WAV vs FLAC revisited

[FrantzM]

Hi

I wouldn't want to be too strong on the semantics ... The simple fact that the formats (I use the term loosely) are different does induce people into believing that there are differences .. How do we remove this expectation bias from the results?? There is a solution but it is so loathed by us audiophiles, that its mere utterance would derail a thread .. I will not , thus mention it but it seems that such protocol has not shown any differences between FLAC and WAV ...

 

[amirm]

Hi

I wouldn't want to be too strong on the semantics ... The simple fact that the formats (I use the term loosely) are different does induce people into believing that there are differences .. How do we remove this expectation bias from the results?? There is a solution but it is so loathed by us audiophiles, that its mere utterance would derail a thread .. I will not , thus mention it but it seems that such protocol has not shown any differences between FLAC and WAV ...

Please don't mention it .

Mitch, the solution I am thinking about is this: we measure jitter/ground noise (can't do with your hardware unfortunately) and then make the problem 100X worse and see if it makes a difference. We can run a program that exercises the system in various ways, including one that decodes FLAC, while playing a WAV file to see if there is any impact on the output. If there is not, then we know this theory is kind of busted. I say kind of because every PC is different so the the situation may not apply to others but at least we will have something to talk about .

 

[Vincent Kars]

It looks like Gordon Rankin did this type of measurements.
Whithout finding any differences
http://www.audioasylum.com/forums/pcaudio/messages/9/99274.html

 

[Phelonious Ponk]

Hi

I wouldn't want to be too strong on the semantics ... The simple fact that the formats (I use the term loosely) are different does induce people into believing that there are differences .. How do we remove this expectation bias from the results?? There is a solution but it is so loathed by us audiophiles, that its mere utterance would derail a thread .. I will not , thus mention it but it seems that such protocol has not shown any differences between FLAC and WAV ...

I won't say it either, but I will say that it is the answer to this, and so many questions. But it is, evidently, more fun to keep the questions open.

Tim

 

[Vincent Kars]

An Linn engineer about the DS:

1. If we measure the power rail that feeds the main processor in the DS we can clearly see identifiable disturbance patterns due to audio decoding and network activity. These patterns do look different for WAV and FLAC - WAV shows more clearly defined peaks due to regular network activity and processing, while FLAC shows more broadband disturbance due to increased (but more random) processor activity.

2. If we measure the power rails that feed the audio clock and the DAC we see no evidence of any processor related disturbances. There is no measurable difference (down to a noise floor measured in micro-volts) between FLAC and WAV in any of the audio power rails.

3. Highly accurate measurements of clock jitter and audio distortion/noise also show no difference between WAV and FLAC.

The extensive filtering, multi-layered regulation, and careful circuit layout in the DS ensure that there is in excess of 60dB of attenuation across the audio band between the main digital supply, and the supplies that feed the DAC and the audio clock. Further, the audio components themselves add an additional degree of attenuation between their power supply and their output. Direct and indirect measurements confirm that there is no detectable interaction between processor load and audio performance.

http://forums.linn.co.uk/bb/archive/index.php/thread-7711-6.html

 

[Phelonious Ponk]

This from a server - a machine with the computer and the DAC in the same box? Surely the old, the original solution to all of this processing noise -- the well-isolated external DAC -- is as effective if not as elegant?

Tim

 

[Vincent Kars]

I’m not familiar with the Linn hardware.
But I do know that e.g. the Bryston BPD-1 is a one box solution but they did ‘everything’ to isolate the analog part from the computer part.
Isolating is imho not necessarily a 2 box solution e.g. 2 boxes with a galvanic connection might work as good or bad as a 1 box solution.

I get that, and I'm always a bit surprised when I get what you're saying, Vincent. But surely it's easier to accomplish galvanic isolation in a 2-box solution?

Tim

 

[rbbert]

There are many programs available (all or nearly all free) to interconvert FLAC and WAV or AIFF. I have been under the impression that the converting algorithms are all identical within platforms, i.e. a UNIX utility for Mac and Linux and a different one for Windows. Can someone here confirm this or otherwise give a correct explanation?

 

[fas42]

Yes, provided that the writer of the algorithm has not decided to be "clever", and add some sort of filtering or other fiddling to the processing, the end result should always be the same. In other words, irrespective of platform and creator of the utility, if you have a certain FLAC file and convert to WAV, then the actual musical data should always be 100% identical to the result of another conversion.

Frank

 

[rbbert]

Yes, I have XLD and xACT for Mac, and dbPoweramp, FLAC front-end and Trader's Little Helper for Windows. They all will convert a WAV or AIFF file to FLAC and back again, with the checksums agreeing between starting and ending files. But, they all convert files at very different speeds, XLD being much faster than xACT on my MAC, and dbPoweramp being faster than TLH or FLAC front-end on my PC.

 

[fas42]

The speed of conversion, all else being equal, is all due to the "cleverness" of the programmer! He could be using a different algorithm, a subtle but "clever" variation of a standard algorithm, or a completely standard algorithm which has been "race-tuned" to go as fast as possible! In programming, there are many, many ways to skin a cat :b ...

Frank

 

[microstrip]

I am using DBpoweramp to prepare a beta JRiver Music Server with some of my CDs. When ripping a particular CD to FLAC I get the information:

AccurateRip Verified Using Pressing Offset +703, Confidence 4 [new crc ... ] in all the tracks.

Looking at the artwork and label of the CD is seems poor quality. Is this Pressing Offset an indication it was a non legal version?

 

[Vincent Kars]

Try: http://forum.dbpoweramp.com/showthr...rified-Using-Pressing-Offset-6-(-possible-bug)

 

[Thomas.Dennehy]

With disk storage space down around US$0.05/GB (that's five cents), it is really necessary to store music in compressed format these days? FLAC for download, WAV for playback.


TGD

 

[amirm]

With disk storage space down around US$0.05/GB (that's five cents), it is really necessary to store music in compressed format these days? FLAC for download, WAV for playback.

TGD

It is because uncompressed .wav files don't have universally supported metadata tags. So to the extent you want to have a rich experience in your media player with your files, some sort of lossless format is necessary. I use WMA Lossless for that reason.

 

[amirm]

I should add that to the extent you want to use a solid-state-disk (SSD) to store your files, then space saving does translate into real dollar savings as SSD is a lot more expensive than hard disk.

 

[microstrip]

I should add that to the extent you want to use a solid-state-disk (SSD) to store your files, then space saving does translate into real dollar savings as SSD is a lot more expensive than hard disk.

A relevant aspect. Do you think there is real quality improvement using a SSD hard disk to play your files using a state of the art DAC? As I use DBpoweramp to rip CDs I am faced with the choice of FLAC compressed (- 350M per CD) versus uncompressed (~660M ). Is there any real advantage in using uncompressed? If so, is there any software that at a later stage can systematically uncompress all compressed files?

Sorry for so many questions, but I prefer to ask than finding it after I have ripped 100 CDs the wrong way .

 

[RBFC]

A relevant aspect. Do you think there is real quality improvement using a SSD hard disk to play your files using a state of the art DAC? As I use DBpoweramp to rip CDs I am faced with the choice of FLAC compressed (- 350M per CD) versus uncompressed (~660M ). Is there any real advantage in using uncompressed? If so, is there any software that at a later stage can systematically uncompress all compressed files?

Sorry for so many questions, but I prefer to ask than finding it after I have ripped 100 CDs the wrong way .

The theory is that the mechanical portion of the standard hard drive injects noise into the signal, depending on what connection you choose between the hard drive and the playback gear. I'm using a FireWire 800 cable with the power lead cut inside, so that there is no power connection between the drive and the playback computer (which has internal SSD). The hard drive is plugged into an AC outlet for power.

Lee

 

[Vincent Kars]

Is there any real advantage in using uncompressed? If so, is there any software that at a later stage can systematically uncompress all compressed files?

Lossless uncompressed vs lossless compressed (WAV vs FLAC most of the time) is very controversial.
Some recent examples:
http://www.audioasylum.com/cgi/vt.mpl?f=pcaudio&m=101899
http://www.audioasylum.com/cgi/vt.mpl?f=pcaudio&m=101549

I do think that if there is an audible difference it indicates a system very sensitive to software induced jitter so a system error.
http://thewelltemperedcomputer.com/KB/WAV-FLAC.htm

Technically it very easy to convert from one format to another.
dbPoweramp comes with an excellent format converter.
As lossless=lossless compressed or not, you can always convert bit identical to another lossless format.
You can even convert MP3 to WAV!

One of the problems is the lack of a standard for tagging WAV.
Some developers solve this by writing ID3 tags in a info chunk.
Have not tested this but there is change that if you use dbPoweramp to rip and JRiver for playback the tagging will work.
But most of the time it won’t

Best you can do is rip a couple CDs to FLAC
Convert them to WAV and compare them in a listening test.
Do it unsighted otherwise you will almost certainly hear a difference.

SSD
What you can try:
- Disable the page file
- Using JRiver to play from local HD with and without memory playback on.
- Play from a USB thumb drive or a SD card
- Play from a NAS or another PC
Might give you an indication what different hardware does on your system

 

[rbbert]

What if you do the test "sighted" and you hear no difference? Too much "expectation bias", or there really is no difference? I've done a few sighted tests in my own system, comparing different CD-R media, and some different cables, and have generally been unable to hear a difference.