A playback rig with one laptop will eventually crash. Maybe once a year, maybe once a decade — but “eventually” always arrives mid-song. Redundant playback puts two laptops in sync and an interface that auto-switches between them so the audience never knows the primary died. Here’s how to build one that actually works under pressure.

What redundant actually means

A redundant playback rig has three properties:

Two playback machines running the same session, both in audio sync.
An auto-switching interface that takes two USB inputs, listens to the primary, and instantly swaps to the backup if the primary stops sending audio.
A failover that requires no human intervention during the show. If the operator has to push a button to recover, it isn’t really redundant.

The switching interfaces

iConnectivity PlayAUDIO 12

The default purpose-built option. Two USB hosts, 12 outputs, audio-aware failover. If the primary stops sending audio for a configurable threshold (typically 250–500 ms), it switches to the backup output and stays there until the primary recovers. The drummer’s click and FOH never miss a sample.

Radial SW8-MK2

Different approach: 8 channels of audio passive-switched between two physical sources. The SW8 listens for a watchdog tone on a sentinel channel — if the tone disappears, it flips the AB switch on all 8 channels at once. Requires you to commit one of your interface’s outputs as the sentinel.

DIY with two interfaces + a mixer

Possible but fragile. Two interfaces feed two channels of a small mixer; an engineer mutes/unmutes manually. Only do this if there’s a dedicated playback tech watching the whole show. For unattended (worship) rigs, use a real auto-switcher.

Syncing the two DAWs

Both laptops must be playing the same audio in lockstep, sample-accurate. Three ways to achieve that:

MIDI Clock or MTC (basic)

One laptop is the master and sends MIDI Timecode to the slave over USB MIDI or a 5-pin DIN. The slave follows. Easy to set up; the downside is it drifts a sample or two per song, which is usually fine for short cuts.

Ableton Link

Both laptops on the same Wi-Fi (or wired Ethernet — much better) network. Link keeps the transports in tempo sync but is NOT sample-accurate. Good for jamming, not great for backing tracks.

Word Clock + identical sessions

The professional approach. Both interfaces lock to the same word clock (one is master, one is slave). Both laptops start the same session at exactly the same timestamp. As long as nothing nudges either transport, they stay in sample-accurate sync indefinitely.

AbleSet, AbleNet, or a custom OSC bridge

Modern controller apps for Ableton run on one machine, send transport over the network to both laptops, and keep them bit-identical. AbleSet is the popular off-the-shelf option — runs on an iPad, controls both laptops, handles start/stop/jumpto and song-list navigation.

Keep both machines bit-identical

The most common cause of a “failover that didn’t work” is the backup session being slightly different from the primary. To prevent that:

Use disk-image cloning (Carbon Copy Cloner on Mac, Macrium Reflect on Windows). When the primary changes, clone to the backup the same week.
Same OS version, same DAW version, same plugin versions, same audio drivers. No exceptions.
Audio files on identical paths. If the primary loads /Users/matt/tracks/song1.wav, the backup loads the same path. Don’t use relative paths that get confused.
Test on the backup, not the primary. The week’s rehearsal should be run on the backup laptop so you discover problems before show day.

Rehearsing the failover

Test the failover by yanking the primary’s USB cable mid- song during rehearsal. Confirm:

Audio doesn’t pop, click, or drop a beat.
The drummer’s click is uninterrupted.
The interface shows it switched to backup.
You can re-plug the primary and the system stays on backup until you manually swap back (so a flaky cable doesn’t cause flapping).

Do this once a month. The failover that worked last quarter may not work this quarter after a driver update.

What redundancy doesn’t fix

Bad source audio. If the stems are corrupted, both laptops play corrupted stems.
Console-level failures. Redundant playback doesn’t help if the console crashes. For mission-critical worship rigs, also consider a fully redundant console (X32 + X32 mirroring is a thing).
Power. Both laptops on the same power conditioner means a popped breaker takes both down. Use a UPS, and run the backup off a separate circuit when possible.
The operator. If the music director launches the wrong song, both laptops launch the wrong song. Train two people to run playback so an absence isn’t a single point of failure.

Who actually needs redundant playback

Worship teams running weekly services you can’t restart. Top use case.
Theatre productions where the music director can’t pause for a reboot.
Corporate events where the playback is load-bearing for the keynote.
Touring bands at festival main stages — the festival’s reputation is on the line, not just yours.

Casual club shows? One reliable laptop and a backup SSD in the case is fine. The price of redundancy is real (~$1k for the switcher + a second matched laptop), so match the complexity to the stakes.

Plan the rig

Two laptops + a PlayAUDIO 12 + two interfaces is a 4–6U rack. Lay it out in IEM Rig and the planner will sort the cabling. Related reads:

Ready to plan your own rig?

Open IEM Rig →

Redundant Playback: Two Laptops, an Auto-Switching Interface, and Why You Need One