s a great idea to use (but not over use) to improve any source that has a lot of dynamic fluctuation.

[Jon Fairhurst]

Think of a simple x-y graph with a 45 degree line. That represents a non-compressed system. What ever level you put in (x-axis) translates to the same value out (y-axix).

Now imagine that the line goes up at a 45 degree angle to a given point, and then bends a bit to the right. The stuff to the left of the bend (low amplitude signals) are unchanged. The stuff to the right (loud signals) have a lower gain. That's compression.

The two main knobs are the "threshold", which sets the level at which the gain changes., and the "ratio" which tells you how much to reduce the gain.

If you set a low threshold (-20 dB) and a high ratio (10:1), you will have HUGE compression. If you set a high threshold (-3 dB) and low ratio (1.5:1), almost none of the signal is compressed, and when it is, it's not compressed by much.

The last two knobs are the attack and release time. A fast attack (say 1ms) will make the compression come on almost instantly. For instance with a loud snare beat, the whole thing gets crunched. If you set a slower attack (say 20 ms), you will get the full attack of the snare, preserving the "punch", but the rest of the snare sound will be muted.

The release time determines how long the compression stays active. With a short release time, the compression is always going on and off. You might hear a pumping sound. With a longer release time, the gain turns down for a loud sound, and it stays turned down, just in case another loud sound will follow.

Let's say you have a long attack and a short release (10ms), and there is a repetitive snare drum. Now every attack comes through, and the back end of every snare sound is mute. Now try turning up the release to a high level (500 ms). The first snare attack will be loud, but the compression will remain active for the remaining snare hits (assuming they're within 500 ms), and all of their attacks will be muted.

One last thing. Notice that straight compression reduces the overall gain. You can manually turn the gain back up to compensate. Often a compressor has an "auto gain" button that maintains your peaks. This tends to make quiet sounds louder without overloading on the peaks.

If your stuff is playing in a movie theater, you don't need much compression. The speakers can play loud sounds, and it's quiet enough that you can hear soft sounds. If you're stuff is on TV, stuff gets compressed a bit. Radio is compressed to death - people listen to it on small speakers in their noisy cars.

I hope this is helpful.

[Bill Ravens]

Depends....all music, these days, is compressed in order to get the RMS level as high as possible. Musicians want their stuff to play as loud as, or louder than others to be competitive, especially for on air radio broadcast.

[Jon Fairhurst]

For mastering purposes, people use limiters and sonic maximizers. They're very similar to compressors, but often more advanced. And, again, the compression on a CD will be different from radio, broadcast or film.

But you can also use compression within a single track. For instance, let's say that the dialog you recorded for your film is too loud at times and too soft at times. You can manually tweak the envelope, but that takes time. Let's say you're competing in the last hour of the 48-hour film project. Just compress the dialog track and go! It's not as perfect as a hand tweaked mix, but it's better than unprocessed, and it's super fast.

Drums are often compressed. You know that Phil Collins tom tom sound? Can't do that without lots of track compression.

Want a killer voiceover? Clone the voiceover track. Leave one uncompressed. That will provide the peaks and keep the track sounding alive. Compress the snot out of the other track. That track will boost up the quiet vowels. Mix to taste. Scoop the signal with some aggressive EQ (turn down some of the mids), and your thin, cheesy voice can sound BIG.

So, yeah, you can compress the overall mix, but you can also compress individual tracks, or groups of tracks, such as a string section.

But be careful. Too much compression turns the sound into a lifeless mush. If you compress away all of the loud peaks, the whole thing can sound like it's been smothered with a pillow.

[David Jasany]

Jon, thanks for the excellent explanation of compression. It was a big help for my understanding.

[Jon Fairhurst]

I'm glad you found that helpful.

Of course, my explanation will only get you so far. Try compressing some tracks. At first the differences will sound subtle - and maybe you won't hear much of a difference at all. Over time you'll hear exactly what each control does, and you'll know exactly what you want to turn up or turn down.

At least that was my experience when first messing with compression.

[David Delaney]

What is the best plugin for audio compression?

Thanks for the info BTW

[David Delaney]

What is the best plugin for audio compression?

Thanks for the info BTW

[Jon Fairhurst]

I just use the track compressor on the individual tracks and groups. I use Wave Hammer for mastering/limiting the whole shebang. I haven't had the budget to get Waves, or other high-end processing. I'm still working on building my collection of pro sample libraries...

Here's a thread from a composer's forum on high-end mastering limiters that may provide food for thought: http://vi-control.net/forum/viewtopic.php?t=5827

[Seth Bloombaum]

Ozone from Izotope is awesome. I've never done a direct comparison to Waves, which I've used in other studios and certainly has the great reputation, but I've done all kinds of audio and video production with Ozone and always been impressed with it.

[Douglas Spotted Eagle]

iZotope products are da' bomb. The full WAVES suite is sweet too, but if I had to choose ONE over the other, I'd take the iZotope product every time.

FWIW, compression and understanding it are pretty well required in the digital realm, especially if you're an old-schooler that liked to slam tape @ +9dB to get "warm." Can't do that in digital.
Learn the art of working with compression, you'll see your skills and ears take 3 leaps forward.

[Greg Boston]

Just to add my .02.

I have played guitar many years and back in the 80's I bought a Boss compressor/sustainer. By compressing with a long release time, you can get the notes on your guitar to sustain like crazy. Obviously, you don't want this all the time but it's great to have as you can reduce the force applied to plucking the strings and still get volume.

So that's an example of using a compressor on an individual instrument in a mix.

-gb-