Why do so many quality DACs sound better with WAV than FLAC, AIFF, and others?
- Thread starter caesar
- Start date
Actually, it could to some extent based on EMC, but this would be a good test to do to check it with perceived SQ.
My point was that even if you are buying WAV/AIFF files, they have almost certainly been through a FLAC stage at some point. Obviously that doesn't apply to native DSD recordings, but a fair number of files sold in DSD format are PCM sourced and have probably been stored as FLAC at some point.
Not necessarily.
Not the ones I look for: they are either recorded directly to DSD or sourced from analogue Master Tapes.
Here again, even those which have been converted from PCM to DSD (some of which make their way to SACD) may have no intermediate conversion to with FLAC.
Both of us are posting true statements, so I'm not sure what the problem is. In any case, my first post on the topic was in response to a statement that the poster buys WAV files from HDTracks, and my point was that he is really buying FLAC files converted to WAV, and that is also true for most other retailers selling WAV files.
Nope.
You may need to re-check your post #26 above.
Again, check your post #26 above where you quote me. I have never bought from HDTracks nor do I ever intend to, so who are you referring to?
Sorry, re-check the title of the thread; it's not about DSD
And yes, I was mistaken; I was intending to reply to marslo who mentioned buying WAV files from HDTracks.
Where do you think the difference between WAV and FLAC comes from? In both cases, the PCM data sent to the DAC is identical, it seems to me any difference could only come from extraneous (non-data) information in the signal stream, which the Regen is supposed to decrease or minimize?
I have certainly found a huge difference, as reported by others, between flac and wav.
When I was dac-hunting last year, an audiophile buddy and I were listening to the Bricasti, Ayre and other units on a regular basis using flac files.
At one point, I don't remember why, I played a WAV version of something that immediately shocked us.
After a few repeats, we were totally convinced that WAV sounded much better than flac and all future tests/comparisons were done with wav as the default.
My system now: totalDAC d1-dual, pass aleph P, ayon monoblocks, all balanced. digital source is a netgear NAS running squeezebox LMS as server, streaming to a sonore client running squeezelite over a home gigabit LAN with a 24 port netgear router connecting a ridiculous number of IP connected devices around the house.
So one possibility is that the 24 port router and all the IP traffic in the house could be adding jitter and maybe a better regeneration of the data in the ethernet stream would help.........I don't know, but as an engineer I crave an answer........and as an fussy audio person, I know flac/wav makes a difference that i don't want to live without once i've heard it.
Yes, bits are bits and should be the same going into the dac so I have to assume jitter and/or noise is the cause.
To test this, I took a WAV file from a CD, converted it to flac, then converted the flac back to wav, then that wav again to flac.
In all cases, the WAV files were identical and sounded the same.
In all cases, the WAV files sounded better than the flac files.
Clearly flac encoding/decoding does not change the bits (we all knew that, right?) but somewhere in the NAS steaming to client to usb to dac, wav files are being presented better somehow to my dac.
Science and engineering are also partly an art.
I remember a professor in grad school who told me (40 years ago!) that I couldn't possibly hear differences between DACs as the CD was effectively perfect because 16bits had an snr of X dB, case closed.
When I was dac-hunting last year, an audiophile buddy and I were listening to the Bricasti, Ayre and other units on a regular basis using flac files.
At one point, I don't remember why, I played a WAV version of something that immediately shocked us.
After a few repeats, we were totally convinced that WAV sounded much better than flac and all future tests/comparisons were done with wav as the default.
My system now: totalDAC d1-dual, pass aleph P, ayon monoblocks, all balanced. digital source is a netgear NAS running squeezebox LMS as server, streaming to a sonore client running squeezelite over a home gigabit LAN with a 24 port netgear router connecting a ridiculous number of IP connected devices around the house.
So one possibility is that the 24 port router and all the IP traffic in the house could be adding jitter and maybe a better regeneration of the data in the ethernet stream would help.........I don't know, but as an engineer I crave an answer........and as an fussy audio person, I know flac/wav makes a difference that i don't want to live without once i've heard it.
Yes, bits are bits and should be the same going into the dac so I have to assume jitter and/or noise is the cause.
To test this, I took a WAV file from a CD, converted it to flac, then converted the flac back to wav, then that wav again to flac.
In all cases, the WAV files were identical and sounded the same.
In all cases, the WAV files sounded better than the flac files.
Clearly flac encoding/decoding does not change the bits (we all knew that, right?) but somewhere in the NAS steaming to client to usb to dac, wav files are being presented better somehow to my dac.
Science and engineering are also partly an art.
I remember a professor in grad school who told me (40 years ago!) that I couldn't possibly hear differences between DACs as the CD was effectively perfect because 16bits had an snr of X dB, case closed.
cat6man, thanks for posting your experiences.
I have a sneaking suspicion your culprit is LMS. It is among the most resource hogging, bloated server packages out there.
It has many know issues with the way it decodes FLAC, ALAC, etc.
I use MiniMServer which the lightest, most agile package you are going to find. And it is free.
The Sonore, btw, is simply superb. Good choice.
The above about LMS is just my theory, I certainly can't say for sure.
OS X Yosemite on a brand new Mac Mini w/ solid state drive. I use it as a NAS with G-Tech drives attached which house a 12 TB library.
It has become obvious (if it wasn't at first ) that this thread is misnamed, since the DAC probably has nothing to do with any perceived differences. I also wonder if anyone using a newer dedicated music player / streamer (e.g., Bryston, Aurender, Lumin, etc.) notices differences between the various file formats.
) that this thread is misnamed, since the DAC probably has nothing to do with any perceived differences. I also wonder if anyone using a newer dedicated music player / streamer (e.g., Bryston, Aurender, Lumin, etc.) notices differences between the various file formats.
It has become obvious (if it wasn't at first
Yes, exactly, it should had a title asking why there may differences between SERVERS with FLAC/WAV etc.
FYI, I have a BDP-2 on the way. According to Bryston they can her no differences what so ever between FLAC and WAV.
Back in 2009, asked the same question they said they "thought" WAV sounded a bit better, but were not sure.
It has become obvious (if it wasn't at first
You correct. But unlike yourself, Andre, and Amir, many of us have impoverished experiences when it comes to digital audio. Thank you for helping clarify.
"impoverished"?
Excellent word! It is all good. One of the pit falls of digital audio is to over think it. If I listened to every suggestion on various forums I would spend 24 hours a day
implementing "tweaks" and testing files, cables, vibration devices, power supplies, software, formats etc instead of listening to music.
Don't sweat it.
As I mentioned above, it is very obvious with Naim in my friend's system. But they are trying to sell a cd ripper/ upnpn server product, and get the customer to shell more $ and get locked into Naim's digital architecture.
When my friend converted FLAC files to WAV using some trick on his minimserver on his nas, it was a night and day difference.
"impoverished"?
One of the pit falls of digital audio is to over think it. If I listened to every suggestion on various forums I would spend 24 hours a day
implementing "tweaks" and testing files, cables, vibration devices, power supplies, software, formats etc instead of listening to music.
Don't sweat it.
Experiential poverty is what leads to so many wars in this hobby.
Thanks, I hear you. It's great advice for one's sanity.
Here is how the architecture works in a networked system. You can have FLAC/ALAC etc decoded at your renderer..the component that received the files and feeds your DAC, or you can have them decoded at the SERVER side. One way or the other, the files must be decoded. MiniMserver simply decodes FLAC to WAV if you tell it to before it sends the file through the network, no trick here. And that is what I do btw.
I have a 12TB library. Converting everything to WAV or AIFF would be insane.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |