Sound is better from uncompressed downloaded files?

[dsnyder]

Hello,

 

You are correct that some of these things actually work. Check the last link in my sig...a photo of my listening room which shows that my speaker cables are lifted off the floor using stands made from sets of three chopstics and some twine. LOL!

 

Again though, this is in the music/sound domain where these sorts of tweaks make sense. Let's be careful about making similar claims of sonic improvement in the data domain. Unless your data domain tweak actually alters the PCM or bitstream data (eg, upsampling), I humbly submit that any difference in sound that you hear may actually indicate a shortcoming in your playback system. If you can successfully identify the reasons behind the change in sound that your data domain tweak seems to make and address it (fix it without making a change downstream that merely masks the difference), your playback system should sound even better than it did before or after the tweak.

 

For example, some folks have reported that if they take a FLAC file and convert it to a WAV file before playing it, the sound quality of their systems improves vs. just playing the FLAC file directly (decompressing on-the-fly). If this is the case in your system, it's not because the PCM data are different. There's something about the additional CPU workload, extra load on the memory or I/O bus, or perhaps even a bug in your playback software or operating system that is causing this difference. Other background processing that your computer may do from time-to-time (not related to music playback) is almost certainly also compromising sonics even when you play the uncompressed WAV file. Decompressing FLAC to WAV up front is a nice workaround, but if you can find and address the root cause for the difference (eg., find a better media player application, address a performance bottleneck, add more RAM, buy a faster CPU or disk drive, etc.), the sound of your system will improve that much more.

 

Make sense?

 

 

[bdiament]

Hi dsnyder,

 

..."For example, some folks have reported that if they take a FLAC file and convert it to a WAV file before playing it, the sound quality of their systems improves vs. just playing the FLAC file directly (decompressing on-the-fly). If this is the case in your system, it's not because the PCM data are different. There's something about the additional CPU workload, extra load on the memory or I/O bus, or perhaps even a bug in your playback software or operating system that is causing this difference. Other background processing that your computer may do from time-to-time (not related to music playback) is almost certainly also compromising sonics even when you play the uncompressed WAV file. Decompressing FLAC to WAV up front is a nice workaround, but if you can find and address the root cause for the difference (eg., find a better media player application, address a performance bottleneck, add more RAM, buy a faster CPU or disk drive, etc.), the sound of your system will improve that much more."

 

"Make sense?"

 

Actually, while it might appear to make sense on the surface, I think you might be missing something. The FLAC file contains "the same data" only after it has been expanded.

 

I don't believe CPU or hardware deficiencies are the issue and would think there is a relationship here with what I (and a good number of colleagues) hear with sample rate conversion algorithms. I listen to and test a lot of SRC algorithms and have found that even with the very best algorithms, results are not the same with on-the-fly conversion as they are with off-line (i.e. not while listening) conversion. Invariably, there is a performance hit with on-the-fly conversion and the result simply does not sound nearly as transparent (i.e. the most like the unconverted original) as it does when the same algorithm is used off-line.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[audiozorro]

of a spinning hard drive, there are several improvements to be had in the digital domain such as defragging your hard drive. Again, the differences are real and subtle.

 

Here are some expert tips that were written for recording audio and video, but are just as applicable for audio and video playback:

 

http://www.jakeludington.com/ask_jake/20050225_optimize_your_pc_for_audio_and_video.html

 

 

[elcorso]

Hi Barry!

 

I'm happy you meet this post, another recording engineer...!!!

 

I do admire David excellent work analyzing file structure (music tracks): Length, bit by bit check, end of file matching after conversion, with no conversion at all, and sending both by email, and making comparisons before and after. A REAL labor of love!

 

But, I still have the same problem: FLAC (and then expanded to AIFF or WAV) doesn't sound as good (to my ears) as direct download from not compressed formats.

 

I made a lot of different tests, but has the same result. Of course with good recorded music, as yours in "Equinox" (DVD-no compression), and 24/96 WAV Zip (no compression on Zip) downloaded from Blue Coast Records. And from others: MA Recordings, AIX recordings, Tone-Pearls Records, 2L, etc., etc.

 

I know, FLAC (after conversion) is no so bad, but no so good as the no compressed ones. Then, if I pay (and not peanuts) I would like to have the choice between both downloads.

 

More: When I convert WAV to AIFF, in order to add art, I do notice some deterioration, but minimum.

 

Even more: I didn't make any listening test (I didn't have the time, or I was tired), but I got different file size, from converted downloaded FLAC files converted to AIFF, than AIFF direct download...

 

Then, something is "Lost in Translation", as the film, or is something added? And in downloads? I don't know, but human hearing is faster than our eyes capability. I have a close friend (a very good classic guitar player) that says that if our eyes had the same "power" of our ears, we could see a 40 watts bulb lamp at 40 kilometers (about 25 miles), at night of course...

 

And regarding audiozorro post, defragging is a very important issue on PC, but not in Mac OS Extended hard disk estructure and files format. But on Mac we some other issues that affects music reproduction also.

 

This will be my last reply to my original post, I think a lot of members are tired, or feel uncomfortable regarding this post. Too much money has being expended on a lot of downloads, as is my case.

 

My main goal was, and still is, to get "real" hirez music, and don't be fooled with 24/96, 24/192 "resolution".

 

Cheers!

 

Roch

 

[1audio]

ZIP and FLAC are very similar. FLAC is optimized for real time uncompression where ZIP is really meant for static data. Also the file sizes will depend on the imbedded metadata and other non-audio content that is handled differently in different containers.

 

Again, can you tell the files apart if you don't know the source?

 

[elcorso]

Welcome again, Mr. Martin,

 

ZIP files utilise compression algorisms, and it is easy to be transmitted beacause they are better "understanded" by web browsers and email app's than FLAC, WAV or AIFF files. There are a lot of documentation regarding this matter, but I don't want to bother you with this.

 

I downloaded ZIP files (from Blue Coast Records) containing NOT COMPRESSED by ZIP, NOT COMPRESSED WAV files. Then this files are less error-prone (I suppose this by listening test).

 

And even there are app's that get you to:

 

Easier (and less error-prone) creation of Zip files

 

http://blogs.mathworks.com/pick/2009/09/18/easier-and-less-error-prone-creation-of-zip-files/

 

• "FLAC is optimized for real time uncompression where ZIP is really meant for static data".

 

Yes, but I never liked to play FLAC files directly from any of my computer music players. But this is a matter of "musical taste". And then, if you open "Activity Monitor" app on a Mac (there are also similar app's for PC), you will see an "extra" activity from your processors when you play FLAC, because there is a "conversion in the fly", then there is not such thing like "real time".

 

And yes, you are right, ZIP is for "static data". From Wikipedia:

 

"Static data means the data that occurs one time during its life time. Once it is created it cannot be deleted or modified"

 

But I don't play the music from the ZIP file, but in WAV or AIFF after the file are UNZIPED.

 

• "Again, can you tell the files apart if you don't know the source?"

 

I didn't completely understood your question, maybe you are writing in "compressed English"(and this is not a joke): First, I don't know any of the sources: Where the file come from? I don't know, and I don't care, because recording is NOT my business, my only passion is to listen to nice, well recorded, music! But a forum member said, the source is maybe from DSD...

 

What I can compare between 2 files, that are (supposedly) the same, but one from a direct AIFF download, and the other from a converted FLAC to AIFF, then in comparison of the two AIFF's, I like better the direct download AIFF. That is the only thing I can tell you, but again, is a "matter of taste".

 

In my last post (and supposed to be the very last) I forgot to mention Dr. Johnson, from Reference Recordings: Excellent music (an also excellent recorded) 24/176 WAV's on DVD.

 

And please forget me, I don't like to make statements regarding computers kernel, math, nor algorisms, but I was "forced" to this, since you make the questions.

 

Thanks and happy listening,

 

Roch

 

[1audio]

I guess the file is zipped to incorporate the error checking of the zip process. I had never seen that done before with uncompressed files. The file transfer process over the web has lots of error checking as well (and some legs may have compression as well e.g. IP Payload Compression Protocol (PCP)). I have never seen a compressed zip file have errors either if its processed correctly.

 

If it makes one feel better I suppose one can avoid the evils of compression (and many other misunderstood technologies). I would prefer to use less bandwidth and use a container that has good standardized metadata support, something missing from .wav .

 

My point about comparing files was to see if lossless compressed and expanded files are mixed with never compressed files randomly can the processed files be detected.

 

The link about ZIP files had no insight beyond facilitating adding Matlab stuff to a compressed file reliably.

 

[sandyk]

Cookie, Barry and Roch

You are all correct !

However the problem is at the decoding level due to PSU interaction/"Jitter" when played back on the fly.

I have a fairly resolving Class A amplification system,as well as a Class A Headphone amplifier fed from a highly modified Musical Fidelity X-DAC V3,

and when I play back .wav files of the decoded .flac files that it originated from,

using a USB stick plugged into my WD TV Live HD, which now uses a high quality +12V Linear PSU,

it sounds like a veil has been lifted in comparison with the .flac version played from the same USB stick.

However I have DL 24/96 .flac files from Linn Records in 2008 (Claire Martin-Too Darn Hot). and it has been my experience, that as I improve the sound dampening etc. of my PC, that subsequent conversions to .wav files.sound a little better.

I now convert such files from my HDD directly to a Corsair Voyager GT USB pen that has it's power supplied from a very low noise,very low impedance +5V Linear PSU instead of noisy internal SMPS. The differences are quite marked, and those same files from 2008 now sound as good as my best ripped 24/96 DVD-A .wav files,and no longer just a little better than a well recorded 16/44.1 CD.

 

 

SandyK

 

P.S.

Although I am approaching 72 years of age, through this system, I can appreciate the differences of well recorded 24/192 ripped DVD-As such as "Carly Simon - No Secrets" compared with other 24/96 .wav files , or even more so compared with the original CD.Keep up the good work guys !

 

[PeterSt]

G'day there Sandy !

 

I've been just struggling through this thread, and I guess people should forget about resulting files not being equal. They just are, unless something is seriously wrong with your computer. This is not to be assumed, nor is there any 10% loss of data anywhere. Of course, your backups on DAT from 20 years back may not be as reliable as you hoped, and if that it counted in to reach some 10%, well, maybe. Otherwise ? otherwise it is zero. Period.

 

But Sandy ... of course we remember you from some "ridiculous" applications to data media and in the end stuff which is subject to this thread. And as you know, I always believed you. Well, from an IT perspective I can't, but from an audio listener perspective I will. Or maybe I have to. If I don't hear it, it will only mean I don't hear it. But there's a difference in not wanting to hear, or not being able to. The former may be because of "IT thoughts" and the latter can be a not resolving system.

But as I always promised, there's also the angle of diving into the IT matters and find out whatever *can* make the difference, no matter it can't to begin with.

 

Riddles riddles.

 

I guess it has been two years by now that you came up with your appliances, and I never could let go of it.

Btw, this is different from FLAC being decoded in real time, which easily (actually the MOST easy) makes a difference compared to playing the WAV file directly. It really doesn't need a thread full of guesses like this one, if you're only more into what the computer can cause to harm sound. So, yes, the real time decoding does matter, and it is easy to see how. Ok, for me that is, but ain't I spending my life on it.

Nu further explanation, because I could write forever. Don't believe it if you want.

 

Today, there is so much more, that we're only at the beginning of it all. Two or three people may exist by now who have the guts to take some basic influencing things for granted (it includes me, and all are IT pro's), and go further from there. Many things can be explained now, but some cannot yet. Stupid example :

 

Playing back from RAMDisk sounds better, and while I did my stinking best to mimic a manual copy of the data (music) into the RAMDisk programmatically, when people do it themselves it sounds better.

By itself I have no doubt at all that this exists (starting to learn the reasons why), but at not being able to mimic this manual copying with the same result, there's still some bits and bytes missing in the complete scenario.

 

While this is an example of something which is not under my control, there are quite some which are by now, and which were not only 4 months back. Stuff like the example, and unexplainable, until ... well, you can. It only takes that guts to assume some basics for granted (no matter there's no real logic in it at first), and proceed from there. Well, how did XXHE ever start eh ?

 

After reading this thread I thought not to respond to it. This, while there SO much to tell about it all. But it can't be done with understandable sense. Not yet. It will though, just wait.

But when I saw the last post from you, Sandy, I got triggered by the message I like to spread :

 

Only if you have the sheer guts to believe in what you hear, and next maybe spend a life on explain it with the "techniques" you may carry, audio WILL progress.

 

It doesn't matter whether it's someone like Cookie doing it, Barry, David in his way, me in my way, or you Sandy. If you start off with the difference being there, it is a matter of finding out why. Oh, we could skip the latter, but it's less good for progressing on it all.

While not everyone is capable of explaining it, it can turn out to be dangerous to come up with "thus reasons" anyway. The 10% thing is an example. Dangerous, because the person concerned may be shot at because of just coming up with the wrong explanation. This does, however, not mean the person wasn't correct in his or her observation in the first place. So, let's try to separate that from the observation itself.

 

Sandy, if I may say so, you are different. You always withholded from explanations, and not being an IT guy is your luck. It is more easy now to have that guts, and proceed with it. Keep on shouting to the world until so many agree that there must be something, and in the end someone may pick it up and sort it out.

 

Just wait. There will be so much more of this ...

Peter

 

 

 

[sandyk]

I was surprised to recently read a post by you where you didn't find the higher res comparison files sounding better on your set up, as that is not my experience. (they wouldn't DL for me)

The thought crossed my mind, that perhaps there may be a little too much evaluation of your various "XXHE engines" by a "committee"

Too many people in your forum have very different system resolutions, and many people often pick the sound of a file that sounds louder, instead of which is the least fatiguing when you turn up the volume, as well as no loss of detail when you turn the volume down a couple of notches either.

I would like to see you rely more on your own ears and high quality system.

Kind Regards

Alex

 

[Quest]

One day some months ago, I did a blind test on my girl friend for a AIFF file versus a ALAC file (converted from same file)... and she picked out the AIFF file as the one she liked 10 out of 10 times, citing tonality as the main difference.

 

Next day I converted all my music to AIFF.

 

It's good to question, but when harddisk space is abundant, sometimes you don't need to.

 

I didn't dare repeat the test with WAV as the lack of tags would be a deal breaker for me.

 

[bdiament]

Hi Roch,

 

Aside from differences in all of our systems, I've said for years that different people seem to have different sensitivities to different aspects of audio.

 

I am at a loss to explain some of the things I read here on CA and on other audio sites (both audiophile and pro). Rather than attempt any proclamations of Universal Truth (which I find silly), I can only report on my own experience.

 

I find a sonic price is exacted by the so-called "lossless" formats when they are played directly but not when they are expanded in a separate process, prior to playback. In other words, I have not been able to discern (by listening or measuring) a difference between a source .aif file and the result of turning that file into a FLAC and then expanding the FLAC back to .aif (again, in a separate process, prior to listening). When I conduct a null test with the two files, the result is dead silence, 100%, to the sample, which says to me the files I compared are identical.

 

After reading reports here of perceived sonic differences between .aif and .wav files, I conducted several tests with different source files. Once again, I could not hear or measure a difference and all the null tests, again produced dead silence, 100%, to the sample.

 

I have not detected any losses between original files and the same files uploaded to a server, then downloaded again.

 

I have found significant losses with all forms of digital tape, after a period of storage, even under the best, climate controlled conditions. Error counts increase measurably and in the worst cases (fairly often with the DAT format) sections of audio are lost; playing them back results in a sound very much similar to that from a fax machine(!).

 

I've tried several third-party power supplies (sealed lead acid battery as well as linear) with my ULN-8 and have found some actually degrade the sound while the best sound indistinguishable (to me) from the stock supply; the only differences being they take up considerable space in the listening room and cost hundreds of dollars.

 

I've tried several "server" applications that are claimed to "sound better" than iTunes. Yet, I can null the results from any of them against any of the others -as well as against the original masters used to create the CDs ripped into these apps. (I experience clear differences in user interface though. ;-})

 

So, as I said at the start of this post, I'm at a loss to explain many of the things I read here. Some folks talk about hearing things I do not and some folks talk about not hearing things I do.

 

Always, I remain open to the possibility that a new experience might alter my perspective one day. There was a time I didn't believe turntables or cables could sound different. But in fact, I've never heard two that sound the same. There was a time when I didn't think what a component sits on could alter its sound. Now I realize anything a component sits on will alter its sound, though to my ears, more result in mere changes rather than improvements.

 

As far as my own work, I must stay true to what my own experience tells me. This is why Soundkeeper's files-on-disc releases are only offered in raw PCM formats (.aif and .wav). I currently do all my recording and mastering in raw .aif format. All the music on my server is in this same format.

 

I find it of great interest to read/hear about other folks' experiences, whether they are in line with my own or at complete odds. There is always something to be learned.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

[Panelhead]

I am learning shortcuts from the postershard work. My only recordings have been vinyl rips. Have a 2TB drive to hold the files for editing.

From what I gather storing as ALAC will be fine. Just move to ITunes library as AIFF. The computer plays back from memory (ram) but this is the same file as the the hard drive. The expansion or decompression from ALAC to AIFF still occurs real time.

I am learning.

 

George

 

[bdiament]

Hi George,

 

"...Just move to ITunes library as AIFF. The computer plays back from memory (ram) but this is the same file as the the hard drive. The expansion or decompression from ALAC to AIFF still occurs real time."...

 

If I understand you correctly and you are importing into iTunes in .aif format, isn't the file being expanded during the import?

 

If so, I would refer to this as "off-line" expansion - unless you are listening to the file as it is being imported.

 

You are listening to the files from iTunes in the format in which they are stored in the iTunes library. If the file in iTunes' library is an .aif, there is no expansion during listening. This would only occur if file in iTunes' library is an ALAC.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[Panelhead]

Barry,

My original was to move the vinyl rips into iTunes as 88.2/24 ALAC files. This is to save some disc space. Now I think 44.1/16 or 44.1/24 AIFF will be fine. My Apogee Duet ultimately limits the quality of the files recorded.

 

George

 

[losingmyreligion]

 

If I might add to AudioZorro's post - here are a few data specific scenarios that could result in differences in sonics:

 

1) "on the fly" FLAC (or other lossless) processing

2) differences in endian-ness (i.e. WAV vs AIFF)

3) meta-data processing (e.g. WAV vs AIFF)

4) differences in drive offset (as it relates to reading from optical drive during initial file copy to disk)

5) files stored on disks in (significant) need of defragmentation (vs. NOT)

6) files stored on SSDs vs spinning hard disk

7) files streamed wirelessly from external storage (versus local storage)

8) files accessed from NAS (i.e. storage device with it's own operating system) versus local storage

9) files stored on USB external drives (which use computer's processor) versus local storage or perhaps Firewire/esata

 

These are likely only some of the aspects of data files (and how we access them) that could conceivably influence the sound produced by one's DAC, despite that (in most cases) file comparisons would elicit the same checksum, and in all cases the PCM data would be identical.

 

The morale: be careful how you store your data files.

 

 

clay

 

 

 

 

[soloburnmike]

David I was really okay with your post as being objective and sticking to your observations until this part:

 

"My only concern is that the credibility of Blue Coast could be affected because of their outspoken claims that WAV sounds superior to FLAC in their testing...and that there could be a difference in sound among FLAC compression levels. The PCM data are the same. Any verifiable difference in sound quality can therefore only be attributed to a shortcoming somewhere upstream in the playback system."

 

My intent here is NOT to fan the flames - just state a few observations and facts. I have been involved in the development of pro audio products for 20 years. I cut my teeth at Studer the company most of us remember as making some of the most amazing multi-track recorders on the planet - and probably recorded more grammy records than any other recorder. I had product management duties for all types of analog and digital audio products over the years at Studer, Digidesign, Tascam, etc. I was involved in development of some of the first pro digital products and after that the chip sets for pro gear. That was the bleading edge of pro audio and Cookie has lived it with me.

 

Cookie Marenco, Blue Coast's founder has the most amazing ears I have ever heard. I brought her in countless times to help me develop products. She has helped my companies win TEC awards, AES awards, even technical Grammys and Oscars. I joke that if Cookie says "well it doesn't completely suck" then I have a winner!

 

From recording monitors to A/D converters Cookie would nail down problems even my internal programmers and hardware engineers couldn't find. At Digidesign in the 16-bit days she claimed to hear something that no one else heard. The EE's did bit comparisons all day long and laughed. The software guys thought she was nuts. Sure enough about TWO YEARS later the bug was isolated and fixed because another "golden ears" heard it. Cookie stuck to her guns and she was correct.

 

So if Cookie tells you she hears something, I have a list a mile long of industry winning products that tells you she does. There isn't anything subjective about that. She is known throughout the pro world for this reputation and she may honestly be the best.

 

In the pro world the test gear gets you close and people like Cookie take you the rest of the way. Her's the punch line: there is no way on this green earth bit comparisons and test gear find the details like the human ear does - period. I know that sounds crazy but talk to any of the good engineers at the best pro audio companies in the world and they will tell you the same thing.

 

Look at an AP plot for a $79 mic and a $5,000 mic. They look pretty much the same. We get products as close as we can with testing, then we take them to the pros for verification.

 

It's people like Cookie that find the devil in the details that we pro audio people love. So please don't throw someone under the bus based on your observations just because you don't hear the difference. I completely get you don't hear it. I don't hear many of the things Cookie hears!

 

I vote for everyone on this thread to grab your martini, close your eyes and kick back with your favorite song. Isn't the music what its supposed to be about anyway as opposed to statements about someone's credibility?

 

FYI - just so you know I am a real person, you can find me on the internet: Michael Newman - I now do digital marketing for musical instrument companies. But if I start developing ANYTHING in the audio markets again you know who's going to help me. Peace and love high rez folks.

 

I know it's time for me to grab a drink!

 

 

 

 

 

[ziggyzack]

The EE's did bit comparisons all day long and laughed. The software guys thought she was nuts. Sure enough about TWO YEARS later the bug was isolated and fixed because another "golden ears" heard it.

 

So how was the bug isolated and fixed? If there was a bug, and they could fix it, then something must have been broken... i.e. sending out the wrong bits. So the EE engineers missed it earlier - cocky little sh*ts! But Cookie is saying she hears differences when the bits are identical, and attributes it to digital generations, email, etc., which is nonsense.

 

[Elprior]

Hi there,

 

With all due respect to all recording engineers here, this data "generations" and corruption issue is really a non-sense.

 

I've been a software architect for the last 10 years - just upgraded to design authority - not to mention the latest 20 years as a software developper.

 

We are using several types of storage for the source code, most of them being handled by software configuration management (SCM) softwares, storing either in a database or on a dedicated file system. We are exchanging data through emails/nas/SCM/... using tcp/ftp/xmpp/... protocols dozen of times everyday. I have yet to see any corruption of data due to the resuling "generations".

Sometimes the physical support is failing, but then the data is simply no longer accessible at this physical point or recovered to another physical point, and the new copy just looks exactly the same.

 

If audio files were affected by email protocols, then the text of those emails would have to be too (all of this is really just data).

 

Go ahead, write yourself an email and forward it, again to yourself, as many times as you might want to try it, and compare the initial content with the embedded (forwarded) one.

My guess is you will stop this little game before even being able to find an error in the original text.

 

As for the sound being modified by physical location. Well you may just want to try to move it to another location, and load/decode it in memory, then read it from there.

 

Ok, now you will tell me that the playing software is not always using the same memory addresses, and sometimes it's closer from the kernel addresses, so that data is corrupted again, and it makes the sound harsher (or smoother, who knows). Maybe you will end up with specifying the addresses to use, just to make sure.

We are doing that all the time with our native libraries, but that's because we want to be able to allocate the maximum amount of continuous memory to a program, not because we fear that it may affect the data.

 

To wrap-up, this "generations" joke would lead our programs to evolve by themselves as they are being used (is it not a virus we are talking about ?).

Maybe an audio file can slowly become a bit-perfect audio player by exchanging it through emails ?

 

Darwin would be proud...

 

Elp

 

[sandyk]

" But Cookie is saying she hears differences when the bits are identical, and attributes it to digital generations, email, etc.,"

I posted similar back as far as 08-15-2009.

I even provided links to uploaded comparison .wav files with identical checksums for interested members.Very few took me up on the offer, though.

4 CA members posted that they did did indeed hear audible differences.( 2 of those members have verified this in recent posts.) In the meantime I have uploaded dozens of comparison .wav files for members of another forum, and virtually all of them have been able to hear the differences , and correctly nominated which track was the more recent rip.

(The majority were using headphone amplifiers.)

I have also had confirmation from an English musician who still tours, as well as from a retired U.S. Professor of Music (>80 years old!) who toured with the cast of Porgy and Bess in 1954.

 

 

[PeterSt]

But Cookie is saying she hears differences when the bits are identical

 

Yes, I left out the remainder. But :

 

Never forget that this (quoted part) is non-sense within itself. I won't talk about playback software (bit perfect), but just take Vista and compare it with W7-SP1. Both measure bit perfect - hence the bits are identical. Anyone who doesn't hear the difference between these two OSes is DEAF.

 

But the quote in the full context :

 

But Cookie is saying she hears differences when the bits are identical, and attributes it to digital generations, email, etc.,

 

Yes, I too will agree that this is non-sense.

But as I said before, once we have the guts (or stupidity, also good :-) to believe in this just because we perceive it like that (differences) it is a matter of finding the proper explanation. So what I would say on this one is :

 

Yes, totally non-sense within itself. Can't be. Out of the question. Do you hear ? good.

 

but

 

Since I tend to believe people who claim such things anyway, now go seek for what can actually be happening. To name an option : how will that newly received email be stored opposed to whatever original you listened to first. It won't be at the exact same spot for sure (out of your control). What does that do ? what does it imply for the OS ? how can it be dealt differently with - ehh, wrong - how WILL it be dealt differently with, because there IS a difference.

 

If one thinks *this* is non-sense, then maybe it is time to follow the development of one of the two players which explicitly deal with such things. Look what is behind it. Look who is behind it and what the attitude is. Look what is (technically) being attempted, trialed, done - and with thousands of people judging and reaching the same conclusions. Look why these players ever and ever proceed. Always outputting the same bits, I guarantee you that.

 

For some counterweight :

 

At this moment I wouldn't know of any "mechanism" that will allow for a different(ly intended) sound when a sent FLAC is decompressed to WAV and next use that WAV. Or IOW, whether a sent email or download may change the sound of very indirect reasons (see above), the file itself won't be infected. And so, yes, a FLAC will (!) sound different at real time playback, but whether a WAV is sent by email or a FLAC is sent by email (or anything else), it really won't make a difference if you only decompress the FLAC into WAV. Mind you, if the result is 100% the same that is, and this is already not the case when offsets are different (see earlier post about this suggestion too). With a different offset, the file will sound different throughout. Nothing difficult to reason out for me, and not necessary to layout either.

 

Again the moral : take a perceived difference for granted, but find the right explanation, instead of not being able to justify it because you happen not to know it all (err, I don't either).

 

Also, look what people are actually telling you. For me this is easy again : Cookie tells about a 10% loss in etc. etc., and THUS finds justice in emails being wrong or whatever. Well, Cookie *will* hear the things as she hears them, but doesn't know it all. The 10% loss is BS (sorry Cookie), and thus explanations derived from that are too.

The base -perceiving differences- is still there though, if only that *is* the base, and not a 10% loss etc. or otherwise it's a placebo.

 

Once more : look how these players develop. They develop from these initially perceived non-senses, and certainly not from the other way around. Sign up to Hydrogen and find yourself fixed forever. I hope you understand what I mean.

 

Peter (steamed up a little :-)

 

 

 

[Miska]

take Vista and compare it with W7-SP1. Both measure bit perfect - hence the bits are identical. Anyone who doesn't hear the difference between these two OSes is DEAF.

 

Or they have non-broken system. IMO, any hardware where same data output sounds different depending on OS version, storage format or something similar is fundamentally broken and needs fixing.

 

I can hear differences between different resamplers, dithers and noise shapers, but not between realtime decoded FLAC and WAV. Nor between different OS versions. If I would, it would be extremely annoying since it would make development of those algorithms nearly impossible since I wouldn't know if the difference is because of difference in the math or something more or less random.

 

If I'm deaf, I'm happy being selectively deaf. And prefer staying that way.

 

 

[f1eng]

Do I understand correctly that Blue Coast records "ship" their WAV files zipped?

If so what about the lossless compression of zipping the file allows it to be reconstituted by unzipping it in a "superior" way to converting FLAC to WAV?

If the file is "shipped" directly as WAV this question is moot, of course.

Frank

 

[bdiament]

I find it interesting how so many folks in Internet audio trot out their credentials just prior to making pronouncements of Universal Truth (add reverb under those words).

 

While I have not had the same experiences Cookie mentioned, all I can say for sure is that I have not had the same experiences. I can't speak for her experiences or anyone else's and I'm wary (to say the least) of folks who do. To say things like "nonsense" about someone else's reports reveals more about the person saying it than it does about the one they are disparaging.

 

I've seen plenty of commentary on this and other fora that are completely contrary to my own experience. But how do I know someone else didn't experience something that just eluded me? I don't.

 

I remember the early days of digital when I simply found it a poor substitute for good analog recording and playback. At an early AES panel of mastering engineers discussing digital and CD, mine was the only voice in the room (shyly at the time) critical of the new media. I found (and still do) an appreciable increase in harmonic coarseness and an appreciable diminution of focus and fine detail in digital back then, all superimposed with a smeared treble that was often discomforting. Many colleagues looked at me like I had two heads.

 

Most of my colleagues echoed the marketing spiels in the magazines (both pro and audiophile), saying "Just look at the frequency response!" to which I would respond "Just listen to the music!" It was only years later when concepts like "jitter" became widely known that those folks started to notice problems with what they initially touted as "perfect".

 

Folks who argue that identical data cannot sound different have clearly not compared any pressed CDs with the masters from which they were made. Further, there appears to be much confusion between data and audio. We don't listen to data - at least I don't. (If you can think of the sound of a fax machine, you'll have heard "data".)

 

Arguments against Cookie's points, saying that word processing and spreadsheets would also be corrupted neglect the extremely important fact that word processing and spreadsheets are not time based, where audio is.

 

As I mentioned earlier, I have not heard many of the issues Cookie speaks about. I have not heard many of the things other folks speak about too. And apparently, other folks have not heard some of the things I speak of. These are all facts.

 

To denigrate another's comments or to dismiss them as "nonsense" is not to speak facts, it is to excrete an opinion.

 

I find it interesting that the folks from whom I've learned the most on my own audio journey all have in common that they spoke to me of their own experiences and not of anyone else's. These folks had/have the confidence (as well as the knowledge) to speak in the first person.

 

My point here is to request a bit of mutual respect. I wish folks would disagree when they hear it differently but frame it in terms of their own experience. Just a personal wish. All too many Internet audio fora quickly descend into cesspits. I find differences in perspective are things I can learn from, when folks have the self confidence to not denigrate others.

 

Thanks for listening.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

 

[PeterSt]

Or they have non-broken system. IMO, any hardware where same data output sounds different depending on OS version, storage format or something similar is fundamentally broken and needs fixing.

 

Miska, No.

I know that you are saying this all the time, and I heard you implying that out of all you are not deaf, but how many users do you need to tell you that their systems are broken ?

 

Also please notice that there's a difference in people not perceiving differences in general, and you - implying explicit listening and know what you are doing.

 

All together, when this would be your professional judgement against mine - I'm afraid you must loose. Maybe it is time for you too to start reading what this is all about. It just can't be denied, *plus* I sure know what I'm doing at tweaking the lot. Blame the hardware, fine. But then everybody's hardware except yours.

 

Earlier I suggested that all players are prone to these differences, except yours. I'll withdraw that now, because it's rubbish of course. You just don't hear it. Accept that and be happy indeed. But leave out the broken harware excuse as if you know all, while not working on it at all.

I do.

 

Do I care ? yes. Just because this thread is all about that.

 

"Hey Cookie, your hardware is always broken ! did you know ?"

 

Of course you didn't/don't say that, and it's not even in your context. So I'm making fun a little, while the underlaying subject is dead-serious and shouldn't be denied.

 

Peter