This is one of those terms that I’ve always felt I understand what it means, however, I can’t define it. Even though this is a little embarrassing I’d love to know the actual technical definition when equated with audio gear.
In another thread I said the following about mic pres
“I am not sure if people classify pres that are clean until they hit breaking point high headroom or if they call pres that gently compress with subtle distortion until they finally distort in an awesome and glorious way high headroom? I am not sure if high headroom pres are ones that you can crank 50 db with little noise floor”
When people say SSL or RND consoles have high head room summing sections I think they mean that you can send them a lot of signal until they start to audibly distort. Then again they say that they can hear an SSL 4000 distort in a pleasing way so does this mean they have low headroom because you hear the distortion, even if it sounds good?!?
Not sure. If the SSL 4000 has loads of head room than all compressors have even more head room:)
It’s the number of db over the nominal level (usually 0db) that a signal can go without clipping, so I would say it makes no difference whether there’s compression involved or not. What matters is the level of signal that is output. Compression is just a way of attenuating the level going in.
I figured it had to do more with self-noise (hiss, buzz, etc) than it does with distortion. For whatever reason, I seem to associate it with signal/noise ratio.
When they say the. SSL distorts pleasantly, it is not an indication of more measurable headroom. It would give you more “functional” headroom, in the sense that going past the clean limit of the circuit still gave a useable result.
Headroom can be pretty handy on an analog console.
But things can get ugly pretty quick without it on a low to mid level console, for example…
From hitting the initial gain stage at a healthy level, the signal can get pretty hot (clip) with plenty of additive EQ, then subsequently lose a lot of that signal level via compression if no make up gain is added.
At a glance, the meters may look ok, but in reality the signal has already been into clip and back.
In this scenario, the signal level should be lowered on input, also taking into account the additive EQ on the meters.
Subsequent compression should be made back up to the nominal level as required.