Open Access

Table 4

Analysis of computational performance and local self-monitoring latency for different operating systems and configurations. All devices were running in the same session, connected to the same local network, using ovbox version 0.21.2, with a Focusrite Scarlett solo USB audio interface and one audio input channel. The column “CPU” denotes the CPU usage as reported by the audio server, which is the average time used to process one audio block divided by the duration of one audio block. Overruns and underruns of the audio device (“xruns”) were counted over a period of 900 s. The monitoring round-trip time (“RTT”) was measured from analogue input to analogue output.

Hardware Operating system 2 ms block size
1 ms block size
CPU
xruns
mon. RTT
CPU
xruns
mon. RTT
% 1/min ms % 1/min ms
Raspberry Pi 3B+ (BCM2837B0) Raspbian 10 (32 Bit) 5.10.103-v7l+ 62 5.3 9.25 100 >9000 5.4
Raspberry Pi 4B (BCM2711) Raspbian 10 (32 Bit) 5.10.103-v7l+ 24 <0.05 9.125 34 1 5.125
Linux desktop (Intel i5-8500) Ubuntu 20.04 5.15.0-97-lowlatency 8.2 0 6.04 15 0 3.9
mac mini M1 2020 (Apple M1) macOS Monterey 12.7.3 4.6 <0.05 9.04 5.7 <0.05 7.04

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.