Any sound quality difference between FLAC & WAV?

M

Mr Peabody

Well-known member
Joined
Nov 19, 2013
Messages
3,308
Location
St. Louis, MO, USA
I can't sseem to find credible articles comparing FLAC to WAV. Most are other discussions on other sites. So I thought if I can only get help from others, bring it here. Most everyone is already into playing files here.

My first concern is whether either sounds better. If having a choice which is best to rip a CD to? I realize FLAC compresses and decompresses where WAV remains the same size. FLAC fans claim no loss of data, where another person claims the folding and unfolding of the FLAC file could cause distortions or time issues.

So educate away. :)
 
I rip to ALAC, and buy downloads in FLAC, or DSD. They all sound great with Lumin. I think the gear used to store/play the files is more important than how the file is ripped. Obviously, I mean uncompressed.

Of course, being audio, someone else will feel the opposite. :)
 
WAV - sounds best. You can also then select files and easily make a CD of music to play in the car. Also, easily convert to another format.
FLAC/AppleLossless - both equal for second IMO.


Sent from my iPhone using Tapatalk
 
I very much prefer WAV to Flac in my setup. But it may be a system situation. Maybe my setup creates noise as it uncompresses the file. I'm guessing. I don't know the real reason. A friend sent me a couple hundred albums, some Flac, some WAV and some AIF. The WAV was best across the board. The AIF was not that bad. I might actually call it good. The FLAC was just dead sounding. Something was wrong. I tossed everything accept the WAV.
 
Kingrex said:
I very much prefer WAV to Flac in my setup. But it may be a system situation. Maybe my setup creates noise as it uncompresses the file. I'm guessing. I don't know the real reason. A friend sent me a couple hundred albums, some Flac, some WAV and some AIF. The WAV was best across the board. The AIF was not that bad. I might actually call it good. The FLAC was just dead sounding. Something was wrong. I tossed everything accept the WAV.
Click to expand...


Someone correct me if I’m wrong but you could have taken the Flac files and converted them back to Wav with no loss.
 
You know. I started another thread and asked why I couldn't upsample and save the file to stroage, then playback the high rez file later. I was told it did not provide any gain. As I learn more I sometimes question that. I gather a part of getting the 44.1 to 88.2 is the filter to role off would not be so steep and ringing could be eliminated. Resolution would not improve, but a reduction in noise is always welcome. Now I have been told if I go from 44.1 to 96, I have deviated from a simple doubling of the rate and the math becomes complex so the sound suffers. Anyhow, that is totally different from what you suggest. it makes sense to me that DBPoweramp would probably have the ability to take a FLAC file and uncompressed and convert to WAV. I don't have any of the files any more. Maybe someone has tried this and wants to chime in.

Audiophilehi said:
Someone correct me if I’m wrong but you could have taken the Flac files and converted them back to Wav with no loss.
Click to expand...
 
Audiophilehi said:
Someone correct me if I’m wrong but you could have taken the Flac files and converted them back to Wav with no loss.
Click to expand...

True. For those who believe WAV and FLAC must sound the same and heard no difference, the reason is that they are exactly the same digitally after FLAC decoding.

For those who heard a difference, the reason is that the process of the FLAC decoding needs additional CPU and memory accesses, which cause additional noise. Additional CPU usage also means slightly increased power requirements. Power is of utmost importance to audio SQ.

(Note: WAV is not good for tagging. For a non-compressed alternative consider AIFF.)
 
While it is true that Flac labels better than Wav, it is also true that we depend on the server software that we use later to read the files. A Wav file obtained and labeled with dbpoweramp, is perfectly read (in its labeling) by Mimimserver and horribly by Twonky.
 
Yes, that's one of the reasons MinimServer has been the official recommended server software for Lumin since Day 1.
 
Something else to consider; if you load the entire file or folder from a spinning HD into memory or SSD in order to play it, it uses less CPU processing to decode FLAC as part of that transfer than it does to just transfer the (larger) WAV file (it also takes less time). You can observe this by watching in Task Manager as you do it.
 
I had some friends over last year and one guy wanted to know if WAV sounded better than FLAC. So he brought over some music that was both in FLAC and WAV. There were 5 of us present and we did a blind test and found we all thought WAV sounded better. It was subtle but it sounded better. 100% of the time we preferred WAV.

I am not saying that WAV is inherently better but on this one test it was and there could be may reasons for that.
 
I used to think that data was data but it looks like there may be more factors in digital file playback that may be relevant in different systems.

Recently, I bought a Melco D100 to rip (and likely re-rip) CDs into my Melco N1ZH/60 NAS. Interestingly enough, my local Melco dealer and some other dealers in the UK have experimented and concluded that (a) the Melco D100 (not unexpected from a dealer’s perspective) but that (b) WAV rips sound better than FLAC. Due to travel commitments, I have yet to do my own comparison but my dealer has been a straight shooter thus far.

As for articles, there is an interesting write-up with some empirical data/findings on WAV vs FLAC sound quality on EnjoyThe Music.com and HiFiCritic.com.

Links below, with parts 1 & 2 of the discussion paper available for download from hificritic.com.

www.enjoythemusic.com/magazine/viewpoint/0716/Why_Do_WAV_And_FLAC_Files_Sound_Different.htm

https://www.hificritic.com/flac-wav-sound-quality-research.html

For those who may find the article difficult to digest, I have set out below the conclusion of the authors.

QUOTE: Why Lossless FLAC Compression Degrades Uncompressed WAV File Quality

1. Resolution of metadata associated art (MDA).

2. Degree of MDA compression as created in typical photo-editing software.

3. Degree of FLAC compression according to settings in dBPowerAmp.

4. CPU load during decompression and conversion of FLAC files to PCM format.

5. Allocated buffer size in playback software, music server, and/or digital to analog converter.

UNQUOTE
 
wklie said:
For those who heard a difference, the reason is that the process of the FLAC decoding needs additional CPU and memory accesses, which cause additional noise. Additional CPU usage also means slightly increased power requirements. Power is of utmost importance to audio SQ.
Click to expand...


This is EXACTLY what I heard when I compared FLAC to AIFF files and why I chose to rip my discs to AIFF.
There was background noise/haze that was unmistakable to my ears. It affected the sound in other ways besides just the background noise, as well.
 
I would use a device that ripped the disc to hard drive. I'm not a big computer guy. I also wouldn't be moving the files from one place to another so if the album info was with the WAV file that should be good. Hopefullyat least the same amount of info as if ripping to Windows Media Player.

I did see the Enjoy The Music article but couldn't see the graphs so got lost after a bit and gave up.

Thanks for everyone's input
 
nicoludio said:
As for articles, there is an interesting write-up with some empirical data/findings on WAV vs FLAC sound quality on EnjoyThe Music.com and HiFiCritic.com.
Click to expand...
That article is total nonsense. The authors tried to get it published in Stereophile and were rejected, since JA reviewed and saw that it was total nonsense. That's how it ended up in those 2 publications.
If there is a difference "heard", something is broken, equipment or more likely, test method (as in none).

cheers,

AJ
 
wklie said:
True. For those who believe WAV and FLAC must sound the same and heard no difference, the reason is that they are exactly the same digitally after FLAC decoding.

For those who heard a difference, the reason is that the process of the FLAC decoding needs additional CPU and memory accesses, which cause additional noise. Additional CPU usage also means slightly increased power requirements. Power is of utmost importance to audio SQ.

(Note: WAV is not good for tagging. For a non-compressed alternative consider AIFF.)
Click to expand...

I'm curious. If decoding FLAC requires enough CPU and memory resources to noticeably affect the sound (and BTW, have you ever looked at the CPU and memory resources that takes, using the Resource Monitor? Barely enough to register, less than copying a WAV file from one drive to memory or another drive), then why do so many people say that upsampling music before playing (to 24/352.8, or DSD 512, or whatever) and/or MQA sound better than native playback? Either of those processes uses at least an order of magnitude more CPU and memory resources than decoding FLAC..
 
You must log in or register to reply here.
Back
Top