Many drum machines and sequencers have multiple audio, MIDI and CV / Gate outputs.

Sometimes these are dedicated to specific instrument voices or tracks or they may be freely assignable. In an ideal world, all instrument voices, samples, voltage gates and MIDI events when triggered at the same step should output simultaneously but depending on design this may not always be the case.

MIDI is a serial protocol. Only one event at a time can be transmitted down a single cable even if multiple events are quantized to the same grid position and assigned to the same destination port.

Add polyphony and multiple continuous control data across more than one MIDI channel to a single multitimbral device and things can get messy real fast.

Connecting multiple sequencing devices if not synchronised together accurately also results in misalignment of the output signals.

Using the standard Snellen eye sight test format as an example shows how daisy chaining multiple sequencing devices can blur overall rhythmic feel of the mix audio. The more devices you have connected but not correctly synchronised or, if the generated output events are out of alignment with your audio grid reference, the less focused the feel is of your music.

Aside from the important rhythmic feel issues introduced by sync jitter, poor event alignment and offsets / latency between multiple devices - consider the sonic aspect alone. It is common in music for the bass and kick drum to occur at the same time. In terms of frequency content they overlap considerably. Shifting the alignment even in the sub-millisecond range will obviously change the sound as phase changes and comb filtering occurs. Add tempo-sync or event jitter to the source of either one and the sound of your combination kick drum / bass note will constantly vary in tone. Instead of of a tight, steady, solid bottom end holding the track together you get the wobbles - rhythmically and sonically.

Focus clarity is critical in photography, videography or graphic design work. Lens element grade, sensor pixel count and a solid tripod / mount for photography and video. Nib sharpness and a steady desk or a high resolution monitor for graphics.

Music production is no different just the medium and tools used.

Time and Sound.

At first glance it may seem like some output timing-offset differences wouldn’t matter much in real-world usage unless the numbers were ridiculously large. How many voices, samples or v-gates would ever play on exactly the same grid-tick or step position right?

Kick. Hat. Clap. Tom. Conga. Bass note. Vocal sample. Brass Stab.

Eight voices / samples all on the same grid-tick.

It’s an easy test for audio. With any sampler / sequencer, load a fast attack transient sample and make sure it’s front-trimmed accurately and then copy and rename it eight times. Create eight tracks and record identical, quantized quarter-note hits across all of them and then allocate each of the individual samples to the dedicated assignable outputs.

For drum machines with fixed voice outputs just put active steps on every instrument at the same positions - 1 / 5 / 9 / 13 etc.

Press play and record the individual outputs simultaneously.

Test results reported here are all captured to a dedicated recorder @ 96kHz.

Sequencers with multiple MIDI outputs are a little trickier to test without a logic scope.

MIDI Interfaces and stand-alone routers / thru units can have similar issues depending on design.

Converting MIDI events to audio works well for testing.

Don’t bother with DIY - get one of these - it will outlive you.