WAV 192 Khz stops while buffering FLAC 192 Khz doesn't

When you stream WAV the bit rate is higher than when you are streaming FLAC. Sounds like your LAN can't cope with the higher transfer rate.

Yes, the WAV file is bigger.. Typically by about 30% but varies on the amount of compression and the type of audio being compressed. A wired LAN connection should have no issue, however considerations

A) NAS I/O (or where your media server) throughput issues

B) faulty Ethernet patch lead in the media path.

C) using a software controlled switch such as on an old style bundled ISP bundled ISP router switch.

Just what i thought Simon. It's my ethernet, lately i've had some trouble with my internet connection.

WAV 96 is just fine.

I never thought it but WAV sounds better then FLAC.

if the music is getting from your NAS to your streamer via your router, then a quick thing to try is putting a simple switch (eg a Netgear GS105) in place where all the LAN connections to your router are and connect the switch back to the router with a single cable. You can easily buy from the usual online places and they cost about £25. This removes all the uncertain handling in the LAN side of the router which often isn't well suited to streaming large amounts of data.

best

David

Looks like the OP is using mains over ethernet if I interpret the post correctly.  Most likely where the problem lies. 

Simon & nudgerwiliams have it ......

I would add to install a switch between NAS & Naim & connect all with Ethernet,  and a branch from switch to router.  Switch costs £20 or less.

It really is the only 100% way - avoid Ethernet over mains & wireless - zimples

I know that's the solution. But i wondered why there isn't any buffering with FLAC files.

Also I transcode some iTunes AAC files into WAV. AAC streaming at say 300 kb/s. If it is transcode to WAV it is 1411 kb/s. Where are these extra bits come from?

The server will just pad out the file with extra zeros.

Hi Pallie, interesting question. AAC and WAV-PCM do not use same method for encoding the data. AAC is a lossy type which uses psychoacoustic masking and other techniques to discards some of the data and information but in a way that the brain hears the majority of what  was originally there in a transparent as possible way. WAC-PCM simply is a sample stream of the actual audio with nothing taken away.

Now the AAC signal needs to be decoded so we can play it, that is it will be reconstructed into PCM or analogue. In doing so some of the original content has been discarded. However to capture this new audio in PCM it would require this reconstructed audio to be converted into a stream of samples. For PCM the bandwidth is the same no matter how 'complex' the audio so for PCM it makes no difference whether details in the audio have been discarded, a sample is a sample.

Therefore AAC converted to WAV will have the higher bit rate of the original PCM 1411 kbps) , but contain the information of the AAC (lossy encoded 320  kbps). This phenomenon is called entropy.

Simon

This is a far more interesting example of entropy than my thermodynamics lecturer ever came up with back in the last century.......

indeed

Perhaps I should have said this phenomenon is an example of what is called entropy within Information Theory. (one of my pet subjects at Uni and guided the early part of my career developing standards for network protocols .. those were the days)

Simon