A FET compressor catches peaks fast, an opto levels a vocal slowly. In series each does the job it is good at. Why it works, and how to set both.
A lead vocal asks a compressor for two things at once. It has to stop the loud syllables from jumping out, which takes something fast. And it has to even out the level from one phrase to the next, which takes something slow enough that you do not hear it working. A FET compressor is good at the first job and an optical compressor is good at the second. That is the reason the chain people describe as a 1176 style FET into an LA-2A style opto turns up on so many vocals: each compressor is asked to do only the job it is built for, and neither of them has to work hard.
A FET compressor uses a transistor as its gain element and can react in microseconds. That makes it a peak catcher. It grabs a sudden consonant or a shouted word before it gets past the rest of the vocal, and a fast release lets the level come back before the next syllable. Pushed hard, that speed is audible as grab and a little edge, which can suit an aggressive vocal and work against a soft one.
An optical compressor uses a light and a light sensitive cell, and the cell does not respond instantly. Its attack is gentle and its release depends on how hard it has been working: after a short peak it recovers quickly, after a long loud passage it lets go slowly. On a vocal that reads as levelling rather than squashing. The phrases come closer together in level and the performance still breathes. What an opto will not do is stop a sharp peak cleanly, because it is too slow to see one coming.
The order follows from that. With the FET first, the opto never hears the sharp peaks, so it does not clamp down on every loud syllable and pump. It hears a vocal whose peaks are already under control and levels it gently. Put the opto first and the peaks reach it untouched, so it either misses them or reacts to them with a slow release that pulls the next words down with it.
In TANDEM the FET stage runs into the opto stage on one insert, and the VOCAL folder has four starting points. Lead Vocal Front is the classic chain, FET into opto, about 5 to 6 dB in all. Lead Vocal Even lets the opto ride the phrases by 3 to 4 dB and uses the FET only on the loudest syllables. Each of the 19 factory presets was tuned on a real recording of its own source and leaves the output at the same loudness, so switching between them is a fair comparison rather than a jump in level. TANDEM is sold on its own and is also part of All-Access, and the 7-day trial opens it the way a membership does, with no card.
Turn up INPUT until the FET catches the syllables that stick out and leaves the rest alone. ATTACK runs from 20 to 800 microseconds and RELEASE from 50 to 1100 ms: the fast end catches more, the slow end lets the front of each word through. A ratio of 4:1 is the gentle start, 8:1, 12:1 and 20:1 clamp harder, and ALL is there for when you want the vocal crushed.
Raise PEAK REDUCTION until the quiet and the loud phrases sit closer together, then bring the level back up with GAIN. COMPRESS is the softer of the two modes. LIMIT starts a little earlier and holds the level a little harder.
Press POWER to take TANDEM out and listen. With POWER off the signal is the input bit for bit, so if the compressed version comes out quieter, turn up FET OUTPUT or OPTO GAIN until the two match before you judge. A louder version always sounds better for the first few seconds.
Two compressors in series raise a practical question: how much are they taking between them? With two separate plugins you read two gain reduction figures and add them up in your head. TANDEM shows the total instead. With the METER key on GR, its one VU meter falls by the gain reduction of both stages together, and GR dB under it reads the same number. BEFORE and AFTER show the distance between the peaks and the average level over the last three seconds, before TANDEM and after it, so you see how much the vocal has been flattened rather than guess.
When the two stages together average more than 8 dB of reduction, one TOO MUCH light comes on. It changes nothing in the sound. Heavy compression can be the right call, and the light only tells you that you are there. The TANDEM quickstart goes through every control on the face and the one that lives only in your DAW's parameter list.
TANDEM runs a FET stage into an opto stage, either one alone or both, shows what they take together on one VU meter, and ships 19 level-matched factory presets.
A FET compressor uses a transistor as its gain element and reacts in microseconds, which makes it good at catching peaks. An opto compressor uses a light and a light sensitive cell, reacts more slowly and releases according to how hard it has been working, which makes it good at levelling.
Usually the FET. It takes the sharp peaks off first, so the opto after it hears a steadier signal and can level it gently instead of clamping down on every loud syllable.
There is no single number. TANDEM's lead vocal presets take about 3 to 6 dB from both stages together, and its TOO MUCH light comes on when the two average more than 8 dB, as a warning rather than a limit.
Yes. The IN key on each stage switches it in or out on its own, so TANDEM runs as a FET compressor, an opto compressor or both in series.
No. TANDEM reports zero samples of latency at every setting: neither stage looks ahead and nothing is oversampled.
Yes to both. TANDEM is one of the plugins All-Access covers, and the 7-day trial opens it the way a membership does: sign in to Nolvra Hub, with no card and no separate activation step.
Guides
You may also like
MixingMasteringDynamicsA clipper changes the shape of the wave, a limiter turns the level down for a moment. That difference decides which to use, and why most chains use both.
MixingEQVocalsSeven free plugins suggested instead of soothe, checked on each maker's own page: which are really suppressors, which are really free, and what they need.
