Apple Silicon native hosts/DAWs (list)

Before i return the DTK I just want test a few more plugin-hosts with my plugins

  • Apple Logic
  • Reaper

What else is there?

Adobe Audition and Traction Waveform too I think

1 Like

I recently added M1 support in my own host in its v2.3, currently tagged as RC1.
Release notes at KXStudio News

Latest Ardour 6.6 release also added arm64/M1 builds

FYI both of these tools do not have signed binaries.


VST3 test-host by Steinberg developer (with download-link)

DaVinci Resolve can host VST plugins and has an Apple Silicon native build.


regarding Steinberg VST3 testhost:
The plugin only get audio-input in the left channel, even if the plugin opens with stereo input configuration. Can somebody confirm this behaviour?

I also have checked the AudioPluginDemo from latest JUCE/develop with the VSTPluginTestHost provided by the link above and something I found is, the editor (I guess unintentionally) gets destroyed when the info-dialog (i) from the Steinberg host gets opened. When then the editor (intentionally) gets closed the plugin crashes - because the editor doesn’t exist any more. Other vendors plugins doesn’t crash, so this might be a problem with the Juce VST3 implementation. It would be cool when the JUCE people join together with the Steinberg people and look for a solution so there will be no surprises when Steinberg releases Cubase for ARM. @reuk @ed95

Thanks for raising this. Interestingly, I was able to reproduce the issue on an Intel Mac too. I’ve now pushed a fix which explicitly disallows creating a second editor in the case that an editor already exists. This patch fixes the issue in the test host.


Correct me if I’m wrong, but isn’t the VST3 spec allowing more than one editor for the same processor? I think I’m even using that feature in internal testing hosts I made.

1 Like

The spec allows it, but doesn’t require it as far as I can tell. Supporting this feature in JUCE plugins is a larger piece of work which would require deprecating parts of the AudioProcessor interface, and rewriting a significant part of the VST3 wrapper. It seemed better to fix the issue quickly, rather than to wait on this longer-running piece of work.

Were you able to reproduce the audio-input only in the left channel? It might be a special problem of the DTK.

I’ve tried this now, and I can reproduce the issue using the DSPModulePluginDemo. It seems that the host is correctly providing a single input bus with two channels. The second channel in the input bus is initially empty, so it seems that the VST3 test host is not filling the input buffer as expected. I think this is a bug in the test host.