PasTim;630098 Wrote: > A basic scan takes several minutes for my ~14,000 tracks. Custom scan > takes a further 1 to 2 hours to load non-standard tags. If SB server > scans every time it starts up I'll miss my own funeral waiting for the > music to start. I am one of those strange people who turns their PC > off at night, and doesn't run SB all the time. > > > Well I thought that. I tested the network effect by downloading some > large files (more music) over the internet at the same time as playing > 24/96 flacs via flac native format. Total network utilisation was > steady at around 8-9% and I had no problems (although much of the > traffic was in the opposite direction from the PC's perspective - I may > try some other way of stressing the network). > > When I use PCM, network utilisation varies quite a bit, from 1-2% to 8 > or 9%. I can sometimes play 24/96 for several minutes with no issue, > and then it stops for rebuffering. The PC CPU is hardly running at > more than 3%. > > A major difficulty I have is that I don't have the knowledge to really > work out what's going on, and I have a very workable method of playing > music (i.e. via flac). PCM may be better - I'm not really sure since > I've spent too much time worrying that it's going to stop any minute. > > > So I've run 7.5.3, 7.5.4 and 7.6 (the latter only briefly - it may have > been better but the other issues rule it out). I guess I'll stay on > 7.5.3 for a while. > > I still cannot work out why a smaller buffer would improve quality. I > would, naively, have though a large buffer would allow the system a > chance to run more smoothly and reliably. Have you determined a > reason? > > Do you think the buffer size will have an impact when using flac format > over the network?
I played quite intensive with 24/96 over the weekend. (I downloaded 24/96 Oscar Peterson - Night Train from HD Tracks - I love it.) I experience that random rebuffering on 24/96 "PCM" streams too. This rebuffering also happens if I quickly switch several tracks. Everything becomes kind of instable. I do have a note about it on the blog since the very beginning of my project. Probably I ran into that problem earlier. I think it's time to address the issue. I'm well aware about the standard Logitech answer: "Stream flacs!!" Looks like a compromise/workaround to hide certain shortcomings or flaws. On a first glance this problem shouldn't have anything to do with the toolbox itself. If the buffer is running at 20000 there shouldn't be any mod impacting the actual throughput. The toolbox actually should improve the throughput. >From a networking and processor load perspective there is plenty of headroom left - that you confirm too. The Touch input-streaming-buffer ( which has nothing to do with my buffer mod) is affected most when running 24/96 PCM streams instead of flac streams. There must be a problem in that area. Afaik Logitech realized some streaming related inefficiencies, which are supposed to be improved in 7.6.. I'm not sure though if above issue would be affected with those 7.6. improvements. If 7.6 wouldn't be that unstable (see beta forum) I'd give it a try by myself. Does anybody have a clue if 7.6. will ever come to live? Nothing seems to be moving since a year or so. If anybody is streaming 24/96 as PCM without any problems, please let us know about your config. -- soundcheck ::: ' Touch Toolbox 2.0' (http://soundcheck-audio.blogspot.com/2011/01/soundchecks-squeezebox-touch-toolbox-20.html) ::: by soundcheck ------------------------------------------------------------------------ soundcheck's Profile: http://forums.slimdevices.com/member.php?userid=34383 View this thread: http://forums.slimdevices.com/showthread.php?t=84742 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
