collie:

“channelConfig”: “1-2”,

Can you update this to use the new channel config object shown in the documentation I linked? I’ll take a look and see if I can figure out the SR issue. It’s strange. Can you check the output for errors?

I dont get robotic audio when using the default Apple drivers. But sr still 44100

R: 48000
Buffer Size: 64
Input device: Apple Inc.: Cabbage3-AudioDriver
Number of input channels: 2
Output device: Apple Inc.: Cabbage3-AudioDriver
Number of output channels: 2 [CabbageAudioApp.cpp (730) initialiseAudio Thread ID: 0x202280f40]
Reading CSD file: …csd

–Csound version 7.0 (double samples) Mar 4 2026
[commit: 32fe6c37f4d382e1b25d4a5fccf0e0fccc3ffa85]
using libsndfile-1.2.2
sr = 44100.0,
kr = 689.062,

I’ll be posting a mini guide for newbies in this thread.

Cheers and good luck with the project!
Julio

Also when I use either of the following ins/outs I get robotic audio:

• the normal Mac ins/outs for either the MacBook itself (MacBook Speakers/Microphone)
• the default ins/outs for my Apple earphones (External Headphones)

Only when I create a custom aggregate driver (2 mono ins, 1 stereo out with one additional unused mono in) does it fix the robotic voices issue, but I still get sr = 44100 always.

I’m not sure if the sr - 44100 and default audio driver issues are related, but I would assume so.

sr = 44100.0, kr = 1378.125, ksmps = 32
0dBFS level = 1.0, A4 tuning = 440.0

error opening file '/Users/colmdelaney/Downloads/Chordable-Mar26-v130/debug.csv'

from file /Users/rwalsh/Downloads/Chordable-Mar26-v130/Chordable-Mar26-v130.csd (1)
fprintks
gSlogfile:S,
"%.4f,%.2f,%.6f,%.6f,%.0f,%.0f,%.0f,%.0f,%.3f,":S,
ktime:k,
kfreq:k,
kamp:k,
krms:k,
kmidiin_raw:k,
kmidiin:k,
kmidi:k,
kharmtest:k,
kaccuracy:k

B  0.000
INIT ERROR in instr 1: note deleted (1 init errors)

When I change this to a valid path, it runs fine, and ksmps is repoting the correct value. So if your seeing ksmps is 10, that’s a sign that something is not running correctly with your .csd file. If you enable Cabbage: Log Verbose, you’ll get a lot more info printed. Should help us track this down

–Csound version 7.0 (double samples) Mar 4 2026
[commit: 32fe6c37f4d382e1b25d4a5fccf0e0fccc3ffa85]
Csound Command ERROR: no outfilename
overall amps: 0.0
overall samples out of range: 0
0 errors in performance
Elapsed time at end of performance: real: 0.000s, CPU: 0.000s

You’ll need to call the Csound that is installed to /Applications/Csound, i.e,

cd /Application/Csound
csound ~/Downloads/ksmps.csd

Let me know what you see here..

sr = 44100.0,
kr = 4410.000,
ksmps = 10
0dBFS level = 32768.0,

Even if I let “ksmps = 64” , then later include “setksmps 64” within the instrument, I get “ksmps = 10” in the console.

If I dont include the #define MY_BLOCK_SIZE lines it just ignores any ksmps declarations.

But the thing is… even with the #define command it STILL shows “ksmps = 10” in the console output. Even though the audio runs perfectly and ksmps is clearly not set to 10 (if it was the audio buffer wouldn’t run correctly and it would sound horrible, but it sounds fine!).

collie:

#define MY_BLOCK_SIZE #64#
ksmps = $MY_BLOCK_SIZE

This is no different to doing ksmps = 64?

I have to include the following at the top of the instruments section:

#define MY_BLOCK_SIZE #64#
ksmps = $MY_BLOCK_SIZE

Strangely I’m still seeing this printout in the console:
sr = 44100.0,
kr = 4410.000,
ksmps = 10

But when I create a new aggregate audio device in the Audio MIDI Setup App with 2 ins and 2 outs (using the same ins and outs as the first time, but adding another input which is not used in the file), then set “nchnls = 2” the audio sounds great.

In the two files attached, the code is identical apart from the UI (one for Cabbage2, one for Cabbage3) but the difference in audio is night and day. Cabbage2 is perfect/clean, in Cabbage3 there is a strong resonance frequency above the input pitch which scales with amplitude, making it the most prominent frequency. It seems the pvscaling is working fine but something is off with the audio buffer/processing.
Also no matter what I change k-rate or ksmps to, I get “kr = 4410, ksmps = 10” in the console. Also having issues with 0dBFS = 1 vs 32768.

In cabbage2, the console IDE returns:
sr = 44100.0, kr = 344.531, ksmps = 128
0dBFS level = 1.0, A4 tuning = 440.0
audio buffered in 1024 sample-frame blocks

In Cabbage3 the VSCode console returns:
sr = 44100.0,
kr = 4410.000,
ksmps = 10
0dBFS level = 32768.0,
A4 tuning = 440.0
audio buffered in 128 sample-frame blocks

ksmpsbug-cabbage3.csd (592 Bytes)

ksmpsbug-cabbage2.csd (625 Bytes)

In my csound file (cabbage3 + csound7) I am settings ksmps to 64, but when I run the file I always get the following in the console:
sr = 44100.0,
kr = 4410.000,
ksmps = 10

It’s as if something is forcing ksmps to 10 which breaks the rest of my file (phase vocoder opcodes) because it all runs on multiples of 64, causing awful audio quality

In the settings.json file I had to delete an extra backslash at the end of the following line which fixed the issue:
“systemAudioMidiIOListing": {\

After that I was able to run the audio input/output device commands.

I didn’t see any related topic on the JUCE forum so for now I don’t worry about it

Not sure what Google has in mind, presumably something similar:
https://developer.apple.com/news/?id=f5zj08ey

I tried to copy and paste the OSCinit example from the Csound documentation in an effect and it is not working. I previously tried to create an effect using the same opcode and couldn’t make it work, do you know why? I tried to install csound plugins and reinstall csound with no luck. I am using the latest pro version on MacOS.

Thanks!

<Cabbage>
form caption("Untitled") size(400, 300), guiMode("queue") pluginId("def1")
rslider bounds(296, 162, 100, 100), channel("gain"), range(0, 1, 0, 1, .01), text("Gain"), trackerColour("lime"), outlineColour(0, 0, 0, 50), textColour("black")
</Cabbage>
<CsoundSynthesizer>
<CsOptions>
-n -d
</CsOptions>
<CsInstruments>

sr = 44100
ksmps = 100
nchnls = 2

gihandle OSCinit 7770

instr 1
  kf1 init 0
  kf2 init 0
nxtmsg:
  kk OSClisten gihandle, "/foo/bar", "ff", kf1, kf2
  if (kk == 0) goto ex
    printk 0, kf1
    printk 0, kf2
    kgoto nxtmsg
ex:
endin

</CsInstruments>
<CsScore>
f0 z
i1 0 [60*60*24*7]
</CsScore>
</CsoundSynthesizer>

/*============================================================*/
UnifiedCSD:  /Users/julienbeisel/test.csd

error: syntax error, unexpected T_IDENT  (token "OSCinit")
 from file /Users/julienbeisel/test.csd (1), line 19:
 >>> gihandle OSCinit <<<
Unexpected untyped word gihandle when expecting a variable
Parsing failed due to invalid input!
Stopping on parser failure
cannot compile orchestra

This is a warning that can be ignored - if you are not using any plugin opcodes. I assume you’re not, otherwise you’d have seen compilation errors

Just getting around to a second attempt at making my app bundled so users don’t have to have Cabbage/CSound already installed. This round I’ve had success with the MacOS Standalone, AU MidiEffect, and VST3 Synth and have some notes:

The standalone worked following your instructions in this post.

The AU worked once I realized I needed to use the CabbagePluginMidiEffect name instead of the plug-in name as that’s what the binary in the component package folder was called.

The other thing I wrestled with is that there was no Resources folder in the VST3 Synth package, so I just copied the CSound framework into the root of the package folder. When that didn’t work I tried again by creating a Resources subfolder and placing it there, instead, which seems to work fine.

FWIW pluginval returns the following error but otherwise validates both plug-ins successfully:

WARNING: e[mError opening plugin directory ‘/Library/Frameworks/CsoundLib64.framework/Versions/6.0/Resources/Opcodes64’: No such file or directorye[m
e[minitAllCsoundChannels (ValueTree cabbageData) ….

Thanks for the instructions and hope my notes are helpful to someone out there on MacOS with 2.10.0.