Audio Latency in Real Time Recording with Garuda

Hello Garuda Community,

I am a teacher who gives live classes online via jitsi and records them on my desktop via kazaam. I have been using various distributions of Linux since 2007 and I have just recently started to dabble with Garuda on my laptop and have been in love with this distro for the last 6 months. Thank you all so much for this beautiful and rock-solid creation! I am also currently running MX Linux on a 2012 desktop as my daily machine. While things work okay on this fossil - needless to say - it is time for a hardware upgrade this day in age…

So I am currently putting together a totally new desktop which will serve both as my new recording device and my work computer. To this end, I am about to install Garuda Cinnamon on the following hardware:

AMD Ryzen 9 5950X CPU, ASRock Challenger D Radeon RX 6650 XT GPU, Asus ROG STRIX B550-F motherboard, WD_BLACK 1TB SN850X NVMe SSD, etc…

I intend to connect my XLR microphone up to a USB Yamaha AG03MK2 audio interface which will be connected to the Asus ROG STRIX B550-F motherboard. This will be my very first time using an external audio interface between my microphone and my computer, so I have a…

n00b question: Since I am going to be teaching and recording live online classes with the microphone hooked up to an external audio interface, could this external device tethering workflow potentially cause audio latency issues with the live class and/or class recording?

Do I need to make any changes to the kernel to lessen audio latency in this particular case?

I am hesitant to run the dedicated real-time kernel on my daily work computer that is less often used as a recording device. From reading about it, it seems that real-time is now included in the mainline Linux kernel. That said, will any tweaking need to be done to the default kernel included with Garuda to make sure I am minimizing audio latency?

PS- I am hoping that just recording the audio inputs from my mic and those of participants in the questions-and-answer session does not require the same amount of real time audio tweaking that recording a band with multiple instruments and vocals would. Correct?

PPS- I am hoping that the external USB Yamaha AG03MK2 device is plug and play with Gardua and that no bundled proprietary firmware or software will get in the way. Correct?

Thanks in advance for your help. :slight_smile:

Welcome, I use a Scarlett 2i2 with a Rode Pod Mic (XLR)
No issues with audio lag and I use Garuda as my daily driver.
I work in IT so I’m constantly on calls either via VOIP or Jitsi (using the app) and record stuff with Kooha

1 Like

Good to know. Thanks for your reply.
So you didn’t change any latency parameters in the kernel then? You are just using the stock base kernel settings that came installed with Garuda, and everything is smooth?
When I was researching this, I read that Linux audio (music) producers were either choosing a low latency kernel (on a dedicated DAW machine) or making some changes to the mainline kernel to lessen audio latency. Maybe the recent inclusion the real-time patches to the mainline kernel makes this unnecessary…?
I am not sure, audio production is something I am very new to.
Thanks again. :slight_smile:

High Resolution Scheduling: 1000hz tick rate for precise low jitter task scheduling.
Zen, cachyos, lqx, xanmod, rt kernels have this as default.

1 Like

No kernel changes and I’m using the vfio kernel now, but the different kernels I have tried all seem to work just fine.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.