Triode wrote: > Hum - mainly because I thought that the use case would be machine > specific.. Also didn't want to put the option into a web gui and then > need to explain it.
Does no harm it being "on by default" on a single core, like a Pi or BB. Big win on a quad or 8 core desktop. Can't "publish" numbers right now, (stuff packed up, I'm moving), but I suspect when I show that on a multi-core, that by using soxr OpenMP multi-thread, wall socket power consumption is lowered, (by virtue of spending more time at lower core speeds by speading the resample load across multiple cores), that most sensible people will want it enabled regardless of what a CPU% or load avg number says, if they have a multi-core CPU, and wish to re/upsample, whatever arch they are running it on. Triode wrote: > > I can include it - how about making it the 't' flag for 'threaded', 'z' > doesn't really make sense unless you know the underlying code? Up to you. Whatever you want to do. For the squeezelite rpms that I've made this morning that are now part of the "Fedora LMS" repo, I have gone back to enabling it by default. I really don't think it should be a compile time option, hence why I posted the patch, which at least if you want it disabled by default, means that it is compiled-in with resampling option and can be enabled via a flag. PS. I have added a note to "squeezelite -h" output on my rpm packaged version, to say that it isn't the "official" upstream version. I didn't want to go as far as calling it a "fork"..... ;) ------------------------------------------------------------------------ JackOfAll's Profile: http://forums.slimdevices.com/member.php?userid=3069 View this thread: http://forums.slimdevices.com/showthread.php?t=97046 _______________________________________________ unix mailing list [email protected] http://lists.slimdevices.com/mailman/listinfo/unix
