The last installment in this series covered Studio One 5, which has problems. I’ll be covering this more in-depth and more clearly.

I’ll also be doing similar testing on a number of other products, but with more data.

Contents

Testing

These tests test a single thing: MIDI passthrough timing while live monitoring.

For most people this means MIDI->Software Instrument. It might mean Hardware MIDI In->Hardware MIDI Out.

IT DOES NOT MEAN playback of MIDI. IT DOES NOT MEAN playback of Audio. IT DOES NOT MEAN Clocks of any sort.

How I test - Part 1

I will be using a hardware sequencer to feed MIDI at a set tempo to the DAW and using a plugin to spit out:

  • The event offset - the amount of time after a chunk that a note should be played.

If the event offsets are 0 then there will be jitter. That means that the MIDI notes are aligned to the chunk size rather than at the position they should be.

Secondarily I’ll be recording a loopback to compare the latency of each DAW’s MIDI->Audio so that comparisons of total latency can be made.

How you can test

A simple test that anyone can perform is to use a loopback like IAC MIDI on macOS or loopBE on Windows then:

  • Setup a track (source) with MIDI notes on the downbeats.
  • Output that track (source) with MIDI notes to Virtual Output 1.
  • Create a track (dest) with MIDI input.
  • Assign Virtual Input 1 to your MIDI Input Track (dest).
  • Assign your MIDI Input Track (dest) to trigger a Sampler with a sharp drum hit.
  • Create an Audio Track and either:
    • Take input directly from the Sampler or…
    • Create a physical loopback (Output->Input) so that the Sampler track is routed to the Audio Track.
  • Hit record and either:
    • Measure the latency between the MIDI note and the recorded beat and/or…
    • Align the second1 beat of the recorded audio to the grid in your DAW .

At this point you can go through and compare the recorded audio to the grid in your DAW.

Try with a variety of buffer settings and other “playback” or “audio system” settings in your DAW. Some DAWs will play without jitter at certain settings but not others.

The behaviour also appears to vary between operating systems for the same DAWs.

Results

I sure hope you appreciate that I spent time in Windows to get Windows results. I went into convulsions twice and chopped of a leg as a sacrifice to the gods to make the pain stop… but here we are, with Windows Testing.

I’ll be testing buffers of 64, 512 and 2048 samples. It’s possible that some DAWs will change behaviour at a value between those (such as at 256 or 1024 samples). DO YOUR OWN TESTING.

I’ll also post the total RTL loopback latency for each DAW while using the same interface.

All tests done at 48khz Sample Rate.

Ableton Live

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Ardour

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Bitwig

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Cubase

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Digital Performer

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

FL Studio

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Logic

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Pro Tools

macOS

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Studio One

macOS

  • Jitter
    • 64 - YES
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Windows

  • Jitter
    • 64 -
    • 512 -
    • 2048 - *Latency
    • 64 -
    • 512 -
    • 2048 -

Meta

This post took:

  • XX hours to do
  • YY hours to do

If you appreciate the information presented then please consider joining patreon or paying us for the time spent bringing you quality content!

Be a Patreon!

If you have any questions or comments, please comment below! I read every comment and respond to most. No registration is necessary to comment, so don’t be shy.

  1. Some DAWs misplace the first beat! This appears to be a gross workaround to avoid the first note being lost in looping or from extra-buffering. The second note always appears to be fine in my testing.