I don’t think you can stream WAV from Qobuz, or any on-line streaming supplier for that matter. All my CD rips are WAV on my Core, as are almost all digital downloads. I’ve always found WAV to sound better - more open and natural. Though my best digital sources are SACD (DSD) and Bluray (PCM). I have moved heavily to digital media outputted to the ND555 after firmware gate as well, so I do go along with the view that reducing CPU load is the way to go here.
4 Likes
Agreed Mike.
That link between local WAV ríps and reduced CPU workload surely is relevant here.
It compounds the theory that the busier CPU workload has created more noise (EMI) that has leaked into the DSP/ DAC’s sensitive clocking. This has necessitated a tightening of the DSP filters to compensate resulting in the less ‘breathing’ organic sound that we both previously preferred.
I make no claims to fully understand the science here. There are many forum members like Simon who will do so far better. But I am confident in how these changes have manifested in sound characteristic changes.
3 Likes
I recall having a discussion with Steve Sells at the Australian launch of Statements in 2014, that wav was better than FLAC, he advised that FLAC required extra processing for playback, that wav didn’t require.
I heeded his advice, so all in my CORE are wav files.
5 Likes
With the later streamers it allegedly makes no sonic difference whether files are in WAV or FLAC, as the streamers have much greater processing power than the originals such as NDS or NDX. Twelve years on I suspect Steve would give different advice. Let’s ask him!! @110dB All my music is kept on the NAS in FLAC, but I still use Asset to transcode to WAV on the fly. Old habits die hard.
It’s worth remembering that WAV Naim rips don’t embed metadata, which will give problems when the Core eventually dies and the files are moved to a different server.
What does this mean?
1 Like
Asset turns the FLAC files into WAV as they are played. It’s called transcoding because it doesn’t convert the files themselves permanently, just as they are played, hence on the fly. Sorry for the jargon.
I’ve been thinking about adding a tube headphone amp to my system and it’d be better if I could only have one output active so I wouldn’t have to manually turn off my active speakers (Genelec G Three) that I’ve got hooked up to the XLR output every time I’d want to use the amp. Of course then it’d also be better if you could set the RCA output level as fixed or have separate volume levels for it and the XLR output which also isn’t possible at the moment.
I’ll probably look for some other solution if I decide get the tube amp.
You said
All my music is kept on the NAS in FLAC, but I still use Asset to transcode to FLAC on the fly.
I do not understand the benefit of transcoding from FLAC to FLAC?
I think he is transcoding FLAC to WAV using the Asset software running on his file server NAS. Transcoding is an another process running somewhere in the system. By moving that process off the streamer and into the NAS he is lightening the load on the streamer, so making a sound quality improvement.
We tested this with 272 and WAV was preferable.
When we upgraded to NDX2 as transport into nDAC/555 we repeated the test and could hear no difference.
FLAC decoding is designed to generate minimal power loading and negligible CPU use (the “expensive” part of the cycle is encoding, to generate the compressed content in the FLAC file). On the modern streamers you wouldn’t expect to hear any difference, especially with cache in play.
We no longer transcode on Asset, but that’s only since Christmas when we updated our DSM. Had forgotten we had it turned on! Vast majority of listening is via Tidal Connect.
Just for fun, and this is not really a repeatable scientific experiment 
I looked at our NDX2 playing the same track in both formats, with Asset transcoding….
These are figures during playback, and I would presume the NDX2 is playing both exclusively from cache after the first second or two…
FLAC
WAV
From this entirely completely unscientific experiment I did observe (over a huge 3 samples) that WAV results in the streamer reporting a lower SDK load, but higher total CPU,
I then tried when watching telly, NDX2 taking Sky Q as input (48kHz) and sending out S/PDIF
No difference really,
As a result of this highly scientific investigation I am no longer concerned about which input of our NDX2 is in use, or which format the stream is!!!
6 Likes
I have tested WAV and FLAC on my NDX2 (with a separate DAC) and the difference is very clear: the WAV files are so much better. Tested my wife blind between tidal streaming, local files I bought as WAV and local files I bought as FLAC and she too found repeatedly (37 years of marriage makes you patient) the WAV files much better than anything else. IMHO, it’s not even close and I would not bother with FLAC. But to each their own.
6 Likes
I disagree. My ears hear a big difference and so does my wife (using repeated blind testing with ps-random order). A very clear difference: the WAV files sound much better. This is of course does not invalidate your test. I It just shows that this is not the right test for the cause of the perceptual difference.
2 Likes
You’re quite within your rights to disagree, but our experience doesn’t mirror yours.
Funny how two groups of people can use the same device and hear different things.
But without conflicting opinions this forum would be no fun.
I have to say that today doing the highly scientific test above, the track was “Young and Stupid” by Belle & Sebastian. Only played a couple mins in each format, but this is a track I know well and I could hear no difference. But then perhaps I was enjoying the music too much to be analytical, that is after all the point of having a decent mid-range separates system 
I’ve turned transcoding in Asset to WAV back on - I wonder if SWMBO will ask what’s happened to the hifi, although I expect she won’t notice a thing.
3 Likes
Indeed, after all some people like non-Naim products (just a joke!). This is why I used my wife as blind tester. I only use acoustic instruments (typically classical music) as tests for obvious reasons. Short bursts of music (to avoid focus on the musical idea and because of habituation issues), ps-random order of music and which is which between FLAC and WAV. She too was also so quick and clear in her judgement. Maybe it’s the air in Western NY :). Or, the most likely explanation: we are all different. My point was simply to entice people to try on their own. I buy my classical music files from Presto which allows you to download both the FLAC and WAV files, which makes for easy testing.
It’s surely system dependant, setup might make a big difference. That’s why, I suspect, you don’t feel such difference. Also the kind of music you play, does have a big impact as well. Classical and jazz music will underline and amplify these differences, while pop and rock music might not.
The wife test never lies!
3 Likes
This could be it. Too much death metal 
Seriously though, when testing things like this we use a range of music including acoustic and vocal.
But any change has to sound good with wall of noise recordings too. In many ways those put far more stress on a system in terms of control, speed, separation, dynamics.
3 Likes
I mainly listen to jazz fusion. Hahah too much death metal indeed!
Don’t have WAV, but same experience here with AIFF vs. FLAC (and ALAC).
3 Likes
Ok…i think I’m ready for the update…!!!
I’ve had enough of the weather so if you guys in Beta can speed things up with Naim I’d really appreciate it 
FYI Dusty’s tongue is firmly in cheek!!!
2 Likes
Better to get it right than get it fast, Dusty.