MP3 help

[nymph4]

I have two WAV File that are the same but one is 44.1 kHz Stereo at 16 Bit and the other is 22050 kHz Mono at 8. Bit

I put them both through MP3 Encoding at 98 kbps.

They both come out  638 kB in Size??

I don't understand this because I thought the WAV File that is 22050 kHz Mono at 8. Bit  already is a por quality so wen I put it throu the MP3 Encoding it should be even Smaller then the other.

Please explane???

[Cityscape]

I'm not quite sure that I understand what you mean.

638 kB sounds about right depending on the length of the track.

[nymph4]

What I meen is   one file starts out very High quality  44.1 kHz 16 Bit  Stereo  this is  CD  quality

The other file is 22050 kHz  8. Bit  Mono  ver pory quality.

Wen the both of them are encoded  the one that is very low quality to start with will cokme out smaller  RIGHT

[Cityscape]

Not necessarily. What is the durations of the 2 files?

[nymph4]

60 Sec.   why??

[rthompson80819]

The other file is 22050 kHz  8. Bit  Mono  ver pory quality.

I think you either are missing a decimal point in there somewhere, or meant Hz.

[nymph4]

yes I ment HZ

Both are  HZ  but one is very low  22050 Hz

[Salmon Trout]

mp3 size= channels x (bitrate x duration)

[Geek-9pm]

mp3 size= channels x (bitrate x duration)

The relationship expressed above  is valid. There are more factors involved that determine the exact size.

[Salmon Trout]

The relationship expressed above  is valid. There are more factors involved that determine the exact size.

I should have said that was for cbr (constant bit rate) mp3 files. In fact the bitrate is normally quoted for 2 channels together so a 128 kilobit/sec mp3 lasting 60 seconds will have a size of (60 x 128000) / 8 bytes

[Geek-9pm]

I wish I could contribute more to this thread. The dentist had sooth pullout  today  un eye  can hardly talk. Anyway, here goos.
Digital speech compression has fascinated me for years, long before the MP3 method became popular. For years the telephone company has used digital compression to reduce the bandwidth required on their long distance systems. Thus they could carry more conversations without expanding their equipment.
The modern MP3 format serves a similar purpose. Moderate compression in the MP3 format sounds good even on music. More intense compression tends to make this sound rather rough and produces many artifacts. An artifact is a strange sound that is not directly connected to the original sound in the musical or harmonic sense that our years are accustomed to.
Here is a very good book that you can read. Well, you can read the first few pages, which is enough to let you know that MP3 is much more sophisticated than the WA V.
The cancel below is recommended reading.
Understanding MP3: syntax, semantics, mathematics, and algorithms - Google Books Result

[BC_Programmer]

I wish I could contribute more to this thread. The dentist had sooth pullout  today  un eye  can hardly talk. Anyway, here goos.
Digital speech compression has fascinated me for years, long before the MP3 method became popular. For years the telephone company has used digital compression to reduce the bandwidth required on their long distance systems. Thus they could carry more conversations without expanding their equipment.

It wasn't really "compression" since all they did was strip out higher and lower frequencies.

Well, you can read the first few pages, which is enough to let you know that MP3 is much more sophisticated than the WA V.

a WAV file can be compressed via Mpeg layer-3.

FLAC is much better, overall.

[Geek-9pm]

It wasn't really "compression" since all they did was strip out higher and lower frequencies.

Digital compression. Can't you read?

How voice compression saves bandwidth
Gary Audin
The Public Switched Telephone Network (PSTN) started with the transmission of analog speech. This worked well for decades until the areas under city streets became saturated with copper cables, one copper pair per call. Starting in the 1950s, AT&T Bell Labs developed a technique to carry more voice calls over copper wire. They developed digitized voice technology through which 24 digital calls can be carried on two pairs of copper wire, thereby increasing the carrying capacity of the cables twelvefold. The voice is digitized into streams of 64,000 bps per call. The technology is called a T1 circuit and the bandwidth for the 24 calls is 1.544 Mbps. This worked well for domestic connections. The T1 technology then became the mechanism for long-distance domestic transmission.

Most of the early voice compression technologies were designed for undersea cables, where bandwidth was limited an...

The above shows that digital voice was in use before you were born. My reference is to the compression that is used after the voice has been digitized. Most often needed for long distance circuits. This research by Bell Labs was the precursor of other compressed digital audio formats that came later on.

[nymph4]

thanks

[BC_Programmer]

Digital compression. Can't you read?

Yes. MP3 does the same thing, and I'm fairly certain that the input is a digitized waveform. In fact, a good chunk of the waveform that needs to be compressed is stripped off by removing the sounds above and below certain wavelengths.

The above shows that digital voice was in use before you were born. My reference is to the compression that is used after the voice has been digitized. Most often needed for long distance circuits. This research by Bell Labs was the precursor of other compressed digital audio formats that came later on.

First: I'm tired of having to google for your references. you can't just paste in a blurb of text from god knows where and take it as fact. I did find this http://searchunifiedcommunications.techtarget.com/tip/0,289483,sid186_gci1239944,00.html

In either case, all it proves is that voice is digitized. hardly notable. the "compression" is simply the stripping off of higher and lower frequencies, leaving a comparatively narrow band (which is why hold music sounds so awful). This means that "space" that was used for sending the higher and lower frequencies  can be used for more calls.

They didn't actually implement a compression algorithm. If they did your ten second google would have surely found that.

[Geek-9pm]

In either case, all it proves is that voice is digitized. hardly notable. the "compression" is simply the stripping off of higher and lower frequencies, leaving a comparatively narrow band (which is why hold music sounds so awful). This means that "space" that was used for sending the higher and lower frequencies  can be used for more calls.

BC, you are out of your field of study.
Say anything you want about programming and I might believe you. But you knowledge of Digital Audio and Psychoacoustics is limited. Do you know what Psychoacoustics  means?
 
MP3 can take a digitized audio stream and compress it, either in real time or as a post process. It does not have to remove the high frequencies. It does not have to compromise the noise floor. It does not have to introduce distortions of any significant amount. It compress by a number of techniques, including, but not limited to
Linear Predictive Coding (LPC) Go look that up if you want.

The reference I gave earlier for the MP3 book is for the enlightenment of everyone. I can not quote it here because it is a new copyrighted book. And it does not have any short little blurbs that make sense out of the context. You have to red a few pages to get the jest of what MP3 really does.

Here is an experiment you can do at home. Make a recording of your own voice using the best stuff you have. Save it as a high quality WAV file. Notice the size. Now use a ZIP program to see if you can reduce the size. In many cases the ZIP program with use LPC to reduce it. Or try another compression method and see. Programs for DATA  compression use no loss or "loss-less" compression. MP3 goes further and will destroy some information. But the human ear can not detect it if done moderately. And it is not simply a matter for spectrum. MP3 removes redundant audio patterns not significant to the human ear

The OP  was about why is MP3 different that other digital formats. In is in the details. There is an alphabet soup of terms and concepts that take awhile to master. Sample rate in MP3 does not means always what you think it means.
I have made recordings at 44100 HZ and saved them in RAW PCM format. Then I converted them to MP3 with not change in the sample rate. But the resulting MP3 can be streamed at 128000 bps and sound very good. But it is still at the 44100 sample rate.

And don't ague with me, resistance is futile.

Go read a book!

[Quantos]

Do you know what Psychoacoustics  means?

You hear voices?

[BC_Programmer]

Do you know what Psychoacoustics  means?

yes (and this is without a google, I swear-) based on psycho (mind) and acoustics (sound) it's evidently the study of how sound is interpreted and heard.
 

MP3 can take a digitized audio stream and compress it, either in real time or as a post process. It does not have to remove the high frequencies. It does not have to compromise the noise floor.

Never said it did. nearly every single MP3 encoder in existence does this by default. only one that doesn't would be LAME, and even then you need to provide options specifically saying not to. Did I mention I wrote a crappy mp3 encoder? it barely works and only encodes at a specific bit-rate, something like 128kbps and I don't think I quite mastered some of the headers since it doesn't seem to play properly in all players (but it does work in some) and also it only runs in win 9x and I'm too lazy to figure out why. I think it was integrated into a little chat program I had at one point as well. My point is, you don't have to tutor me on what MP3 is. I know.

It does not have to introduce distortions of any significant amount.

yes. it does, anything less then 320kbps for me is immediately noticable, drum cymbals are the first to go. at first it's hard to tell- it just sounds "different"; but the CD's sound infinitely better- this is why I switched to FLAC. MP3 is lossless. Works fine where space is an issue (for example, MP3 players) but I'd much rather have the full range of sound with absolutely no loss

It compress by a number of techniques, including, but not limited to
Linear Predictive Coding (LPC) Go look that up if you want.

*censored* does <ANY> of this have to do with telephone lines? seriously? all I said was that telephone voice is transmitted not in a compressed form (which would entail the use of specific hardware on both ends to decompress the data, like a compressing modem does (telephones are not modems)).

The reference I gave earlier for the MP3 book is for the enlightenment of everyone. I can not quote it here because it is a new copyrighted book. And it does not have any short little blurbs that make sense out of the context. You have to red a few pages to get the jest of what MP3 really does.

Again. I know what it does, because I did it as well. It was a few years ago but that hardly means I don't remember any of it.

Here is an experiment you can do at home. Make a recording of your own voice using the best stuff you have. Save it as a high quality WAV file. Notice the size. Now use a ZIP program to see if you can reduce the size. In many cases the ZIP program with use LPC to reduce it.

Um, no. it won't, in any cases. ZIP uses Lempel-Ziv and Deflate (a modified LZ), not LPC. some other compression formats (maybe RAR, 7z, zipx) might have LPC, but Not ZIP.

Or try another compression method and see. Programs for DATA  compression use no loss or "loss-less" compression.

duh.

MP3 goes further and will destroy some information.

this the name LOSSY COMPRESSION...

But the human ear can not detect it if done moderately. And it is not simply a matter for spectrum. MP3 removes redundant audio patterns not significant to the human ear
[/quote]

That's exactly what I said. Omissions are not wrong, it's called brevity.

The OP  was about why is MP3 different that other digital formats.

no, they asked why two different PCM waveforms compressed down to the same size.

Sample rate in MP3 does not means always what you think it means.

Now you're saying you think I know what it means and can statistically evaluate that what it really means is not always what  I think it means?

But seriously, of course it doesn't. this is why I used to term "bitrate" explicitly. bitrate <essentially> (I have to emphasis this here, because apparently omitting anything means that I don't know about the omitted information) determines how much lossy compression occurs. a 96kbps file will always sound worse then a 320kbps file- except of course if they are both silence.

I have made recordings at 44100 HZ and saved them in RAW PCM format. Then I converted them to MP3 with not change in the sample rate. But the resulting MP3 can be streamed at 128000 bps and sound very good. But it is still at the 44100 sample rate.

And don't ague with me, resistance is futile.

I'm not arguing with you, and honestly I don't know who you are arguing with, because all your arguments contest things i never said or even implied.

[/quote]

[2x3i5x]

another question on MP3, what is the difference when you rip with J-Stereo vs Stereo? I have a CD ripper, which has these options, what is better to use when ripping a regular CD??

[Quantos]

another question on MP3, what is the difference when you rip with J-Stereo vs Stereo? I have a CD ripper, which has these options, what is better to use when ripping a regular CD??

That is a damned good question.  Many people notice that the lower frequencies(bass) is affected with j-stereo(joint stereo), most noticably that it is missing or muted.

Which one sounds better to you.  That should be the only thing affecting that decision.

[Salmon Trout]

yes. it does, anything less then 320kbps for me is immediately noticable, drum cymbals are the first to go. at first it's hard to tell- it just sounds "different"; but the CD's sound infinitely better- this is why I switched to FLAC. MP3 is lossless. Works fine where space is an issue (for example, MP3 players) but I'd much rather have the full range of sound with absolutely no loss

I think you meant "mp3 is lossy"? I thoroughly agree with you about the sound of lower bitrate mp3s. I read somewhere recently that a record producer said that these days so many youngsters use mp3 players as their primary music source that they are coming to prefer an "mp3 sound" and he said that it is becoming increasingly common to build this in to certain material e.g. chart-oriented singles.

Your remark about cymbals made me remember a term from my hi-fi buff days decades ago, when a term often applied to the sound of those instruments was "tizzy". This happened when something in the (entirely analog!) playback chain was affecting the upper treble - bad stylus tracking,  poor amplifier design, crappy tweeters, whatever. While Googling for that word, I came across this...

The treble characteristics we want to avoid are described by the terms bright, tizzy, forward, aggressive, hard, brittle, edgy, dry, white, bleached, wiry, metallic, sterile, analytical, screechy, and grainy. Treble problems are pervasive; look how many adjectives we use to describe them.

Learn the Language of Good Sound   Better understand the difference a hi-fi system can make   

http://www.crutchfield.com/S-U5HMY0aV9Dn/learn/homeaudio/introguideexcerpt.html

Although I still retain certain audiophile traits, I have never been inclined to go for the "hocus pocus" side of things - magic cables and the like, and I am not convinced that bi wiring makes much difference.

[Quantos]

I've worked as a professional audio engineer for a number of years before my hearing started to fail(common problem in that line of work), and bi-wiring is not something that will affect anything that the average user will notice.  The same can be said for gold plated connectors on your RCA cabling, unless you also have gold cable in between the connectors, contact resistance is still there.

MP3 however can be lossless, in the digital sense.  The only factor involved in ripping audio should be the users personal preference.

[Salmon Trout]

I've worked as a professional audio engineer for a number of years before my hearing started to fail(common problem in that line of work), and bi-wiring is not something that will affect anything that the average user will notice.  The same can be said for gold plated connectors on your RCA cabling, unless you also have gold cable in between the connectors, contact resistance is still there.

MP3 however can be lossless, in the digital sense.  The only factor involved in ripping audio should be the users personal preference.

I always supposed the reason for gold plating was to avoid oxidized metal surfaces leading to (probably imagined) partial rectification of waveforms... I remember in the 1980s a pal of mine who was much richer than I, who had got himself one of these newfangled CD players, and set it up alongside his Linn turntable with expensive Shure cartridge. He also had a very expensive FM tuner, (fancy antenna on the roof) all feeding a Naim amplifier and fancy Mission loudspeakers. He maintained that he could distinctly hear a difference between the same material on CD and vinyl (I do not doubt this). He said that he could detect a distinct "digital" sound from the CD player, that the effect was "cold" or "over analytical". I do not doubt this either. I preferred the sound of his CD player myself. I cannot remember what make or model it was. He told me that he disliked "digital" sound so much that he was spending more and more time listening to live classical concerts on FM, on BBC Radio 3 (an all-classical music station with a well-deserved reputation for high production quality). He preferred to live with the drawbacks of FM radio because of the lovely clear, warm analog sound. I was maybe cruel to tell him that for about a year the BBC had been using PCM links between the concert halls & studios and thence to the transmitters...

[BC_Programmer]

Also, it doesn't matter how much gold connector you use or high quality cabling if your encoding that signal to something like a 128kbps MP3.

MP3 compression is essentially the audio equivalent of jpeg- the amount of compression directly affects the loss of the original data, as that original data is "replaced" by approximations.

[Salmon Trout]

I think that you have to take into account the complex and subtle nature of subjective experiences. I could go on all day about this, but I shan't. I'll just allude to such things as the difference between e.g. sex with a prostitute versus making love to somebody you love, eating organic food versus factory farmed stuff, the reasons why homeopathic medicine appears to work for many people, the nature of the so-called placebo effect, feng shui, etc etc. Maybe many people's systems do sound better to them because they have spent a great deal of money on magic monocrystalline copper cables and gold connectors and a granite slab for their CD player. Maybe mine sounds good to me because I "know" I haven't wasted any money on "nonessentials".

I seem to remember a motherboard with a vacuum tube audio output amplifier, let's see if I can find it... ah yes a series of Aopen boards that used a Russian made double triode, a Sovtek 6922, equivalent to my old friend the 6DJ8 or ECC88... said to give a "warm, breathtakingly open" sound.

[patio]

Sweet ! !