MP3 Players Destroy Good Audio

A return to the golden ear argument. Well one of the reason that audiophiles will forever suffer is that I've found that they enjoy it. Why else would somebody nit pick the fidelity, stereo separation, stage accuracy and dynamic range of a recording instead of enjoying the sonic performance itself if they didn't relish misery. As horrible as MP3 squashed sound is, I've relearned to listen for the performer and not the engineer. But you are wrong that nobody has digitally matched vinyl's sound quality. Digital recordings can greatly exceed analog's capability, but most of the time the extra effort and arcane standards (24 bit playback and data-stream with reference clock stabilization) that appeal to so narrow an audience that financial investments never get a payback.

So if you truly want the best sound quality a world class musician can provide an audience, buy a concert performance ticket. Get rid of the engineer, the microphone, the speaker and sit in the same room as the performers with nothing modifying the sound between your ears and their instruments than air and room reflections. Anything short of this will always be a compromise.

When I was working on concerts microphones, speakers, and sound engineers were employed.

Many a great concerts have been amplified. Some concerts must be amplified. All amplified performances add an additional layer that can change the sound heard by the audience. Frequently the performers use the house sound man to operate the house reinforcement system. My point though is not that great sound cannot be found from an amplified sound system. My point is that only when a performer does not use an additional sound reinforcing system can a listener be certain that an unusual sound is precisely what the performer produced. When a golden ear hears a sound he/she dislikes, that maybe intentional.

Why would somebody nit pick 0-60 acceleration, cornering g capability, 60-0 stopping distance, top speed and comfort?

Because they matter if you are at all concerned with performance.

"I've relearned to listen for the performer and not the engineer"

Problem with that is you can't hear the performer, particularly their nuances, unless the engineer has done their job properly.

Thank you .. I couldn't have put it better ! I was a great beilever that digital can NEVER equal hifi vinyl performance... till a friend invited me to see his stereo system .. about $5000 worth, which he had brought lock, stock and barrel from NYC to Bangalore...and in a wood paneled long room with no furniture .. in a $1M house. After that, my music, which was sounding pretty good till then, suddenly sounded tinny and shallow.. well now my MP3 sound just as good.. after all, it is the music isn't it? not the bathwater after the baby is thrown out?

I like the way you think, but it doesn't always work that way. I recently attended a live concert in Symphony Hall - Boston. The violin solo was brilliant. However the location of my seat was such that the volume of the soloist was down a few db at my seat. The next evening I listened to the same concert on my FM receiver. The sound engineer had fixed the problem and the performance was much more enjoyable.

Thank you for the compliment.

Your exception actually proves my point. By hearing a violinist live another night you recognized that that night you found one of the few anti-nodes of Symphony Hall. Your observation of sitting in a null for the soloist was later confirmed by the sound engineer's rebroadcast. But in both acoustic presentations, by paying closer attention to the performer and not the room acoustics of your seat or the typically compressed dynamics of a FM signal you enjoyed the music.

That's precisely what Edgar Vilcher would have said, and it has some merit. I knew what the problem was as I listened to the live concet but couldn't change my seat. I enjoyed the performance anyway, but would have enjoyed it much more if I had been in a better location.

People who truly enjoy music for the nuance and virtuosity of the performers will still buy vinyl, but for who listen to popular music sound quality doesn't matter. Would we really care to hear the nuance of a top 40 artist? MP3's have made music more accessible for the masses and those of us who care will still find what we like, whether it be in concert halls or on vinyl.

All compression methods are lossy, there is no such thing as a lossless compression algorithm. The higher the compression rate, the higher the losses. This is all determined by what the algorithm is intended to do. There has been some terrible compression algorithms in the past which was rejected outright by the public. MP3 employs a variable compression algorithm which can be selected to do whatever the user wants to accomplish depending on what their intent is. MP3 can be anything from mildly lossy to highly lossy.

The sampling rate, above a certain minimum limit, has little to do with how much fidelity is retained. You can certainly use a high sampling rate with a lossy algorithm and still get lousy fidelity.

MP3 is certainly a reasonable algorithm to use for listening to pop music with limited fidelity ear phones, you won't get hi fi under these circumstances and I rather doubt anybody is expecting hi fi.

The proper application of MP3 can result in reasonably good fidelity, for example, with good noise processing, old 78 recordings can produce very good MP3 tracks with very good fidelity of what is left on the 78. The sampling rate varies with the quality of the reproduced audio, an 1900 Edison cylinder is not going to have much of an audio range on it even if it was in pristine condition, perhaps 2.5KHz bandwidth. It would be a total waste to record it at 128KB per second, it would add absolutely nothing to the original source.

Theoretically, 24-bit audio should provide the entire sound range of an orchestra or large pipe organ, in practice it may only come very close, the same can be said for other recording methods, the limitation of the dynamic ranges of the various pieces of gear used in the recording chain can limit just how close the final recording is to the 'live' performance. Microphones being only one of them. The dynamic range of a pipe organ can easily overload the best mics while challenging it to pick up the softest notes. The noise floor can easily render anything beyond 18-20 bits useless, it is even questionable if there is really any accuracy beyond 20 bits (1 PPM).

Basically all recordings are a compromise, some more than others. Can we expect to absolutely reproduce a 'live' performance on one's audio system....NO! Can we come very close, yes. Audiophiles seem to ignore the law of diminishing returns, no matter how much money one throws at the problem, they are not going to achieve perfection. As Redfred suggests, if you want to hear it 'like it is', the closest you can get is at the live performance. No one hears exactly the same as someone else, audio is truly in the ears of the beholder.

20 bits... 1 PPM... that's only 60db if memory serves...

I love my vinyl, and part of the reason I do, is because of my stylus, and my pre-amp. those two components were traditionally the most important aspects of any audiophiles system.

By the time I was buying my components, power amps were so well developed that they were all but 'noiseless' and had THD levels that were pretty much imperceptible. Naturally, the amp had to be driving a matched pair of acoustic suspension cabinets capable of handling enough power to do induction brazing under different circumstances.

All this went together with a 1/3rd octave graphic EQ, and a pink noise generator so the room could be rendered flat (1/2db) from about 120Hz - 15k.

Then, with the room volume at about 70dba (or greater) in the middle, it was like being able to walk around on the performance stage. If you do it right, you can hear the flutes breathing, and the violins elbows moving through the air!

How's that for an image? Too much??? Yeah, but ya know what I mean ^_^

Hmmm... I was thinking of:

dB=10*log(max/min) → 10*log(2²⁰/1)≈60.2dB {"point two" heh-heh ^_^ }

Regardless, I think I know where you're coming from. The *best* digitization of audio is certainly capable of capturing as much (or even more) of the dynamic range of the high end of analog systems.

But as you pointed out, the subtle flaws of jitter, aliasing, and that pesky 1/2 lsb error simply never go away, they just fade into insignificance and the warm tingle following that first Single Malt ^_^

I do honestly and sincerely feel that most mp3 encodings that are currently available are far less enjoyable to listen to than my vinyl. But on the practical side, and for most situations, if it can be reproduced in such a form that it is indistinguishable from the FM stereo in you car with a 5 band EQ, then it's good enough for most ears.

There is a time what I go entirely the other direction with mp3's -- audio books and lectures. I try to squeeze those down to 8 bits 56kbps mono; you know, telephone quality. It's wild to get 20+ hours onto a single CD... I used to commute 90+min each way.

The Light is Green!

Digitization reads the voltage of a signal, not the power of the signal. So the decibel calculation for voltage ratios is dB=20*log(V_max/V_min). Wikipedia has a good page on dB calculations.

I'm glad you liked my little synopsis of how digital corrupted itself.

Good Valve audio equipment may bring some of what you miss back. (read Tubes for Valves if you live in the USA!)

I personally find that the "clicks and clacks" of vinyl are not missed.

Vinyl plays well once, so I used to record to tape and listen to that.....The wear with a stylus is not good for the sound....

Playing/cleaning with liquids on the record are too fiddly for me.

I am personally happy with MP3 with good audio equipment......

Yes I agree MP3 is the pits.

It is a cookie cutter format, which abuses original nuances of raw sound profiles. Of course many having sub-woofered their eardrums will not ever again hear this subtlety.

There are lossless formats that deal with this but they are of the dreaded MS brand, .wma windows media player 9; allowing superior compression values without loss of content.

Try shn or flac - not MS and easy to work with once you get the hang of it. Of course they are larger files.

Windows media 9 is free too, the fidelity is better and files are smaller?? What's you're point?

You said "dreaded MS" so I took that to mean that windows media was unacceptable to you - I have no problem with windows media 9, but thought perhaps you didn't want MS, and nothing pointed about it. shn and flac are free as well:

http://etree.org/mkw.html

http://flac.sourceforge.net/download.html

"dreaded MS"

Oops! There go I using irony without clarity again

do you see how often the word believe appears in this thread, that's significant because this is a belief-based myth.

If you can measure the difference, and with today's technology you can, it is clear that digital recording can outstrip vinyl. Yes there are poor MP3 recordings and there are as many poor vinyl ones.

If you care about music you will seek out good recordings in whatever format. It's easy to believe this old story and first generation cds have embedded it quite well but it's about time it was put to bed.

oh and valve amps INTRODUCE distortion with their transformer output, yes they make the sound "warm" but that's not fidelity. They have their place warming the output of guitars and the like, but to reproduce that sound accurately you need semiconductors. Sorry.

You are correct, sir. But there's an interesting psycho-acoustic twist to the even harmonic distortion of tube and transformer. Since the cilia in our inner ear resonate in even harmonic groups. The added even harmonic distortion is not perceived as distortion until one well exceeds 5% pure even harmonic distortion. Often one must exceed 20% even harmonic distortion to be perceived a signal as distorted. But at less than 1% odd harmonic distortion, most people perceive a noticeable distortion. (I wish I could remember the paper and experiment that found this out, oh well.)

I can see your lips move but I can't hear you. Some one turn this thing down, Ky

Nice set of fog horns in the middle and the two pairs of Sousaphones and tubas are a nice touch but where's the rest of the orchestra?

Electronic Wiz

Sitting at home practicing/rehearsing on a acoustic steel string guitar. No gimmicks no nothing. Gives me the full range and I'm in charge not some freak show of technology. I am a fan of harmonics created by dampening or touching the strings just before or after being struck. When recording I need an extra mic on the fret to get that sound over. It practically disappears on mp3 recordings. What a waste, Ky.

PS; The unit in the pic is not and will never be my own. Was sent to me from a friend in Germany.

KY.....

I quite agree with you, for many purposes, such as yours, MP3 or most other encoded formats are of no use to you. Even the least amount of MP3 compression affects true fidelity but that is what it is designed to do.

bwire.....

I must totally disagree with you, there are no such encoding (compression) formats that are lossless, all compression schemes have some losses at the very least. That is the function of compression, to reduce the size of the recorded file, otherwise, there is no reason to do it. Only UNENCODED recordings are 'lossless' and that has qualifiers as well. Depending on the entire recording chain, there can and likely will be some losses (intentional or unintentional), even the best recording system has some small loss, usually of insignificant proportions but, none the less, there will be some small loss.

MP3 is an encoding format based on acoustical masking (perceptual encoding), in other words, a tone can be rendered inaudible by a tone of lower frequency, this effect of human hearing was discovered in 1894 and the first implementation of it was in 1959. Any number of algorithms are used to produce the encoding, all of which are considered quite 'lossy' and the bit rate at which the MP3 is recorded at further affects the fidelity of the end recording. The algorithm used determines just how drastic the 'losses' are and they can be very 'heavy handed'.

MP3 refers to the audio standard MPEG-1 audio layer 3 (hence MP3 for short), JPEG is the video equivalent standard. If the recording specifically refers to MP3, this means it must conform to that standard and use that particular algorithm.

Perceptual encoding works fairly well for the average listener but there is going to be a lot of people who will not like its effects. There are other encoding algorithms which are more acceptable and 'sound' better, these tend to be minimally 'lossy' with only mild compression ratios.

It comes down to whether or not a particular encoding scheme 'works' for your listening requirements or not, just like there are many opinions about what works and doesn't work in audio. What sounds bad to you may sound good to someone else, it is entirely in the ear of the beholder. You appear to like WMA encoding, that is fine, it works for you BUT it is still lossy, just a different kind of lossy which you appear to tolerate.

So, to answer you, yes, I do understand MP3 quite well and I also think that many of the respondents here also understand what MP3 is and does.

APD.....

IF the digital recording is done with great care throughout the entire recording chain, yes it can produce a superior recording to vinyl records but that is very difficult to do and is often short of the mark as the early first generation CDs proved, they have come a long way since those early days.

On the touchy subject of tubes vs. solid-state, yes they do have different distortion characteristics, solid-state having predominately odd harmonic distortion which ears are much more sensitive to but depending on the type of circuit, tubes can also have predominate odd harmonic distortion as well. However, both types of amplifiers can be made with sufficiently low distortion that few people can tell the difference between them. Double blind testing has proven this time and again. It all depends on the individual amplifier and the design. I have both types of amplifiers, designed by myself and both provide excellent sound, virtually impossible to tell which one you are listening to UNTIL you reach an input overload condition in which case it is immediately evident which type of amplifier you are listening to. Solid-state is much more harsh in its overload condition than tubes, an established fact, and the ears knows it. Stay within normal operating limits and you will have a very hard time telling the difference, both types are capable of excellent sound but that doesn't mean it is inherent to either type, bad sound comes in all flavors.

To some degree, I must disagree about tape recording, there exists many reel-to-reel tape masters which are of exceedingly high fidelity and very quiet, before digital recording techniques, they were the standard of audio quality and many of those recording still stand very well with today's recordings. You may be surprised to know that many of your cherished CDs are actually from reel originals, any music that was recorded prior to digital originals are from tape masters and there are still recording studios that insist on using reel recorders yet. Those 30 IPS masters are unbelievable in fidelity.

there are no such encoding (compression) formats that are lossless, all compression schemes have some losses at the very least. That is the function of compression, to reduce the size of the recorded file, otherwise, there is no reason to do it.

Please explain, I am simple and can't relate this to my own experiences, how is it then that when I zip data all the characters remain, where is this loss??

I didn't detail the nine preliminary protocols of the mp3 recording format nor approach the three finishing protocols and I am not at a loss; no pun, concerning the application of each in a audio production sense. I have found that the consumer perception of what mp3 actually is, has no relation to understanding of the format.The thread title is a good example.

For most the listening environments, automobiles, aircraft and those just wanting sound rather than silence, then MP3 is fine because back ground noise fills the blanks or the listener isn't really critiquing the audio.

I record and mix etc. using .wma and then after finishing switch to .wmv and use the audio compression within the video codec some may say I'm a cheatin some to get that 26:1 compression.

Bwire.....

No I wouldn't say you were 'cheatin' some, you've found a process which you like to use and there is absolutely nothing wrong with that. In audio, I've always said if it sounds good to you (or whom ever), then it has done its intended job. It doesn't matter if someone else doesn't like it, they don't have to use it.

Just for the sake of argument and accuracy, mathematically, it can be proven that there is no such thing as an absolutely lossless compression algorithm. In our example of audio for this thread, the 'lossless' algorithms have been optimized for minimal losses, compression primarily occurs through the elimination of redundant 'sets' of data. It can be proven that not all number sets can be compressed and in fact, an attempted compression may actually result in a larger data set than it began with. Sounds weird but that's the way it works. Lossy algorithms intentionally lose data based on the algorithm in use.

Audio compression algorithms are absolutely terrible at compressing text files but text compression algorithms are absolutely terrible at compressing audio files, it demonstrates the requirement of tailoring the algorithm to the specific job and what the intended use is, otherwise, the algorithm is of little use and can actually end up producing much larger files than the original one.

I can not specifically comment on your use of the Microsoft WMV compression algorithm as I am not familiar with how the specific audio data is compressed within the video. Obviously, according to Microsoft, the audio has already been processed using its WMA version 9 'lossless' algorithm before it is encoded with the WMV video algorithm. However, Microsoft does not seem to mention specifically what effects the video compression has upon the audio file. I would presume that the decompression algorithm does not produce significant data loss but, again, there must be some data loss, it is mathematically impossible to not have losses, they can only be minimized.

I applaud your clever use of the WMV format and apparent success of minimal audio error with a large data compression ratio. Personally, outside of recordings with restricted fidelity, I do not use any compression at all.

I used the example of data compression at the beginning of the thread as a vehicle to ascertain your knowledge oops! But it was good too to bring this into the open.

Thanks for kind remarks...

Once an audio signal is digitized, it becomes a series of 0's and 1's. Anything that can compress that type of data without losing any of it can be termed a "Lossless" compressor......the software does not need to differentiate, but naturally one "tuned" to the type of data is better and or faster.....

Its best to digitize at the highest rate you can for the best quality, but if you use a variable bit-rate, then the file will usually not be that much bigger than using a lower bitrate....experiment to see whats best for you.

I find that generally I can get around 10 audio CDs on an MP3 CD at 160 K/Bits per second, but only about 6 at 320 K/Bits per second. Naturally the length of the CD plays a role here, that is just my average.....I don't mind getting only 6 at 320.....

Good morning gentlemen, sorry for the delay in getting back to you.

Perhaps a little bit more clarity is needed as we appear to be in a slight misunderstanding of term use perhaps, lossless encoding is almost entirely based on arithmetic coding with Huffman encoding being a subset, IBM holding many of the patents in this field. All lossless encoding depends on the pigeonhole principle which basically states that no set of data can be compressed 100% and a totally random set of data is impossible to compress at all. This factor results in the fairly low compression ratios of lossless algorithms, generally less than 2:1, most often quoted ratios are on the order of a 1.5:1 to 1.7:1 reduction. Lossless algorithms, by mathematical constriction, cannot compress all of the data, thereby having bits of uncompressed data embedded within the compressed data. Only if the algorithm stays within these limits can one say that mathematically the data is essentially lossless. To some degree, how the algorithm is constructed and implemented has a bearing on whether or not it is truly 'lossless'. For most all intents and purposes we can call it lossless. This, of course, only refers to the mathematical algorithm itself and does not take into consideration any errors outside of the purely mathematical. Because of these constrictions, algorithms must be 'tuned' to its particular application as I mentioned before, otherwise it is a total waste to use it.

In order to achieve higher compression ratios, compression must become lossy, a mathematical fact you can't get around. Here we start getting into data loss, at a compression ratio of 3:1, it is rather mild and nearly no one notices any difference between the 'original' and compressed version.

Since this thread was primarily concerned with MP3 coding which is very heavy handed, with a minimum compression ratio of 11:1 at 128KB per second, we kind of got off the track about the main subject matter with lossless encoding.

http://en.wikipedia.org/wiki/Arithmetic_coding

The primary mathematical argument, http://en.wikipedia.org/wiki/Pigeonhole_principle, also known as the Dirichlet's Box Principal, provides the proof that 100% 'true' compression is mathematically impossible, reference the book, Discrete and Combinatorial Mathematics: An Applied Introduction by Ralph P. Grimaldi. Basically the pigeonhole principle states that no lossless algorithm can efficiently compress all possible data and completely random data cannot be compressed at all.

I think we are actually agreeing more or less with each other but with a slightly different interpretation of terms. In the end, we can agree.

Andy Germany.......

The numerical base, whether a version of binary, octal, decimal, hexadecimal or otherwise, has no bearing on the implementation of compression itself, while one may have a particular advantage in efficiency of computation and memory usage, it does not matter at all what base the data set is in.

Just to provoke ...

There cannot be a 'lossless' transform from analogue to digital unless the sampling rate is infinity ?

heh heh gentlemen, dont get me wrong, and don't bother to react.

A $50 MP player is ok for me, so don't count me.

Shall we then create a new thread?

"Audio Compression Principles, Objectives and Characteristics"

http://cr4.globalspec.com/thread/45190/Audio-Compression-Principles-Objectives-and-Characteristics

A really good post and the links were very informative.

Aa big Texas thank you for Germany!!!

You appear to be working under the delusion that perfection exists, it doesn't! Error correction circuits exist in computers to catch both hard and soft errors, that hasn't been eliminated yet. In many cases, the error rate is quite small but it is still there. The qualifier is what error rate is acceptable, what constitutes an negligible error. You are looking at it from the standpoint of storage and between bit error correction, etc., the error rates are pretty low now, they haven't always been this low. Data storage does not involve computation, it only stores data and does nothing else.

Compression, on the other hand is computationally intensive, some algorithms more than others. 100% compression is impossible, 100% lossless compression is impossible because lossless compression cannot compress 100% of the data! Compression does not operate on just '0's and '1's individually, it operates on blocks or sets of numbers. You cannot compress a '0' or a '1', it must be a set of numbers. Lossless compression operates on the subsets of data which can be represented by a smaller set of data presumably without an error occurring when it is decompressed. A large amount of the data sets in any given 'file' cannot be compressed at all without significant error, therefore lossless compression does not touch these sets or blocks. In general the error rate of lossless compression is quite low but not zero! It is of a insignificant amount unless the algorithm is not doing a good job of it. It must be remembered that compression is a two way street, what was compressed must be decompressed and that adds another layer with possible errors as well.

The above fact is why lossless compression cannot produce large reductions in file sizes, ratios of 1.5:1 to 3:1 is about the limit, anything beyond that is going to increase the error rate or loss, whether or not this loss of data matters is solely up to the user of the compression method, since we were discussing MP3 lossy methods, we are talking data loss with no ifs or buts.

Technically, since the function of compression is to compress data, it is an error when that algorithm cannot compress any said data sets. Yes, I am being picky, I am stating things as a fact of reality, nothing is perfect, including lossless compression. I did not, however, state that those errors made lossless compression unacceptable, I did say that "for all intents and purposes it was lossless." You seem to forget that we are discussing audio here and a very lossy compression method and what constituted an acceptable rate of compression loss and that is what I had been speaking to before the subject of lossless compress entered. What one person hears, another one can't or hears it differently, audio is very subjective and a very personal experience period, otherwise we would have just one great amplifier and speaker type selling all over the world because everyone thought it was perfect and we all know that isn't true.

Most of your remarks regarding my knowledge is not only very presumptuous but also quite wrong and you should refrain from making such remarks when you do not know anything about the other person, stick to the technical remarks and leave out the personal. I could argue against several other of your remarks but that is also off the main subject so I will refrain from doing that.

I consider this subject closed now and do not intend on further remarks as they do not pertain to the main subject matter of this thread.

I bid you a good day sir.

Sadly, you have made some further serious, erroneous and basic mistakes, starting with your personal take on error correction for example........please post web links that support any of your claims, I would be highly amused to read them....

In some of your claims/ideas, you are just far enough away from the correct info to probably hear the comment "close but no cigar!" Its as though you took the training, but only attended 90% of the course, the missing 10% being VITAL to fully understanding the whole of course!!!

For your information, error correction is used when there is a case of bits/bytes being unreadable from the digital storage medium, it allows the missing bits to be recreated EXACTLY as they were before, not close, EXACTLY the same, or a read/data check will be posted for any uncorrectable errors discovered...........

If you really wish to learn something new, please read page 15 of the Hewlett Packard .pdf at the following link:-

http://h71028.www7.hp.com/ERC/downloads/5983-0148EN.pdf

On the second line you can read about "all vendors must use the Advanced Lossless Data Compression (ALDC) algorithm for data compression."

This is a tape example, but in fact similar Lossless algorithm's have been around for digital media of all types for many, many years......

Seriously, I personally doubt though that there is enough time left to educate you and convince you of how wrong you really are, before our Sun expands, consumes all the planets before becoming a red dwarf.........in about 4 Billion years if I remember correctly (give or take a day or three.....)

May I reiterate that you do need to improve your knowledge DRAMATICALLY before making such a large number of false or inaccurate statements on CR4 or in fact anywhere......especially without any links to websites that support your theories. You will notice that I have always supplied such links, so that you could easily prove my statements for accuracy......why don't you follow such a simple method? Its easy when you know how!!! Perhaps links supporting you are few and far between?

As the saying goes, "You can lead a horse to water, but he has to drink all on his own!" and I am certainly giving up completely on you as of now.....enough is enough!! I have better things to do with my time than trying to correct your misguided thinking....

On a more personal note, may I suggest you modify your CR4 name as a "Wiz" in some English speaking countries means "having a Piss or going for a leak".......not being a wizard!!! but it does fit quite well from another point of view...... but of course I leave the final choice of your name completely up to you......I just wanted to make sure that you were not left ignorant on any other points as well.....we don't want anyone laughing behind you back now do we?

Thank you Andy for contributing to this subject, I too have experienced the lossless compression values and enjoy digitized audio greatly.

I'm a fan of windows media player finding it already embedded into the windows platform and understanding the plethora of trials confronted when the media was configured it is quite a piece of work. It never ceases to amaze me how often the significance of this piece of work has had in the Audio/Video industry and most don't know the standards by which the media they see and hear has MS tech written all over it.

Though it is not my intention to toot the MS horn it is the package of codecs I use for everything without disappointments. The compression attributes prior to the release of Wm9 were handy then Wm9 voice is very helpful but the lossless feature allows extemporaneous capture and compression of the full audio spectrum.

As the topic of this thread is the subject of good audio (what you hear—algorithms aside) I can attest to the accuracy of high bandwidth, sample rate recording using a lossless codec and we may produce such that most discriminating audiophile wouldn't realize it weren't an analog type processing procedure. I feel the need to use the term processing so not to be criticized of analog misinterpretation as any audio used in software is without analog constraint.

A comment was made early in the thread of one wanting to eliminate the frequency adjustments made by sound reinforcement personnel or equipment have serious misunderstanding of the manner by which their sound gets to their ears.

Other comments made by knowledgeable individuals have not contributed towards an understanding of the subject, there is a point when dropping semantics and rhetoric enables one to listen just to the music.

Thank you for your support and for the detailed info you supplied with regard to "Lossless" compression and usage of..

The numerical base, whether a version of binary, octal, decimal, hexadecimal or otherwise, has no bearing on the implementation of compression itself, while one may have a particular advantage in efficiency of computation and memory usage, it does not matter at all what base the data set is in.

That is basically what I said and I said it first!!! But it STILL starts as a series of 0's and 1's.....as I have also mentioned twice (at least) already.....that cannot be changed, its a fact.

You get too picky.....and don't read things twice (or more if required) to fully understand what or someone else was saying.....before diving in at the deep end.....

We are only talking about digital data here, tell me, is there another way of representing digital data in a modern PC or similar other than a series of bits?

If yes, I would be a) most interested in how it works and b) is it used in either lossy (MP3 for example) or nonlossy music compression....eg. does it belong here at all?

Thanks in advance.

Lossless audio compression produces a representation of digital data that can be expanded to an exact digital duplicate of the original audio stream. This is in contrast to the irreversible changes upon playback from lossy compression techniques such as Vorbis and MP3.

The End.

Well put Sir!!

Many thanks.

Excellent and valid points always earn a GA from me....

I need to correct a typo, the first implementation of encoding hardware was in 1979, the encoding curves were actually developed in 1959....oops.

My brother-in-law has a very nice system that incorporates the best of both worlds as it uses both tubes and solid state semi conductors.

It is a right expensive system that is further independently grounded with a spike out in the yard.

I myself was fortunate at some point to have had my hearing tested, so I am aware that I actually do not even have ears of equal quality. I had Scarlet Fever as a youth and am left with one ear that hears highs, and another that hears lows.

If I remember correctly the range of human hearing is from 30 to 20 thousand hertz, so the aim of a recordist is to supply that range.

I am under the impression that MP3s are not capable of providing that even to amps and speakers capable of providing that range.

Few people over about 16 years old can hear beyond 16kHz. Very few.

People of my age are happy if they can hear 12 khz......I am 63. Untested but I doubt that I can even hear 10kHz nowadays...

99.9% of a music signal content is probably less than 6khz....

Too much is made about harmonics outside of the normal hearing range I feel......personal opinion only.....

Many years ago, I made copies of CDs onto good cassettes, with a top cassette recorder, NOBODY could tell them apart.....simply nobody......music buffs and amateurs.

I played them in parallel on the same system and switched back and forth......when asked.......but nobody could tell which was which with any degree of accuracy.....there were many "experts" among them who were 100% certain they could tell a difference......before they tried.

Much "waffle" is written and said on this subject......well its something to talk about, but probably less than 1% of the population are able to differentiate, or less. Then far less are even interested in the difference. But the theories are well sold.......!!!

For most people, they simply "IMAGINE" that they can hear a difference......but when push comes to shove, they cannot tell any better than anyone else.......

Quite true, but many find this sound better as that is what they were used to many years ago.

I am not one of them by the way......I prefer quality with semiconductors!!

In Europe its big business selling valve amps/preamps......

Way back when, I bought a CD player and was so impressed that I sold my vinyl collection: big mistake. As I have collected over the years, I have bought CDs that sounded so good that it would bring tears to my eyes, and others that sounded like they had been mastered off of an old 8-track tape that had been played about 10,000 times. Generally, they seem to lack the mid range strength of the vinyl, but the bass and treble are astounding. There are things that I can hear on the vinyl version that I can't hear on the CD, and vice-versa.

That having been said, I listened to my daughter's MP3 once; that was all I needed. Now I just go downstairs, fire up the Onkyo, and let the Cerwin Vegas sing to me.

Why aren't you using shn or flac files instead of mp3 if you want lossless? Oh, you don't want to have to bother to rip them yourself! The vinyl to tape idea is great if you can stand the hiss of tape, but vinyl to shn or flac is the way to go if you want to hear everything that was there to begin with and still be able to listen on a computer or portable player (not on an ipod though, but why use an ipod when there are more versatile options? oops, there is another topic).

I have the distinct impression after reading the posts, not a one knows what MP3 is... sad.

Everyone here is going on about the quality but lets remember that the main reason to use mp3 is to put the music into a format that allows a great amount of portablity!

If you want great quality go listen live!

I use mp3 because i want to carry music round with me. I remember my first walkman and having to carry tapes round. Then minidisc smaller but still have to carry extra bits round if you want any kind of range.

Now i carry an ipod nano around, it has thousands of tracks on it and videos as well. It sounds perfectly fine to me. I play a whole range of things on it and its fine. It will never win any awards for high quality but thats not what it is designed for.

mp3 is not designed to be high quality but for most of us it is perfectly fine for doing the job that it was designed to help with!

I suppose with IPod you're stuck with mediocre mp3 but with Zune you can have .wma true lossless audio it's what you're willing to settle for I suppose