Sound is better from uncompressed downloaded files?

[Audio_ELF]

Just to answer myself...

 

Originally AIFF used the big-endian format. With Mac OS X a variant was created using little-endian. This is technically using AIFF-C (which supports compression) but instead of true compression uses a codec known as sowt. When you use iTunes (or drag a CD to the desktop) the files created are actually AIFF-C/sowt.

 

AIFF (Audio Interchange File Format) is based on the IFF (Interchange File Format) created by Electronic Arts as part of the Amiga OS. However IFF could also be used for (and was I believed primarily designed for) images.

 

Eloise

 

[elcorso]

I'm still waiting an explanation from HD Tracks, since they don't have an alternative download on AIFF or WAV for hirez music, FLAC only.

 

Maybe on Monday?

 

Jazz At The Pawnshop at 24/88 is still intriguing me. I have (believe) all the releases from LP to SACD, an even the first CD release ripped to my computer, has a better soundstage and detail

 

Also, will be interesting to know the computer ripper and playing software opinion in this matter.

 

I read about Cookie Marenco "pedigree" and is incredible, a lot of great musicians was recorded by him.

 

I know the big analogue versus digital dilema, and a friend told me that in Vegas last CES was a lot of stands with an analogue/LP display only, but this is also because Hi End manufacturers, that doesn't has DAC products, are scared with the inevitable switch to digital, with the hirez tracks.

 

In my case the turntable and LP's have a place on my rack, but I doesn't listen to them anymore, since about 4 years. This way I can't tell you wish one is better to my ears...

 

Roch

 

[dsnyder]

Hello,

 

I should have some time today for a listening test. Here's what I have in mind, and I welcome everyone here to follow along at home using their own systems:

 

 

[*]Download three 16-bit, 44.1kHz tracks from the Blue Coast website

[*]Use dbPoweramp (or similar) to convert the tracks to FLAC

[*]Archive the three FLAC files into a single ZIP file

[*]Email the ZIP file to my Gmail account

[*]Unpack the three FLAC files from the emailed ZIP file

[*]Burn the three unpacked FLAC tracks to a CD-DA (Compact Disk Digital Audio) CD (using Mediamonkey 3.2.x...format that is playable in any CD player)

[*]RIP the CD to FLAC using dbPoweramp

[*]Burn the three FLAC files to a UDF formatted data CD-ROM

[*]Using the files on the data CD-ROM, go to step three above and repeat steps 3-9 on successive generations of CDs five times (using a total of ten blank CDs--five music/CD-DA and five data/CD-ROM format)

[*]Use dbPoweramp to convert the three FLAC files from the last CD-ROM to WAV files

[*]Perform careful A/B tests between the original WAV files that were downloaded directly from Blue Coast vs. the versions that were subject to multiple generations of conversion to FLAC, ZIP, email, CD-DA burn+rip, CD-ROM burn+copy, and convertion back to WAV

 

 

At twenty cents a piece, this will cost you a couple of dollars in blank CDs, but hopefully the results will put this debate to rest. If there is data lost in a single iteration of this process, most of us should be able to hear this after repeating the process five times. I'll let you know what I find, and I hope that you'll give it a try too. Cheers.

 

-- David

 

P.S. It may be possible to do the same exercise with 24-bit, 96kHz tracks, but you'd have to use something like Cirlinca's DVD-Audio Solo to record/rip DVD-Audio discs. In that case, you'd need blank CD-ROM discs and five DVD+/-R discs.

 

[Phil Townsend]

Cookie

The very idea that there might be a ten percent loss yearly makes me fell a bit uneasy.

Could you elaborate a bit on this subject?

 

Thank you

 

Phil Townsend

Santa Fe

 

[4est]

Gee David, I really hate to be a dick and put airs on, but instead of challenging someone like Cookie, you might want to consider why it is that you cannot hear the differences instead of claiming it is silly. I also feel that you may be missing his point. I think that he is claiming to hear differences in FLAC vs .wav, not just that it has been emailed. Besides using a sub $200 DAC, it appears as if you are running it through a home theater processor. The HT processor alone could screw things up enough to ameliorate the differences in the files. At least try running the Emu straight into the amps if possible. Whether or not you are aware of it, most AVpres digitize the analog inputs subsequently doing their own DAC before the outputs. I am sure you can connect the dots. That is even if the analog section is any good either.

 

Once again, please excuse me for seeming haughty as it is not my intention. I just thought someone ought to point this out. Furthermore, when someone whom has the credentials that Cookie has makes an (apparently bold) statement, lesser souls such as me think twice before attempting a refutation. As to the reality of his claims, I'll leave that to others with more knowledge than I.

 

[Panelhead]

One of the few in the commercial recording industry. Like all women, her hearing is better than ours.

I am getting confused. But never believed the lossless compression story. AIFF only for now.

 

George

 

[Miska]

claiming to hear differences in FLAC vs .wav

 

Now, THE question is; is it because of what is IN the file, or is it because how the file is played?

 

These are two completely different things. Because someone hears difference between WAV and FLAC doesn't mean that the difference is in the data these files store or fault in the way data is stored. The fault may be as well in the way how the file is played back, completely external to the file format itself. It may be even because of the computer hardware, instead of file format or software.

 

 

[dsnyder]

Hello,

 

Fair comments, but the gear that I have "is what it is" and all that I have availble to test with for now. This is part of the reason why I have invited all of you to perform the same listening test on your systems, most of which are better suited than mine! Cookie has asked us to listen first, so I'm taking her up on that in the best, most thorough way that I know how, and I encourage you to do the same if you have not already.

 

While my playback system is not ideal for this sort of test, I am actually using the Emotiva UMC-1 as an analog pre. You are correct that by default the UMC-1 runs signals from analog inputs through an analog to digital converter, applies EQ and crossover in the digital domain, and then converts back to analog. However, I _never_ use it that way when listening to music. I'm running the analog outputs of the Creative E-MU 0404 USB into the UMC-1 with it set in "Direct" mode. This does no A-to-D conversion, EQ, or processing (not even front/sub crossover).

 

I wish that I had a better system for performing this listening test, but that's why I'm running five itterations--to magnify the impact of any generation loss to a level that will make it audible even on my humble playback system.

 

-- David

 

 

[dsnyder]

Hello Cookie,

 

Quote: I've reread your response a few times, but just to be clear, did you do a listening test? Or are all of these data viewing comparisons?

 

All of the tests that I described in my first message were data tests...no listening involved.

 

Since then, I have done my best to contrive a listening test that is thorough enough to expose any data loss and its impact on sonics in my own system, such as it is. This test involves multiple generations of file conversion, recording, and ripping to optical media of different formats, and email. Hopefully others will join along and report their results too.

 

If our computer systems can not reliably transfer, store, archive, and convert PCM (and DSD) data, we should discover why and address those problems before we trust them as components in our music playback systems. Cheers.

 

-- David

 

 

[aenima]

Hi Cookie,

 

Having read all the stuff you posted here, I still can't believe that you are a sound engineer as you claim. Wow, I bet you have learnt little about the basic science behind digital audio and digital information storage, let alone hardware knowledge related to this. I will not embark any more on the "FLAC vs WAV SQ" topic because I can see you don't know what you are talking about. 10% data loss per year? C'mmon, don't you know this is the most well-known troll over the internet audio forums? Oh if you really dig into that you will find some more funny things like "digital dust" or whatever. Yeah, take some time and amuse yourself. If I were an executive at IBM I would fire your "IBM tech guy" friend immediately for making such an utterly arrogant statement. I don't care who you've been working with on your resume because that doesn't prove that you really understand "science". Last but not the least, ever heard of the word "conflict of interest"? Maybe you shouldn't have told people that you work for a record company in the first place. Nyquist and Shannon are crying.

 

[bdiament]

Folks who are familiar with Soundkeeper and comments I have made here and in other fora, know that we don't release anything in any data-reduced format, preferring to stay with the raw PCM we use for recording and mastering.

 

I do believe listening tests are key - particularly since most of the stuff we talk about (both hardware and software) is designed for no other purpose than to be listened to.

 

What I find interesting is that many folks on Internet audio fora, seem to believe their personal experience on their own systems (even if these include relatively inexpensive computer "soundcards", untreated rooms and questionable setup) will reflect what is possible (or not possible) for someone with extensive listening experience and some top notch gear to hear. In other words, "I don't hear it on my $200 soundcard and compact box speakers so anyone that claims to hear it on a $5000 converter with $5000 speakers in a fully treated room - even if listening is what they've been doing all day for the past several decades - is being 'silly'."

 

I'm always wary of general pronouncements of Universal Truth (add reverb under those words ;-}) in place of a simple "I tried it and I don't hear it"... or "I tried it and I do hear it".

 

What I've found to be very informative in addition to listening tests is the null test. For those that may not be familiar with the term, in a null test, two files are synchronized - to the sample. (Perfect sync is critical because in its absence, it is very easy to get false results.) The polarity of one file is inverted and the output of the two files is summed (i.e. combined).

 

When this is done, the output is the difference between the two files. Anything the two files have in common will be cancelled (i.e. nulled) and will not be passed to the output. If this is done with two identical files, the output will be dead silence as there will be full cancellation (i.e a complete null). If this is done with two files that have very small differences, the output may be very low in level but it will not be dead silence. And of course, if the two files are more than a little different, the output will reveal this.

 

The interesting thing about the null test is it doesn't lie. It is not possible to have the slightest difference between files and produce a null test that will not result in some content in the output. (Summing +1.00 and -1.01 can never equal 0.00; it will always result in -0.01.)

 

Similarly, it is not possible to have two identical files and produce a null test that will result in anything other than dead silence. (Summing +1.00 and -1.00 can never equal anything other than 0.00; it will never result in -0.01.)

 

I am fully open to the possibility that what I've said in the previous two paragraphs is mistaken. I currently believe it is not mistaken in the slightest and would invite evidence to the contrary from those who would disagree.

 

***

 

I tend to trust experienced listeners, using high quality, well set up systems with which they are intimately familiar. All to often in my experience, things that were supposed to be "impossible" in theory or that sounded silly at first, turned out not to be so in practice. At the same time, I've also read plenty of "listening reports" where I question the writer's conclusions. (Or maybe they just have very different experience from my own.)

 

I've been able to take an .aif or .wav file, make a FLAC from it and re-expand the FLAC back to .aif or .wav (in an off-line process, i.e. not while listening). The result nulled completely with the original, 100%, to the sample.

 

Unfortunately, I have not figured out a way to incorporate real-time FLAC expansion (i.e. while listening) into a null test. Unlike the off-line expansion, I find the FLAC, expanded while listening, does not sound identical to the original file. So Soundkeeper does not offer FLAC. (It is one of my personal beliefs that convenience and Quality ride on opposite ends of a see-saw; as one goes up, the other must, of necessity, go down.)

 

Similarly, after reading reports of claimed differences between .aif and .wav files, I listened. If there are differences, I could not hear or measure them. In every null test I've done, the result has been dead silence, all the way down, 100%, to the sample.

 

I've already reported elsewhere on some tests of some "server" apps some report as sounding "better than iTunes".

 

There is no doubt that data-reduced formats like FLAC are convenient and make for more practical download times, particularly with high res files where the raw PCM versions can be fairly large. Those who do not hear a difference stand to save 50% on their storage consumption.

 

I do not believe CPU usage (as has been mentioned elsewhere) is responsible for the differences I hear during real-time FLAC expansion. I don't know what the mechanism is, only what the results are, to my ears. (I've heard of very similar responses from other colleagues, not one of whom uses data-reduced formats for their own listening.)

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[1audio]

Would destroy the banking and finance industry more that they have done to themselves. Google could not exist. There are data loss issues with hard drives but those have recovery mechanisms in them. The actual data loss as in failed data or drives is much much lower. Further most software would not be usable if the data was not many times more reliable (BSOD anyone?).

 

Here is some research that refutes the 10% number and puts an error rate at 1 bad block in 10E-14 or 1 in 100 Terabits or better.(a really really huge number) http://arxiv.org/ftp/cs/papers/0701/0701166.pdf

 

I would suggest a simpler test to be randomizing the original and the replicated tracks and a double blind test. This is so charged an issue that I don't think any non blind test would be satisfactory.

I have tried the blind flac vs. wave direct playback test here with very good stuff and didn't get a detectable difference. Perhaps the Berkeley DAC and Crosby Quads aren't up to the task.

 

The issue of the processor having more work and that affecting the sound with a compressed track may have some validity if its not adequately addressed in the hardware. However, if we are drifting into a realm where identical data streams will sound different, depending on where they resided earlier, it would suggest that either the streams are not identical or some other aspect is very different. If the reconstructed track is bit for bit identical (its very hard to fool an MD5 sum) we then start to get into an area where everything gets really fuzzy. One could wonder whether the location of the track on a hard disk will affect its sound. A really fragmented disk may have latency issues that affect reading the file. And different disks could sound different (they all have really different supply noise characteristics and it seems SSD's are noisier in my very small test sample).

 

 

 

[ziggyzack]

If digital isn't resilient, what's the point of it? Go back to analogue. Vinyl is cooler anyway.

 

[elcorso]

Panelhead, I'm sorry, but I stated in some posts than my main language is not English, but latin-romance: Spanish, French, Italian and Corsican.

 

I was very happy when the Esperanto language was invented, but nobody like it, because it is too complicated.

 

Then "guy", "gay", et al, is still confusing to me.

 

I know women has better ears than we have, my wife is telling me all the time to turn down the volume control!!!

 

I am with you regarding the "lossless compression story": Who invented MP3 should be in jail right now, because that's why kids and adolescents are so agressive on this days, they listen to noise instead to music!

 

But MP3 was also invented because the low storage, fast download "benefit". Thanks God women's were created by Him in real size!

 

Cheers!

 

Roch

 

[dsnyder]

Hello,

 

I was able to complete my own eleven step digital file torture test and listening tests today, and I have a few observations that I'd like to share:

 

 

[*]I selected "Rondena", "Muriel", and "Tomate" (sp) as the three tracks for my test, which were 43MB or so after being converted to FLAC and ZIPed up--too large for me to attach to an email from Comcast. Therefore, I had to limit my testing to just a single track..."Rondena", played by Gregory James. I emailed the ZIP attachment for each iteration from a Comcast webmail account to Gmail.

[*]The whole process was tedious. I had to create a folder structure on my computer to keep things organized. Each step and iteration within that step has its own sub-folder.

[*]The CRC checksums for each successive CD-DA rip with dbPoweramp were different. I thought that the first two might be different due to different file offsets, but I was surprised to find that none were the same!

[*]The file sizes for the successive RIPs were all close, but not identical. They differed by as little as a few hundred bytes and as much as about 3.5KB (works out to about two hundredths of a second if my math is correct).

 

 

For my listening tests, I took the original WAV file downloaded directly from Blue Coast records (unpacked from their ZIP file) and the final WAV file that was produced by the last setp of my process (outlined earlier in this thread) and loaded the pair of them into a playlist in Mediamonkey five times for a total of ten tracks. I then randomized the play list a few times (from the Now Playing pannel, there's no way to see which track is which without opening the "Properties" dialog to observe the actual filename) and set the list playing.

 

I tried my best, but I was not able to discern a difference among the tracks. I heard nothing in the timbre of the guitar, the soundstage, attack, decay, background noises, the audience, etc. that might distinguish one track from another.

 

As folks have pointed out on this thread, there are a few possible explanations for this:

 

• I may lack the aural acuity or critical listening skills to discern the differences
  
• My playback system is not sufficiently transparent to reveal the differences
  
• Besides a hundredth of a second, give or take, of offset from start, there is no difference (the PCM data in the files are identical)
  

 

My next step is to have a look at the actual PCM data, but for now, I prefer to believe that the last explanation is the best one for this test when executed using my computers, my optical media, my email accounts, and my playback system (YMMV).

 

As computer audiophiles, we should all hope that this last explanation spot on for each of our systems, because if it's not, we are better off dumping this digital junk and going back vinyl as another contributer suggested! :-)

 

-- David

 

 

[dsnyder]

Hello,

 

To complete my torture test, I used the free sox utility to have a look at the PCM data in the original WAV file from Blue Coast records vs. my copy of the file (which had been converted from WAV to FLAC, ZIPed, emailed as an attachment (probably base64 encoded) from Comcast to Gmail, downloaded, unZiped, burned to a CD-DA CD, RIP'ed using dbPoweramp, recorded to a data CD-R, ZIPed again, and then that whole process repeadted four more times before converting the FLAC file on the last data CD-R back to WAV). Yikes!

 

It was easiest to just have sox convert the two files from *.wav to *.dat. This produces a (huge) file that looks like this:

 

; Sample Rate 44100

; Channels 2

0 0 0

2.2675737e-05 0 0

4.5351474e-05 0 0

6.8027211e-05 0 0

9.0702948e-05 0 0

0.00011337868 0 0

0.00013605442 0 0

 

 

The format is time offset and then the samples from the two channels. I edited the files to trim the "0 0" samples from the start and end. Next, I used a simple awk script to extract just the 2nd and 3rd colums so that I would have only the PCM amplitude data for both channels (the time offsets are calculated by the sox utility and not actually part of the WAV file's PCM data, as far as I can tell). The resulting two files, one created from the original Blue Coast WAV and the other from the final WAV output of my torture test were identical both in length and in sample values.

 

If you get different results on your system, I encourage you to do some troubleshooting to figure out where the corruption is happening. Identify and fix this problem, and your enjoyment of computer audio will only improve. Cheers!

 

-- David

 

PS., I did this PCM analysis after my listening test (see earlier in this thread). If you try this with your system, I encourage you to do the same so that what you hear is not biased by the data.

 

[1audio]

I must give you credit for going to the extent you did to verify the accuracy of the different stages of the digital audio distribution chain. That's a lot of tedious work. I hope we can soon move past this.

 

[Guest Analoganimal]

...yes, in an ideal world we should all agree about what is better sound and have a smooth hierarchy between worst and best sound.

 

...problem is that audiophiles rarely agree about what is better and what is worse, so depending on your taste, agenda, inclination etc. what you actually think you hear will vary substantially from what the guy (gal) next to you think that he (she) is hearing.

 

...just visited the CES in Las Vegas and heard a variety of audio systems from he most mundane up to the grandest (in dollar terms). In my own book listening to an audio system that "sounds like a hi-fi" is something negative, while a set-up that is able to "mimic the actual sound of a live musical event" is the desired goal. Funny that the best sound at the 2011 CES (to my ears) came from a speaker system built in the 1940s playing back digital files (converted from old analog tapes).

 

...Blue Coast Records seem to say here is a sonic difference between listening to a WAV file that was originally made as a WAV file (although I suspect some recordings may have been made as DSD or DXD (352.8kHz) and later converted to WAV) versus a WAV file that was once converted and transported (via upload/download etc.) as a compressed FLAC (lossless) and then reconverted back to WAV.

 

...I personally listen a lot to WAV files with a special SDTrans 192 Rev. 3 battery operated SD card player (see DIYaudio.com) capable of direct playback of DXD 352.8kHz sampled master files (e.g. those recorded and offered by 2L.no), and I also play FLAC and WAV files from my MacBook Pro booted in 64Bit kernel with M2Tech Hiface EVO and AyreWave playback software. It is recommended for best playback sound to always convert FLAC stored files to temporary WAV files before playback, and not just play back the FLAC files "on the fly" (because the conversion has to take place real-time while playing). My DA converters are Berkeley Audio Design Group "Alpha DAC" and Fidelix Caprice DAC with ESS Sabre 9018 core.

 

 

...However, in my experience, the quality loss from temporary storage in FLAC, then conversion to WAV before playback, is negligible as long as the original file is of the best quality possible. However, I must say that listening to DXD master files of 352.8kHz sampling is ear-opening and "quietly spectacular" in a very natural sense. With 2L.no you can listen to the same recording in several sample formate (e.g 96kHz, 192kHz as well as 352.8kHz). The biggest problem is that very few hardware combinations are able to directly utilize and play the 352.8kHz files.

 

...looking at the Blue Coast records site, I found that many of the files they offer in WAV are actually only 96kHz sampling, and sometimes even 44.1kHz (CD quality). It does seem that they offer SACD (DSD) which is fine. I actually also listen to the SONY SCD-DR1 SACD player also capable of being used with external (SONY) SACD DAC, and recently a technology partner has managed to create a separate ESS Sabre 9018 based DAC that can input the DSD signal which is received from a slightly modified SACD player. The result is stunning.

 

...so, my own conclusion is that Blue Coast Records probably is best enjoyed through the DSD medium, and that the matter of whether WAV files are "packed" as WAV for transport or not does not really mater (because their WAV files offered may not be of the ultimate (possible) quality anyway). On the other hand; it is the MUSIC that matters, so very many of my own files comes from EAC ripped files from CD which automatically mean that they are 44.1kHz sampled. Many of these files actually sound wonderful, so I am not implying that there is anything wrong with lower sampled files.

 

...however, if a recording company today with currently available equipment offers their recordings for sale to high-end consumers, there is no excuse for them to not make higher sampled (better sounding) versions available.

 

 

 

 

 

 

 

 

[BobH]

I find it most revealing that someone with the experience, talent and industry understanding of Cookie Morenco should be, apparently, beset by the same insecurities as a lot of this forums' members. It is amazing to me how little trust there seems to be in the science that underpins the 'digital' age.

 

Perhaps there truly is a lot about all of this that science simply does not understand. Personally, my ears are way past trusting - the clearest sound I get these days is my tinitus! So I'm quite happy to be a bemused bystander to most of the 'discussions' that take place on this forum, these days.

 

What I can say is that, when it comes to recording, Cookie Morenco knows what she is doing! It is that fact that has given me pause for thought, as far as this conversation is concerned! I fail to see what she has to gain by sticking her head so prominently above the parapet. When the doctor says it's safe to release me back into the community, I shall have to compare my wavs to my flacs again!

 

[dsnyder]

Hello,

 

At the heart of this debate over how the handling of digital music files (adding/altering metadata, transfer through email, and conversion among lossless formats, etc.) may have a residual impact on their sound lies each person's definitions for "data" and "music", and if there is a distinction, at what point the transformation occurs.

 

If one believes that a WAV, AIFF or FLAC file fundamentally is music, then audiophile notions of how that music should handled can come into play. Music is intricate, fragile, and must be handled carefully to preserve as much detail from the original recording as possible. Transfer over wireless networks should be avoided. Before downloading from an on-line music store, be sure to use a high quality Ethernet cable of no more than 5m in length, directly connected between your PC and DSL router or cable modem for shortest signal path. Intel network cards are better at preserving mocrodynamics, but Broadcom NICs tend to deliver digital downloads with a larger soundstage. Save your downloads directly to a firewire or internal SATA hard drive (avoid USB and SSD because of higher noise levels). Downloads done late at night tend to have better sonic purity than downloads in the middle of the day due to less traffic on the network. Make sure that both your computer and your router are connected to a good power conditioner during the download and any subsequent transfer. Do not convert your downloads to other data formats...there's no such thing as lossless conversion or especially lossless compression! Do not attempt to add or correct the metadata since doing so can have a subtle but irreversible impact to the sound quality of your digital music files.

 

Contrary to the view above, I personally believe that a WAV, AIFF, or FLAC digital music file is just data. It's not music or sound any more than a spreadsheet, expense report, or last year's tax return. All data files are nothing more more than sequences of bits: 1s and 0s. The contents of these data files only have meaning within the context of application software that can read and interpret them. While the physical properties of storage media and signal transfer can have an impact on the integrity of data, modern computer systems and data networks have mature and highly reliable ways of detecting errors and correcting them with bit-perfect accuracy within multiple technology or protocol layers below the application. In the rare event that an error can not be handled transparently, there are facilities to notify the application and user of the failure so that corruption does not occur unnoticed. While it's possible that low level corruption can escape detection, this is so rare that most computer audio systems will not be affected by it during their years of useful service.

 

Since a digital music file is no different from any other data file on your computer, the same handling rules apply. You can make a copy of a copy of a copy of a copy of...a copy of a file one hundred times and the one hundredth copy of the file will be bit-for-bit identical to the original. There's no generation loss. It doesn't matter if you transfer/store files using wireless or wired networks, USB or firewire storage, SATA or SAS disks, CIFS or NFS network shares, Mac or PC hardware. The mechanisms for protecting data file integrity work the same regardless of the file extension, type or contents. If you find this not to be the case with your particular computer system, you should find out why and have it repaired.

 

So, at what point does data become music? In a computer audio playback system, at what point in the playback chain does audiophile handling of the music come into play and actually matter? These are important questions. I believe that this transformation occurs when time is applied to the amplitude values in our data files. It's time, after all that makes sound (and therefore, music) possible.

 

All of the bits in a music data file are frozen in time; they exist in whatever state they are in all at once and unchanging (unless something goes horribly wrong with the hardware). When those bits are interpreted as amplitude values and those values are associated with a uniform series of clock pulses in real time, at that instant the data begins its journey through our playback systems. It's at that instant (and not before) that handling of these delicate signals can have an impact on the sound that eventually reaches our ears.

 

While the signal is still in its digital form, jitter, or tiny variations in the timing of the clock pulses can have a significant impact on sound. Computer audiophiles use high end USB to S/PDIF converters with ultra high quality clocks, DACs with buffering and re-sampling, audiophile USB and S/PDIF cables, and all sorts of other techniques to minimize jitter before the amplitude and timing information reach the DAC chips. Once the signal leaves the DAC, at least the same level of care is taken to preserve the music in its analog electrical form until it reaches our loudspeakers. At that point, the signal becomes acoustic energy and so acoustic treatment and design become important.

 

At all steps in the signal path past the point at which time is applied to to the data, audio engineering approaches to signal transmission and handling come into play. Prior to that point, computer science and best practices that apply to data integrity are all that one must concern themselves with in the world of computer audio.

 

I went over a lot of details here which I hopefully got right. I'm confident that any mistakes will be corrected by folks on this forum in short order and for everyone's benefit, including mine! My purpose in all of this (including the tedious exercise, which I hope you took part in as well) has been to hopefully dispel the notion that modern computer systems are incapable of reliably handing music data files without introducing minute levels of corruption or degradation. If it were true (or commonly accepted as true), this notion would be damaging to the growth of the computer audio industry. Audiophiles who do not currently purchase high resolution digital files or USB DACs may not be willing to even give them a try if they believe that computers can't be trusted to handle these files without loosing bits here and there. Thankfully, computers seem handle data just fine, so let's give the all clear get back to listening and enjoying music.

 

-- David

 

[Audio_ELF]

4est/Forrest said: "Whether or not you are aware of it, most AVpres digitize the analog inputs subsequently doing their own DAC before the outputs."

 

Actually I don't think this is true. "Most" (a relative term at best) route analogue audio avoiding AD/DA conversion unless you specifically engage some form of processing mode. At a minimum the majority have 5.1 bypass inputs which bypass all but the volume control circuitry. By "most" I include (to my knowledge) Rotel (since the RSP1068), Lexicon (since MC12), Krell, Arcam and NAD have proper analogue circuitry. TAG and Audiolab have proper 5.1 bypass. Only Meridian generally force AD conversion.

 

Edit/PS to above. This also describes most recent AV receivers. Generally a "stereo / pure direct" mode where no sub woofer or other processing is allowed indicates proper bypass. These statements refer only to recent AV equipment. Older models (ie Rotel's RSP-1066) are more likely to have the AD stage on "normal" (ie non 5.1) analogue inputs.

 

As for the bold statements being made about transfering music files, I have asked for clarification of Cookies claims... If she is claiming that (a) digital data changes over time and (b) emailing or transferring data from one storage device to another causes data to change (at an audible level) and © that two files with identical checksum (and therefore identical data within) can sound different... well then I turn off until someone provides some evidence beyond wild listening test claims. This is not a FLAC vs WAV argument.

 

Eloise

 

[4est]

Please excuse my previous post as it seems to have side tracked this discussion. I was only loosely attempting to present a plausible explanation of Cookie's argument and why one may not be able to discern.

 

Ultimately I was just attempting to give due respect...

 

[Phil Townsend]

I am reminded of a young lady some years ago named Enid Lumley. RIP

Enid was a total nut case but with good ears. We would run into each other at CES and usually have lunch or dinner. No boy girl stuff... NO! She was not that kind of girl. Or maybe I was not that kind of guy.

She told be one year that her music system sounded better if the cables were lifted off the floor! "What! Raise the cables off the carpet! Are you on drugs Enid?"

Maybe she was... maybe she was not. Didn't care.

So I went back home to Miil Valley, Ca. and lifted the cables. I use small chunks of 2x4 spaced out under those big fat cables. A sure enough things did sound better. Not much but it was better.

 

Next year she told me about the damp towel over her SP3 preamp.

The list of stuff goes on and on.

At the time people would read her column in Absolute Sound and laugh. But as time pass they would try some of the stuff she talked about and found THEY WORKED! Enid had ears. But more to the point she trusted what she heard and reported to her readers.

 

 

 

[rykart]

David.

 

You have really approached this fascinating topic with a scientist's eye! Bravo! The defensiveness of some self-proclaimed audiophiles makes me deeply suspicious. Your randomized test was very cool. You ought to send the resultant files you made to Cookie to see if SHE can tell which track is which.

 

I'm reminded of a certain hi end dealer selling speaker cables for some outrageous amount of money. The magician and debunker James Randy issued them a challenge to prove their claims in a double blind study against lower cost cables. They went on with the same insolent crap about their ears being superior but at the end of the day, declined the challenge.

 

[audiozorro]

music played from a spinning hard disk drive. I don’t want to believe that this is true, because it provides me with another reason to spend money; the rationale for any audiophile seeking superior sonics to replace their HDDs with SSDs.

 

IMO there is always something better but my advice is usually to enjoy what you have and upgrade when you hear the difference that is worth the money you can afford and are willing to spend.

 

Are WAV, FLAC and AIFF audio files different? – yes. Do WAV, FLAC and AIFF audio files contain the same PCM audio data? – yes. Can the WAV, FLAC and AIFF audio files sound different? – yes. Are the sonic differences very small? – yes.

 

Of these three I prefer the WAV format. But since many of the software players that I use can play WAV, FLAC and AIFF, I usually just play them as they are and I do not bother to convert them to just WAV files. Actually I am much more interested in obtaining the audio files in the studio master formats, whether PCM, DSD or analog.