PDA

View Full Version : This is your brain on MP3's



ToSeek
2007-Aug-14, 08:51 PM
Producers howl over sound cut out by MP3 compression (http://seattlepi.nwsource.com/business/327319_mp3sound13.html)


But what is the price of inferior audio quality? Can poor audio touch the heart as deeply as better sound? John Meyer, who designs and builds some of the world's best speakers at his Meyer Sound Labs in Berkeley, Calif., doesn't think so.

"It turns you into an observer," Meyer says. "It forces the brain to work harder to solve it all the time. Any compression system is based on the idea you can throw data away, and that's proved tricky because we don't know how the brain works."

It could be that MP3s actually reach the receptors in our brains in entirely different ways than analog phonograph records. The difference could be as fundamental as which brain hemisphere the music engages.

"Poorer-fidelity music stimulates the brain in different ways," says Dr. Robert Sweetow, head of the University of California-San Francisco audiology department. "With different neurons, perhaps lesser neurons, stimulated, there are fewer cortical neurons connected back to the limbic system, where the emotions are stored."

Tucson_Tim
2007-Aug-14, 08:55 PM
Producers howl over sound cut out by MP3 compression (http://seattlepi.nwsource.com/business/327319_mp3sound13.html)



It could be that MP3s actually reach the receptors in our brains in entirely different ways than analog phonograph records.


I have a friend who swears that his old vinyl records sound so much better than even the best new digital sound.

tofu
2007-Aug-14, 09:08 PM
I have a friend who swears that his old vinyl records sound so much better than even the best new digital sound.

I've heard people say this, but I've never seen an actual scientific test.

Same speakers and amp. Switch between an LP and an MP3 without letting the listener know which is which. Ask them to pick the one that sounds the best.

I wouldn't be surprised at all if the people who say this are listening to LPs on expensive sound systems at home, then trying an iPod with ear-bud headphones in a noisy Best Buy store.

GeorgeLeRoyTirebiter
2007-Aug-14, 09:15 PM
I have a friend who swears that his old vinyl records sound so much better than even the best new digital sound.

IMHO, there might be something to that. I consider CD audio (16 bit 44.1kHz) to be the bare minimum required for decent sound. MP3s are a huge step backwards. They're the digital equivalent of the cassette- sacrificing quality for convenience.

zebo-the-fat
2007-Aug-14, 09:20 PM
MP3 compression will produce a degraded sound, but whether you can hear the difference will depend on the equipment used, the environment (ie noisy or quiet), the type of music and the listeners mood. All very hard to put numbers to.
The biggest problem with modern CDs is the fact that the are recorded with the sound heavily compressed and then the level set too high so that the peaks are clipped. There is no headrooom at all. I can't remember the website, but I have seen a page showing the levels of a CD from the dark ages (about 10 years ago) and a modern one, the old disk only had a few peaks near the maximum level, the newer one had almost all the peaks at maximum.

Tucson_Tim
2007-Aug-14, 09:25 PM
Well, my friend's old "stereo system" has some of the best old components and his speakers are good and huge. So it does sound better than the cheap setup I have at home! :)

Lurker
2007-Aug-14, 09:28 PM
Ahh... and there were those who worried that demon vinyl would be the death of music because it could never capture the nuance of live performance. Progress marches on... anyone who wishes can still consider live performances. Afterall, they are really the only true way to be one with the experience... :)

Occam
2007-Aug-14, 09:34 PM
I actually know something about this phenomena (there's a first time for everything). The reason many swear that vinyl sounds better than CD's or other digital music, is directly related to how our brains interpret sound. To simplify, take the same note played on a piano, violin and tone generator producing sine waves. The note is the same but they sound very different - why? There are not only subtle harmonics that come into play but noise and distortion itself that plays a part. A violin sound, for example, includes the noise of the bow across the strings, the resonance of the wood and varnish - hundreds of different factors that add up to what we interpret as a "warm" sound. Similarly, when we listen to analog recordings on vinyl, there are many additions to the sound that we don't consciously hear but they do add to the experience - the slight variations of speed in the turntable, the noise of the stylus upon the vinyl, the millions of tiny distortions caused by wear and tear. Unless this noise intrudes upon the music (clicks and pops for example) the noise actually enhances our experience, making it "richer" or "fuller". In fact, every sound we hear is a combination of countless subtle layers of noise, different timbres. The frequency range and clarity of a CD is far better than a vinyl recording but it is often described as "sterile" because much of that noise is removed. When we consider mp3's, there is usually noticeable compression as well, where frequencies have been clipped, or even removed entirely. An analogy would be the difference between hearing a person speak in the same room and listening to them on the telephone.

Personally, I always found extraneous noise to be intensely irritating and I'd find myself noting every little scratch, click and pop, instead of being fully involved with the music. Because of that, I much prefer CD's to vinyl but, once again, that is a personal preference that has nothing to do with superior quality of the original sound.

To sum up, noise and distortion is not always a bad thing. Our brains actually like to hear it and we miss it when it's not there. The finest audio system in the world does not compare to sitting in a concert hall and hearing a real orchestra play.

Tucson_Tim
2007-Aug-14, 09:34 PM
A little OT . . .

Have you observed how music medium for the home has evolved? First we had vinyl, which was direct access. (I know, there were media prior to this :)), then we had 8-track where you could only get within 1/4 of where you wanted to listen, then we had cassette where you could only get within 1/2 of where you wanted to listen, and finally we have CDs which is back to direct access.

mugaliens
2007-Aug-14, 09:35 PM
Producers howl over sound cut out by MP3 compression (http://seattlepi.nwsource.com/business/327319_mp3sound13.html)


Ask Ramone how it feels to hear his work on MP3s, and he doesn't mince words.

"It's painful," he says.

Well, sorry, Ramone, but most people can't afford the lovely sonics of a $500,000 recording studio.

As for MP3s, sure, a lot is lost in compression. However, unless it's massively compressed, most people can't tell the difference on iPod-type ear buds, period. The few that can may be audiophiles, but they're not 100% accurate, either.

Now - when I use my $150 MDRV700DJ Sony studio monitor headphones, I can always hear the difference. But that amount of weight on my head isn't conducive to mobile listening, and it creates a terrible power drain on an iPod.

However, using those same heaphones and listening to a nearly perfect LP on my 450 Technics turntable is "painful" compared to listening to it on CD, provided the digitization was done correctly (many of the earlier recordings were too flat, and thus too harsh, lacking in oversampling, etc.).

But hey - the headphone's response rate is 8Hz - 80,000Hz, which is very much well beyond my ability to hear, anyway.

I'm selling a solid magnesium, 65 lb, wire-drive turntable with an $1,100 needle along with my swamp land, if you're interested...

Tucson_Tim
2007-Aug-14, 09:36 PM
- hundreds of different factors that add up to what we interpret as a "warm" sound.

That's exactly how my friend describes it! Thanks for helping me remember.

ToSeek
2007-Aug-14, 09:57 PM
I'm selling a solid magnesium, 65 lb, wire-drive turntable with an $1,100 needle along with my swamp land, if you're interested...

Sorry, I only use titanium turntables myself.

mugaliens
2007-Aug-14, 09:58 PM
To sum up, noise and distortion is not always a bad thing. Our brains actually like to hear it and we miss it when it's not there. The finest audio system in the world does not compare to sitting in a concert hall and hearing a real orchestra play.

Amen to that, Occam. And a lot of audiophiles (I'm one, just not one of the primo snobby ones) don't realize that part of what they consider to be a stellar listening experience is simply due to the fact that their first listening experience with any particular recorded sound sets a pattern to which they'll try and recreate.

Current digital recorders oversample significantly, and maintain that throughout the mixing process. Some claim that produces aliasing when converted back down to 44 kHz, but others claim that sampling only at 44 kHz introduces distortions on every track, while recording and mixing higher evens out per-track distortions.

Regardless, there are digital techniques for minimizing (and eliminating the audible portion) of alising, which are commonly used in the field.

I mean, let's face it - I've got 192 kHz sampling rate on my home computer using FL Studio 7 and an Audigy card. When I output a 1m42s project to MP3, I can select upwards of a 32 bit floating wav depth (16.8), with a 512 pt sinc sampler interpolation, dithered, with a bitrate of 320 kbps.

320 kbps - how's that for oversampling?

And it's contained in the final MP3, which is, for that 1m45s MP3, a whopping 4.1 MB. Compare that to the 19 MB required for a standard CD cut using 96 kbps oversampling and other standard settings.

And trust me when I say the 1/4-sized MP3 sounds better, because it does, due to the far better settings.

On the other hand, I can use standard MP3 settings and the file is just half a MB (616kB), and using my studio headphones I can just barely tell the difference.

So to all the audio engineers out there decrying MP3, I'd say you have a bit (no pun intended) to learn about the technology, especially how, with the right equipment, you can have better quality sound with smaller actual files than with CD.

If the audio engineers wanted to make audio better, they'd simply jump on MP3 and use the higher settings.

One cat actually said he'd like to move to the audio equivalent of HD TV, probably not realizing that HD TV and MP3 are very similar as both are higher-resolution, compressed versions of uncompressed NTSC and CD audio.

Go MP3! (ie, the solution is here. Today).

mugaliens
2007-Aug-14, 10:16 PM
The biggest problem with modern CDs is the fact that the are recorded with the sound heavily compressed and then the level set too high so that the peaks are clipped.

I believe you're talking about dynamic compression, not digital compression. Companders are used to compress or expand the dynamic range so as to increase the audio available to the average listener, most of whom can't hear the lows in the 1812th overature and for whom the highs would shred their speakers if it were left in it's originally recorded dynamic range. By compressing the dynamic range, you bring up the lows and tone down the highs without loosing any of the audio signal itself (every waveform is still there).

This technique has been used in recording studios since the 1950s, because the dynamic range of vinyl is very poor compared to CDs.

quote]There is no headrooom at all. I can't remember the website, but I have seen a page showing the levels of a CD from the dark ages (about 10 years ago) and a modern one, the old disk only had a few peaks near the maximum level, the newer one had almost all the peaks at maximum.[/QUOTE]

Interestingly enough, in the early days of CDs, they did away with compression, but discovered the sound was way too harsh for most, as the lows would be almost inaudible while the highs would be way too loud. They lost sales on audio reviews until they reintroduced compression to reduce the dynamic range while preserving the audio content (and actually increasing the audio content that the listener gets to hear).

One of the most misunderstood things about audio engineering is that they do not try to preserve the original signal - everything is engineered, from removing the cough of an audience member from the original signal by close-micing the violins, to middle-micing the bass to bring out it's rich timbre, to utilizing dynamic compression, to altering the presence of each instrument by individually micing them to utilizing complex software to alter the music hall's characteric reflections, etc., so on, ad nauseum.

If you want audio that's original, you'll have to buy a ticket to the performance. If you want audio that's enjoyable to listen to in your living room, car, or iPod, you'll have to buy the CD.

tofu
2007-Aug-14, 10:17 PM
their first listening experience with any particular recorded sound sets a pattern to which they'll try and recreate.

I feel so sorry for all those kids who grew up listening to Super Mario Bro. on their NES. They are doomed to enjoy techno music all the rest of ther lives!


with the right equipment, you can have better quality sound with smaller actual files than with CD.

What do you think about ogg? I had heard a lot about it but never bothered to investigate. When I installed ubuntu for the first time, I noticed that it came with a program to RIP CDs into .ogg files and that my first time listening to them, but I don't have a sound system that would let me hear the difference.

Lurker
2007-Aug-14, 10:23 PM
I feel so sorry for all those kids who grew up listening to Super Mario Bro. on their NES. They are doomed to enjoy techno music all the rest of ther lives!

Not necessarily... my brother has made it a point to take his kids to live performances of many different types of music.

GeorgeLeRoyTirebiter
2007-Aug-14, 10:52 PM
By compressing the dynamic range, you bring up the lows and tone down the highs without loosing any of the audio signal itself (every waveform is still there).

That assumes the CD was mastered competently. I've seen quite a few that weren't they were compressed to the point of hard clipping.


If you want audio that's original, you'll have to buy a ticket to the performance. If you want audio that's enjoyable to listen to in your living room, car, or iPod, you'll have to buy the CD.

The problem is, the requirements for those different settings are incompatible. With a portable device, or in a car, a large amount of dynamic range compression is needed to make the music heard above the ambient noise. In my living room, I want as much dynamic range as my system can handle. Too many CDs come mastered for the former, not the latter.

tofu
2007-Aug-15, 01:21 AM
Not necessarilyI was making a joke, of course.

GeorgeLeRoyTirebiter
2007-Aug-15, 04:39 AM
This is getting OT, but there's even a Wikipedia page on excessive dynamic range compression in music:

Wikipedia: Loudness War (http://en.wikipedia.org/wiki/Loudness_war)

The phrase loudness war (or loudness race) refers to the music industry's tendency to record, produce and broadcast music at progressively increasing levels of loudness to create a sound that stands out from others. This phenomenon can be observed in many areas of the music industry, particularly broadcasting and albums released on CD and DVD. In the case of CDs, the war stems from a desire to create CDs that sound as loud as possible or louder than CDs from competing artists or recording labels.

However, as the maximum amplitude of a CD cannot be increased, the overall loudness can only be increased by reducing the dynamic range and distorting or clipping the waveform of the recording.

Maksutov
2007-Aug-15, 06:41 AM
Funny, back when my hearing was good, I would attend concerts (classical) and use my memory of those to calibrate my sound equipment. Typically I'd use recordings made at the old Carnegie Hall (before the renovation wrecked its wonderful ambiance). Back then the only thing one could really do was adjust the relative volumes of particular frequencies through a graphic equalizer. There was nothing one could do about the inherently-compressed dynamics of analog, especially with LPs.

But I had no interest in the "warmth" that analog distortion introduced into the recording. All I wanted was an accurate reproduction of what my ears heard during a live performance in the actual hall.

There was an overlap in the late 80s/early 90s where my ears hadn't yet gone bad, and CD recording techniques had finally progressed beyond the "Let's crank this sucker all the way up, it's digital so it can take it!" philosophy. So many of the classical CDs I acquired then were a long way from the first incredibly shrill DGG monstrosities. Instead they were very accurate re the fidelity of the sound produced. And they had something that was close to the dynamic range one experienced in the concert hall.

But then my hearing failed, the tinnitus took over, and I have no idea what the state of high fidelity music reproduction is like today.

Stuart van Onselen
2007-Aug-15, 08:26 AM
Just to note: MP3 is not the be-all and end-all of compressed digital music.

I've heard some snobs refer to MP3 as antiquated!

There are a number of newer, purportedly better, formats available, ones that claim less lossy compression due to better psycho-acoustic models, and others that demonstrate higher compression for the same quality of sound.

And then there are codecs that are completely lossless, if you want ultimate fidelity. The files produced are a little too large convenience for (current) flash-based players, but if you're playing from your PC's gigabyte hard-drive, space is really no concern!

Here's Wikipedia's comparison of audio codecs (http://en.wikipedia.org/wiki/Comparison_of_audio_codecs)

mugaliens
2007-Aug-16, 10:44 PM
That assumes the CD was mastered competently. I've seen quite a few that weren't they were compressed to the point of hard clipping.

The problem is, the requirements for those different settings are incompatible. With a portable device, or in a car, a large amount of dynamic range compression is needed to make the music heard above the ambient noise. In my living room, I want as much dynamic range as my system can handle. Too many CDs come mastered for the former, not the latter.

Both good comments, GLT.