Someone asked me after a show recently how the set “just ran” without me touching much between songs. The answer is snapshots, and a pretty specific way of building them that I've landed on after doing this a lot. This is that method, warts and all.
The idea in one line
One snapshot per song, one changeover snapshot, and a recall-safe list you actually trust. That's the whole thing. Everything below is just how to get there without it biting you.
Start from a real mix, not from zero
I never program snapshots on an empty desk. I build a solid base mix first — either at soundcheck or off a virtual soundcheck recording — and store that as the first snapshot. Call it “00 Base”. Every song snapshot gets made by recalling Base, making the changes that song needs, and storing a new one.
The reason is simple: if every song starts from the same known mix, the differences between snapshots are only the things you meant to change. When something sounds wrong in song six, you're looking at a short list.
What actually changes song to song
Less than you'd think. For most bands it's:
- Effects sends. Big plate on the ballad, tight room on the fast one, tap tempo delay where it matters.
- Who's featured. A guitar solo song gets that channel bumped and its reverb opened up a touch. A harmony-heavy song gets the backing vocals up.
- Playback and clicks. If there's a track, the snapshot brings it in and sets the click to the drummer's mix only.
- Mutes. Acoustic guitar that's only in two songs? It's muted everywhere else. Cleaner mix, no open mic.
What I try not to do is re-EQ vocals per song. If the vocal needs different EQ every song, something's wrong with the base mix or the mic choice, and a snapshot's just hiding that.
The recall-safe list
This is the part people skip and then regret. Recall safe means “when a snapshot fires, don't touch this.” My default list:
- Input gain on everything. If I've had to move gain during the show because a singer's suddenly quiet, I don't want the next snapshot slamming it back.
- The master fader. Obviously.
- Matrix and system outputs. The PA tuning isn't a per-song decision.
- Monitor sends, if I'm also doing mons from FOH. The drummer's “more me” from song two should still be there in song nine.
The test I use: fire a snapshot mid-song and ask whether anything just moved that I'd have to explain to the band. If yes, it goes on the list.
Naming and order
Snapshots get named “01 Song Name”, “02 Song Name” in setlist order, plus “99 Changeover” at the end. Numbers first so they sort. Song names in full, not abbreviations, because at 11pm you will not remember what “BTS2” meant.
If the setlist changes on the day — and it will — I reorder the snapshot list rather than renaming. Ten seconds of work and the go button still does the right thing.
The changeover snapshot
Last in the list. It mutes every input except walk-in music and the MC mic, drops the effects, and puts the desk in a state where the next act can plug in without any surprises through the PA. On a festival with five bands this one snapshot does more for the day than any of the song ones.
Where it goes wrong
Two mistakes I've made so you don't have to. One: storing a snapshot with a solo active. The solo gets recalled with it and you get a very confusing headphone feed mid-set. Clear solos, then store. Two: building snapshots at soundcheck volume and then the room fills. Snapshots store fader positions, not “loud enough”. That's what your master and your ears are for.
Is it worth it for a pub gig?
Honestly, for a three-piece playing 40 minutes, a Base snapshot and a Changeover snapshot is plenty. The per-song stuff earns its keep when there's playback, a big effects difference between songs, or you're mixing the same band on a run of shows and want it to be consistent every night. Match the effort to the show.
Questions, or a better way of doing it? I'm always keen to compare notes.