Next up in the latency/jitter testing is Bitwig!

If you don’t understand what this is, please read the introduction post (This article is being updated if/when changes are made!).

This article cost me $143 to update my Bitwig license for another year.

Other parts in this series:

Contents

Introduction

Read the introduction post. I have updated this post since it was originally published.

For Bitwig I used Sampler.

Bitwig’s sampler has been the easiest to setup and work with in these tests. That may not translate to musical use, but it’s nice. I also love Bitwig’s audio settings panel. The way it communicates buffer sizes is wonderful.

Not everything about Bitwig is great, but the parts necessary to do this test are wonderful.

Reminder, don’t trust dweebs on the internet. Do the tests yourself, or at least double check my data.

64 sample buffer

Reported Latency

  • Unlabeled path - 1.33ms

64 sample buffer

Bitwig - 64 Sample Buffer (Video)

256 sample buffer

Reported Latency

  • Unlabeled path - 5.33ms

256 sample buffer

Bitwig - 256 Sample Buffer (Video)

1024 sample buffer

Reported Latency

  • Unlabeled path - 21.33

1024 sample buffer

Bitwig - 1024 Sample Buffer (Video)

Data

Here is a csv of all the data, including the config files used to plot it using deltafinder.

Bitwig data

Conclusion

  • Bitwig has zero jitter, which makes results from live where there was 2-4 samples of jitter, seem suspicious.
    • The 1-2 sample changes in frames is a measurement variation. Slight difference in where the software picked up the peak.
      • This is in contrast to Live’s results where the absolute peak was itself up to 4 samples offset.
  • Bitwig has less latency than Live by about 1-1.5ms.
  • Bitwig has less latency than the DAWs tested so far.
    • Bitwig has less latency and stable output.
  • Based on my knowledge of the device being used, Bitwig has near optimal MIDI->Audio latency for the device and no jitter.
    • This may indicate that there isn’t a tradeoff between latency/jitter (chunking of MIDI events).

I also tested other VSTi’s for my own amusement, in case Sampler was a special circumstance. All VSTi’s that I tried had similar behaviour. Bitwig’s MIDI->Midi behaviour was also similar.

Good Job Bitwig team!

Meta

This post took:

  • 47 minutes to do everything. Bitwig was simply really nice to use for the purposes of this test.