All about digital audio

[Neutered Sputniks]

After a 1 1/2 phone convo with Nick today in which we discussed just about everything related to audio quality and builds, I was doing a bit of looking around for comparisons between the CD7200mkII and D880-PRS. Lo and behold, I stumbled across this that directly related to part of our discussion - use of Ipods (or other MP3 players) in car audio.

Anyone thinking about using an Ipod should definitely check this article out (I've copied the first bit that really starts to explain why you can't use MP3s for HI-FI systems).


16 vs. 24 bit Audio Recording Demystified

Originally Posted by teh article

Bit Depth refers to the number of bits you have to capture audio. The easiest way to envision this is as a series of levels, that audio energy can be sliced at any given moment in time. With 16 bit audio, there are 65,536 possible levels. With every bit of greater resolution, the number of levels double. By the time we get to 24 bit, we actually have 16,777,216 levels. Remember we are talking about a slice of audio frozen in a single moment of time.

Now lets add our friend Time into the picture. That's where we get into the Sample Rate.

The sample rate is the number of times your audio is measured (sampled) per second. So at the red book standard for CDs, the sample rate is 44.1 kHz or 44,100 slices every second. So what is the 96khz sample rate? You guessed it. It's 96,000 slices of audio sampled each second.




So lets put it all together now. This brings us to the Bit Rate, or how much data per second is required to transmit the file, which can then be translated into how big the file is. Your CD is 16bit, 44.1 so that is 44,100 slices, each having 65,536 levels. A new Audio interface may record 96,000 slices a second at nearly 17 million levels for every slice. If you think that is a lot of data, well, you are right, it certainly is. The Bit Rate is usually expressed in Mbit/sec. But you don't need to do all this math. I'm going to do it for you. This is not an important area in the recording process to get sidetracked on. What is important for you is how this translates to your hard drive storage.


16bit 44,100kHz 1.35 Mbit/sec
16bit 48,000kHz 1.46 Mbit/sec
24bit 96,000kHz 4.39 Mbit/sec
mp3 file 128 k/bit rate 0.13 Mbit/Sec

So you see how recording at 24/96 more than triples your file size.

[Electrodynamic]

Thanks for posting this John! I was floored when you told me about what really goes on with "high fidelity" mp3's vs. a normal CD. I knew there was a big difference, but I didn't know it was THAT much of a difference (concerning bit rates). I knew I loved the way CD's sounded but couldn't figure out why since, according to most people, a 320k mp3 was "close" to CD quality. After you read what you posted you can see why it's no where close.

[Neutered Sputniks]

I'm amazed at how many people don't realize the match involved here. Or recognize how many bad MP3 conversions they've downloaded or even made themselves...

[Electrodynamic]

I made a ninja edit.

[Neutered Sputniks]

So did I

[soydeedo]

First off, most rips today are not done at 128kbps. Most are done at V2 or V0, which are variable bit rate settings that adjust according to how much audio data is present at any given time interval. For simplicity we can generally say that V0 rips are equivalent to 320 kbps [.3125 kbps]constant bit rate files. And since we're comparing with cd's, we're looking at the 16bit, 44.1khz bitrate of 1.35 mbps, there is only a 4.32 fold advantage for cd's to begin with, not 10 fold.

Second, you're completely discounting the fact that there are compression algorithms in place to convey they same data in less space. Now, mp3 is a lossy format, so it's not extrapolated to become exactly the same data, but it makes sacrifices in areas that are generally not audible.

Also, cd's and other raw file formats fill space for every frequency regardless if there is audio information present or not. So you could have a cd of pure silence and still have the same bitrate. MP3's eliminate this enormous waste and only articulate those frequencies that carry information. That is why lossless codecs like FLAC can often generate file sizes that are half [or lower] the size of the original. This part of the process is just like when you make a zip file.

So if you want to compare bitrates, you should do it with FLAC vs MP3, but even then the compression algorithm of MP3 has evolved so much it's not even a fair comparison. It'd be like comparing MPEG2 from DVD's to h.264 from Blu-ray's or even xvid. You can fit more information in the same amount of space with better algorithms. Yes with the video and audio codecs there will not be 1:1 copies, but they are configured in such a way that the distortions do not significantly alter [yes I know this is subjective] the integrity of the original recording.

Anywho, if you want the absolute best, of course stick with your cd's. There will be some degradation moving to lossy formats, but in general it isn't noticeable. Another thing to note is that I don't know much about Itunes' MP4 or Microsft's WMA formats and how well their compression schemes work, so they might be worse. My experience is mainly with MP3's. One last note - Amazon.com uses 256 kbps constant bit rate MP3's, so they are also not a good comparison to homemade rips.

Disclaimer: I might not have every technical detail correct here.

EDIT: Geez this is long. Thanks for getting me worked up, fellas... =P

[soydeedo]

Also, even different MP3 encoding programs will produce different results. The most popular encoder is called LAME, and my comments were based on the evolution of that encoder. I'm certain there are less efficient/faithful MP3 conversion suites out there.

[Neutered Sputniks]

We're not just talking about what you'd carry around on your computer. Regardless of what you rip the music to - you're still going to have major loss of data as soon as you put it onto your Ipod or Flashdrive due to the restraints those particular devices have in conjunction with your headunit.

I have a track that will beat any algorithm out there. I have tried multiple different rips at different bitrates and settings (including WMA lossless) and the track STILL skips because that's how encoding works. It has to eliminate data. And the way it does that is by deleting a certain amount. Regardless of if it's white noise or not, all those frequencies combine (esp when played through a high end system) to create the sound you hear.

Regardless, we're talking about using a low-quality compression to feed a high-fidelity system. You're going to notice loss. You can't lose approx half the sound data (even if it's supposed to be "inaudible" or "white noise") and expect to have the same quality of sound reproduction. Remember, your $$$$$ stereo can only reproduce that which it receives - in programming language, Garbage In Garbage Out.

[soydeedo]

I already stated that there would be losses and that if you need the absolute best then of course you should stick with originals.

I don't believe the part about a lossless encoding losing data, though - that goes against the very definition of a lossless codec. Maybe the WMA implementation is fundoozled? Lossless codecs do not eliminate data, they eliminate data redundancy. Why can you zip text files and unzip them again to generate the exact same original file? It's the same concept except done on the fly with FLAC. I don't know the specifics, but I think it's something like "repeat xHz at x amplitude for the next 30 cycles" and then when you play it back it decompresses automatically to it's fully fledged high fidelity glory.

Anywho, I was mainly responding to this when I replied:

...according to most people, a 320k mp3 was "close" to CD quality. After you read what you posted you can see why it's no where close.

I just wanted to make it clear it's not as far off as that original article made it seem. It isn't perfect by any means, and there will always be exceptions that sound horrible, but usually it's pretty faithful.