regular glitches in audio output at 48kHz

[sevdh]

I'm getting regular glitches (every 8 seconds) with the following patch that outputs a continuous offset to the dac~:

[sig~ 0.2]
|
[dac~]

The glitches occur when the sample rate is set to 48kHz. They don't occur at a sample rate of 44.1kHz.

output: macbook pro speakers
plugdata 0.9.3-1
macOS 15.7.4

It doesn't happen with Pure Data 0.56-2.

[timothyschoen]

Unfortunately I'm not able to reproduce this. Does this always happen for you?

[sevdh]

It took me a bit of time to create a reproduction scenario and had to increase the value of [sig~ ] to 0.4.

sample rate: 48000
buffer size: 512
audio output: MacBook Pro Speakers

Here it is:

• open Plugdata first
• open the attached patch

opening Plugdata by opening the patch doesn't exhibit the problem somehow...

Running the patch in Pure Data doesn't give this problem.

The glitches don't occur with the headphone as output. Therefore I think that the glitches are somehow caused by the macbook turning on and off the power to amplifier for the the built-in speakers over and over and not caused by Plugdata... but then why would there be a difference between Plugdata and Pure Data?

PD-bug.pd.zip

[jayrope]

Please check, if the audio sample rate of your DAW session or Plugdata standaone app matches the one mentioned in Audio MIDI Setup for your output selection. If it doesn't, please make them match and recheck, if error is still persistent. I am on the same OS version than you and can't reproduce your error here.

[sevdh]

Thanks for the idea. I’ve checked it, the sample rates are identical and the problem is still there. The problem is specific for using the speakers as output and does not occur on the headphone output.

…

On 17 Mar 2026, at 19:07, jayrope ***@***.***> wrote: jayrope left a comment (plugdata-team/plugdata#2312) <#2312 (comment)> Please check, if audio sample rate of your session/Plugdata match the one mentioned in Audio MIDI Setup for your output selection. If it doesn't please make them match and recheck, if error is still persistent. — Reply to this email directly, view it on GitHub <#2312 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEBMOIH6CPOM7Z67LE5A5UT4RGH4PAVCNFSM6AAAAACWKPASE2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DANZWHE3TGMZUGU>. You are receiving this because you authored the thread.

[jayrope]

@sevdh Hi Edwin, this pretty much indicates, that the problem is either in Macos or subsequently in your speaker chain.

[sevdh]

Yes and no, when I run the same patch in Pure Data the problem isn’t there so there is some difference between the two. As I already wrote I think the glitches have to do with macOS switching on and off the speaker amplifiers. It switches the power off when there is no signal to save battery power. I’m assuming that somehow it thinks there is no signal and than thinks there is a signal, no signal, signal, etc. If so, the question is why would it do this in this situation with Plugdata and why doesn’t it do this in this situation with Pure Data.

…

On 17 Mar 2026, at 19:33, jayrope ***@***.***> wrote: jayrope left a comment (plugdata-team/plugdata#2312) <#2312 (comment)> @sevdh <https://github.com/sevdh> Hi Edwin, this pretty much indicates, that the problem is either in Macos or subsequently in your speaker chain. — Reply to this email directly, view it on GitHub <#2312 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEBMOIBDJUPXV4H7KN6IDWD4RGK7FAVCNFSM6AAAAACWKPASE2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DANZXGEZTKOBTGA>. You are receiving this because you were mentioned.

[jayrope]

@sevdh Edwin, just to clarify, I am not a dev, am just a user like you, on the same macos and Plugdata versions. Are sampling rate settings the same in both Pure Data and Plugdata?