Windows automatically boosts the prio of threads playing audio, basically treats them as realtime in its prio hierarchy. What I expect GP is talking about is minimum latency achievable, i.e. that with focused tuning, you can outperform Windows.
Regular users of not so tuned apps probably experience this differently because Windows tries to do the tuning automatically and Linux doesn't.
The Windows scheduler has other issues, like random prio boosting to escape from prio inversion.
Regular users of not so tuned apps probably experience this differently because Windows tries to do the tuning automatically and Linux doesn't.
The Windows scheduler has other issues, like random prio boosting to escape from prio inversion.