OT Compressor - Meaning Behind The Values

[sabana]

Oh. So gain(on the OT) affects both upper gain and makeup gain at the same time?

[Schnork]

Gain on OT’s comp is the same thing as makeup gain. Usually every compressor has a gain knob, gain or makeup gain are usually how they’re labelled, on some compressors they have a different name.

In any case it is gain after the compression.
You could leave it at zero and use something else to pull the gain up after the compressor, same effect.
Since we’re reducing dynamic range, we need something to pull the level up again.
Usually you would try to level out the effect of the compressor, so that you have the same level like you had before the compression.

Otherwise you risk being tricked by loudness (loud sounds always better).

[sabana]

Gotcha. So i can just use the OT comp Mix(dry/wet) button on standard comp settings to start with and mess around from there.

[sezare56]

@sabana

standard comp settings

I’d say it depends on audio signal. I’d set mix to max, rate to max, lower the threshold until it lowers the sound. Above threshold, nothing happens (appart from gain).

[sabana]

Thankyou. Quick one. Is the OT compressor used all the time on every track? Or just when you want it. Filter is great. Do you always need the comp?

[sezare56]

@sabana

Do you always need the comp?

Up to you! One on Master Track FX1 may suffice…not an obligation…

[sabana]

Thanks.

[Supercolor_T-120]

@charliebay

and another on Master bus? Or if I put compressor in my submix chain then there’s no point to use it on a single track as well?

Using more than one compressor is totally fine, if anything not as common as it should be. There are a lot of mixing situations where two or more) compressors each doing a little compressing is a better choice than one compressor doing all the work. An obvious example is using a clean compressor with a high ratio and fast attack and release with the threshold set so that it only compresses the loudest parts of a track (loud transients, a spot where filter resonance t out of control, etc.) and then followed a second compressor that has more of a characteristic sound, set up for more gentle compression with a lower threshold as a creative effect. The first compressor controls the more extreme dynamics so the second compressor doesn’t have to and acts more consistently over the duration of the track, so it can be used for what it does to tonal balance, transient shape and all that, rather than controlling dynamics.

No reason the same thing couldn’t be done in the Octatrack. Gentle compression on the master, more utilitarian compression on individual tracks if it’s needed.

There used to be a good video talking about why/when to think of a compressor in terms of EQ rather than dynamics, but audio tutorial/gear review youtube is so oversaturated now that there’s no way to find it without remembering the exact title or channel it was on, and I don’t.

[Blasted_pingin]

You probably need to know when to use a compressor.

It can be used to « glue » stuff together or for instance in case of a guitar if you don’t want too much disparities of volume between finger playing and strumming.
It also can be used to pump up an element amongst other uses. I guess any video explaining what compression is will sort out some examples.

[charliebay]

Thanks, I was asking because what I used to do is using a compressor on two separate submix channels (either as insert or parallel send fx) and then another compressor (not necessary the same one) on Master track just to glue everything together. Usually I am happy with the final mix by I wasn’t sure if I overdo anything by doing so.

Also I read somewhere that if I send material further to mastering engineer, then it should not have a compressor or anything else on master (or as little processing as possible). Is it better to glue everything in mixing stage or leave it to the mastering guys?

[monom3]

generally a good mix makes the mastering easier, balance, stereo image, eq, volume levels etc are probably more important than overcompressing the master to achieve pumping tunes. The engineer would than have less room for improvement and ways to work. Mix while I create a tune helps me ending up with plenty room for eventual mastering.

[Schnork]

I´d leave that entirely to the mastering guys, meaning I´d have no processing at all on my master track when I render the mix. They usually have much better tools, now exactly what they´re doing (if not, choose someone else lol), so yeah, that´s what I do.

Edit:
Maybe if you´re mixing into a compressor it would make sense to render with the compressor on the master - someone who works like that might be able to answer that better.

@MOEast

I think I might have found out something really interesting: each THRS value equals to basically half a dB in gain, all the way from 127 down to the noise floor. A THRS of 107 is roughly a threshold of -10dBFS, and 87 a -20dBFS.
More importantly, the RAT value seems to represent a percentage (out of 128) of total gain reduction! An RAT of 64 reduced the level over the threshold to 50% AT ALL TIME, which is equivalent to a ratio of 1: 2. Similarly, an RAT of 111 reduced the level over the threshold to about 1/8 and 114 to 1/10, meaning that the effective ratios are 1: 8 and 1: 10 respectively.

Btw, that´s really interesting! I set the comp on my OT by ear and I actually like that OT pushes the user in that direction, working by ear and not rely on visuals or exact values - but yeah, very interesting what you found out!

[sabana]

Thanks. So thats where the term glue compressor comes from. Cheers

[Supercolor_T-120]

It really depends on why and how you’re using the compressor on the master buss. If you’re mixing into it then generally I’d say that’s part of your mix and if you just bypass it when you render something to send for mastering you’re not going to actually get the mix you made. On the other hand, if you get your mix more or less done wthout a master buss compressor and then add one near the end (to control the dynamic a bit so you can bring the RMS up for rendering some rough mixes, for example) then you might be better off rendering your final mix without it (or better yet sending the mastering engineer renders both with and without it).

Caveat: I have to master my own stuff 99% of the time, so this is more just passing on “best practices” wisdom than personal experience. When I do stuff that’s going to be mastered by somebody else and I’m going to be using master buss effects (glue compression, etc.) I will put those effects on early in the mixing process (usually right after I get my levels and panning roughed out) and leave them on for the final render. But I’m also pretty conservative with stuff on the master buss, I’m not a big fan of the pumping compressor sound and if I do anything it’s usually sending it out to the GSSL or using a comparable plugin, with maybe 2-3dB of gain reduction at most, just to make things sit together a bit better.

Whatever you do, leave at least 3dBFS of headroom for the mastering engineer as a rule of thumb (but ideally check with them and see what they prefer) so that if they need to do any EQ they’ll have headroom to work with. there’s almost never a good reason for an unmastered mix to peak above -6dBFS in 2021, unless you’re rendering it to check it on a car stereo or something and want to get it a bit closer to a realistic mastered level. For test mixes I’ll usually turn the master up so it peaks around -1dBFS and if the RMS is too low I might add the most transparent clipper I can find (i.e. Airwindows ClipOnly) at the very end of the master bus (right before dither) and let it take off some transients so I can render a few dB hotter. Better yet, find out what track in the mix is actually causing the peaks and get rid of them there, with a clipper or otherwise.

In the OT your options are a lot more limited of course, although because it’s all fixed point processing, in a pinch you could actually clip transients on individual tracks with some clever gainstaging or in the audio editor.