The test I can't cheat

Black Friday is coming, and so is a pile of plugins that look great in the screenshot. So how do you tell which ones are any good before you pay?

I can’t speak for anyone else’s plugins. But I can show you the test mine have to pass, and you can run a small version of it at home.

Try this

A blind test takes about ten minutes in any DAW.

1. Duplicate the track. Plugin A on one copy, plugin B (or nothing at all) on the other.
2. Match the loudness on a meter. Skip this and you’re mostly comparing volume.
3. Loop four bars. Ask someone to mute one copy at random, or flip a coin and write it down without looking.
4. Pick the one you like, then check which it was. Do it five times.

If you picked the same one all five times, that’s probably real (luck does it about one time in 16). If you didn’t, you just saved yourself the money.

From the studio

The blind lab

Every new KERN plugin, and every big update, goes through the same test, just bigger. 115 sessions so far. Short dry recordings, picked at random from my sample library, go through my version and the plugins it competes with, all matched to the same loudness. Then I vote best and worst without knowing which is which. The names only show up at the end.

The update I’m working on now has one extra rule: it only ships if it beats the version you already have. (It’s a lot slower than trusting my ears on a good day, but I don’t get to pick which day it is.)

The plugins

• The maintenance update for all six plugins is still in testing. Every test render matches the current versions sample for sample, so the sound stays the same.
• One small change you’ll see: CHECK’s right-click menu becomes one list with headings, instead of menus inside menus.

P.S. If you run the coin-flip test on something and the result surprises you, hit reply and tell me what you compared.