Dwarf as audio interface in Reaper DAW (Windows 10)

Has anyone managed to use the Dwarf as sound card with Reaper (Windows 10)?

Not yet, but I’ll try it at some point - it could be useful.

That eventually should be built in (you probably know it). But in some upcoming OS release


I was wondering whether this is still in the dwarf’s roadmap and any rough ETA?
re: “dwarf as an USB sound card”

1 Like

oh imagine sending a synchronous wet and dry signal to your DAW while the Dwarf works as an audio interface via usb! :smiley:

Top 30 Dreaming GIFs | Find the best GIF on Gfycat


I have been thinking a lot about this lately, I thought most users are interested on getting their units working as USB audio interface but that is not exactly the case…
Even as @LievenDV says, the idea is to get the unit working together/alongside a DAW, and having the unit as USB audio interface does not really help much.

Say you have the Dwarf working exactly in this mode, presenting itself as USB audio interface. On the DAW you would pick it as the audio interface to use, cool, then DAW receives audio from the Dwarf and audio played/generated by the Dwarf can be sent and recorded in the DAW.
But then you also want to send the output of the DAW to your USB or other audio interface, to go into the speakers, this poses a problem because the DAW audio is being sent to the Dwarf already… Some people would be ok with connecting the Dwarf directly to the speakers, but it is not the ideal usecase.

What I see people want but do not always correctly specifying it is using a MOD unit as part of the DAW without having to rearrange their setup, or at least it seems that way…
That is, the wish is for having some way to connect a MOD unit to a PC and use its FX, MIDI and other processing from within the DAW, integrating the 2.
What comes to mind first as a way to do it is a USB audio interface, but that poses problems as USB audio has very strict timing and MOD units use USB for all regular communication and management. They quickly get into conflicts for example when loading plugin graphics and having to do audio at the same time, both fighting for resources on the same data lines.
For 1.13 we increased the buffering used for USB audio, but from my experiments it is still not enough to get a reliable audio stream :frowning:

An alternative that I have been “tipping my toes” in is passing audio over network, as the MOD units always have a network connection via USB. This in theory is more doable because network packets can be prioritized (so make audio stuff high priority, graphics and other things “normal” priority).
This requires a custom program to be listening on the other/desktop side, but since we are talking about integration of audio into DAWs we already have a way to do this - plugins.
Still need to find something that works reliably, then see the best way to package that.

So my theory and hope at the moment is that we can have a way to send audio from a MOD unit to another side all happening via plugins. Already tried using jack-net stuff, but that didnt work well enough. Next up to try is GitHub - essej/aoo: AoO - Audio over OSC which looks very promising on a first glance, it is also used in Sonobus so it has been stress-tested for p2p audio, soon to see if it can be used for our usecase too…


FWIW, whenever I’ve seen the Dwarf mentioned in this context, that’s exactly what I envisioned - the device acting as a USB audio interface.

This doesn’t seem to be a problem with other devices supporting this functionality. Regarding outputting to speakers, I’d guess that’s more down to personal preference. I tend to think in terms of record with headphones on for monitoring, then use speakers for mixing / mastering. Here’s a couple videos demonstrating the HRPB being used as an audio interface. The latter part of the second one demonstrates how the HRPP is able to send both the clean and processed audio so you can record them simultaneously into separate tracks.

Basic demo: Use The HeadRush As An Audio Interface - Connect To Your Computer - YouTube
Recording both dry and wet: Setup the HeadRush Pedalboard as an Audio Interface with Pro Tools First by ThePedalGuy - YouTube

I am pretty sure that is exactly what people would like actually,

audio in into mod unit (say) input 1
mod unit process audio
mod unit connected to computer via usb
computer received processed signal via usb, does more processing
mod unit receives more processed data from computer via usb, outputs using headphone out and/or out 1/2

here the out 1/2 on the mod interface would become ‘virtual’ outs for the usb only.

i think if unreliable data stream via usb is the problem, you could block the mod interface when you’re using it as an audio interface, or only allow mod interface via wifi/bluetooth. I don’t think the audio should be sent wirelessly/ or other potentially slower method as latency and fidelity is the most important when using it as an audio interface.


Just for information, I have a USB audio interface. I plug the dwarf into that and it works fine. I also can feed the interface output back into the dwarf, which means I can process signals coming from Reaper and then feed them back using the Dwarf, play my instrument through the Dwarf and record in Reaper, feed a midi signal through my audio interface to the Dwarf and feed the audio from the Dwarf back in. Personally I have no need for the Dwarf to act as an interface itself.

1 Like

I would also “expect” the Dwarf to work as a USB Audio interface when thinking about using with DAW.
Other devices I know (e.g. I own a Yamaha THR 10) work like that.
The THR presents itself as audio device with two input channels - dry signal and processed signal.
So I can either record the dry Guitar signal to use with any software modelers/effects/ whatever, or just use the wet signal to record several tracks “as is”.
And at the same time the THR plays back audio coming from the PC, just like any USB sound card (the built-in speaker changes the possible use cases, but doesn’t really change the way to handle it).

On the dwarf this could be achieved by presenting 4 inputs to the PC:

  • signal from input 1 and 2 (unprocessed, taken just after the ADC)
  • signal arriving in the virtual plugs of output 1 and 2 (so at the end of the internal processing line, just before DAC

And then also presenting 2 outputs, directly feeding the PC signal to the outputs parallel to the processed “instrument signal”.
I would imagine turning off the Web UI while connected as Audio Device wouldn’t hurt too much in most cases - either I’d use the dry signal, not needing any additional functions available throug Web UI, or I want to record the final signal, using the Dwarf as a “Live rig”.

Of course defining the inputs and outputs through plugins instead could make this very flexible - you could then choose freely the position in the signal chain to record - just as you can currently do with the Recorder plugin.
That would allow for all kinds of crazy combinations - going Guitar into dwarf, using compressor plugin, then taking that signal into DAW to add software amp modeler of your choice, sending that signal back into dwarf and adding reverb plugin to that signal.
That would fit very well the “anything is possible” approach of MOD ecosystem, but I don’t think that this is actually a very common use case.
Or is it?

(Disclaimer: I’m just a “living room guitar player” and for sure I’m not using any device to its full potential - including the guitar itself. So maybe my ideas sound stupid to more “professional” users.)


I’d also love the Dwarf to work as an audio interface, similar to how @Casanunda explained it.

When I’m on tour with my band, I’ll bring the Dwarf and a laptop anyhow, so with this feature I’d be able to record demos etc. without having to carry another device with me.

side note:
Actually that is currently possible even without using your laptop at all, just using the Recorder Plugin (available in the Beta store).
You just have to assign some actuator to the record / stop function.

Only problem: this approach doesn’t work across pedalboards, so it won’t work for recording your whole gig (if you use different pedalboards during your show).
Also when using snapshots, you have to make sure not to stop recording by accident when switching from one snapshot to the next.

1 Like

Yes like others I’d love the Dwarf to be a standard USB Audio Interface or even better a bit more like my Helix which can do the guitar input - Dwarf fx - pc - Dwarf - out.

But did like FalkTX idea of streaming via plugins. I use Reaper Restream in Ableton that I send out to old PC’s and back running various bits and pieces over Ethernet. It can send Audio and MIDI and is a great little tool.

Appreciate that’s not possible because of Control Chain only but via USB could be handy.

I didn’t have the time to read the thread to the end. Maybe someone already suggested this.

Just thought I’d post this option for OS X users.
I’ve had success creating aggregate audio interfaces. I haven’t tried it with my Duo X though.

I think it would resolve the case you’re concerned about.

I think the same can be done on Linux tweaking ~/.asoundrc or using
alsa_in and alsa_out