Lots of opinion here about wall power, headless, Windows, OSx, and the
like, but I've seen nothing that actually compares the performance of
LMS on different OS platforms with say 100K tracks. I thought that's
what the OP asked for?
But what part of LMS' performance do you want to compare? Scan times?
Web page rendering? Menu rendering on device?
Anyway, I don't think there have ever been thorough comparisons. I've
done some benchmarking when working on the scanner's performance. But
that's only one aspect, and unimportant for many anyway, as it often
happens in idle times.
> I was wondering if any one particular OS gave significantly better
> results.
I think the opposite can be said: on comparable hardware Windows is the
slowest to run LMS overall. Is it significant? Are 20% slower scan time
significantly slower? In numbers: yes. But to you? It really depends on
your usage pattern.
If you're still actively ripping and tagging music on an almost daily
basis, then slower scan times can be important to you.
If you're controlling your system using a smartphone app which caches
all data locally, then LMS query performance might be irrelevant.
If you use the web UI all day long from your PC, then page rendering
might be important. Not so much for the previous user with his smartphone.
If you have 250k tracks, then a Pi might not be suitable, while it's a
perfect solution for the 15k tracks collection which it basically can
handle in memory.
If you'd have to learn Linux just to run LMS, would it be worth a 20%
faster scan time?
That said I'd second what others have said before: if you want a lower
power solution than your PC, then go with a Pi3 or a Linux based NAS.
If you'd like to recycle an old PC, then install Vortexbox on it.
If you have a Mac Mini sitting around, use it.
--
Michael
_______________________________________________
Squeezecenter mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/squeezecenter
