Hello! I was wondering how accurate the clock of a Biosemi Active Two amplifier is. Specifically, since the XDF format, which is the default file format when recording with Lab Streaming Layer (LSL), also records time stamps with every sample, I was wondering which clock I could trust more. The one from the amplifier (which means I'd use the inverse of the sampling frequency, e.g. 1 / 512 s), or the one from LSL (which is based on the computer high-performance clock and is based on the recorded data)? Usually, the difference is pretty negligible, but I'm still curious which clock is more accurate.
For reference, there is an ongoing discussion at https://github.com/cbrnr/mnelab/issues/436 (where I've also linked a relevant prior discussion https://github.com/cbrnr/mnelab/issues/385).