Explain Compression
Apr 21, 2006 at 3:45 PM Post #2 of 15
As far as I know this term is used in three ways:

1. Compression (Lossy): parts of wave file that you probably won't miss are discarded to make a sound file smaller. Used for producing mp3s, wma, etc.

2. Compression (Lossless): like a zip file, the size of a wave file is reduced by storing redundant bits more efficiently, conceptually similar to 11111111 being stored as 8x1. Results in a smaller file but with equivalent information to the original wave. e.g. flac, alac.

3. Compression (Recording): the quieter parts of a wave file are boosted in volume to make the recording sound better on portable and low-end equipment. This has the undesirable effect of reducing the dynamic range of the recording, levelling off the higher peaks. This makes it sound worse on better playback equipment. Subjectively, this translates to a nasty plastic sound. Was used intentionally in some rock music, but now used almost everywhere. This is probably the usage you were referring to.
 
Apr 21, 2006 at 4:33 PM Post #3 of 15
Quote:

Originally Posted by seantx
I see this a lot here and other places. What does compression mean when referring to a recording? What are some examples of recordings that are compressed and recordings that are not?

Thanks!


Compressed recording: At the Drive In's RELATIONSHIP OF COMMAND. It's loud as hell, but the dynamic range is compressed. So if you look at a graph of the sound, you get one big fat bar, instead of lots of little zigs and zags.

Non-compressed recording: Steely Dan's AJA. Quiet parts are quiet, giving the louder parts a chance to really speak. The way it should be - the relationship between the "loud" and "soft" is just as it was in the studio.

For a modern example, probably Nirvana's IN UTERO, recorded by Steve Albini. He's known for not compressing things and leaving the dynamic range in tact. Though I'm sure in the mastering of the album, some of this effect was changed.

-jar
 
Apr 21, 2006 at 5:08 PM Post #4 of 15
Quote:

Originally Posted by Masonjar
Compressed recording: At the Drive In's RELATIONSHIP OF COMMAND. It's loud as hell, but the dynamic range is compressed. So if you look at a graph of the sound, you get one big fat bar, instead of lots of little zigs and zags.

Non-compressed recording: Steely Dan's AJA. Quiet parts are quiet, giving the louder parts a chance to really speak. The way it should be - the relationship between the "loud" and "soft" is just as it was in the studio.

For a modern example, probably Nirvana's IN UTERO, recorded by Steve Albini. He's known for not compressing things and leaving the dynamic range in tact. Though I'm sure in the mastering of the album, some of this effect was changed.

-jar


Mostly agree, especially about At The Drive In, yikes what a heavily compressed CD. But it's really a matter of degree since most recordings are compressed to a certain level so the quiet parts aren't lost, and rock music has always been pretty heavily compressed in the studio, including something like Aja. That's what gives it that nice smooth sound with some good punchiness. You really have to go to classical to find a wide dynamic range. But compared to most modern stuff, yeah Aja is amazing. Most old recordings are. Really sad how much has been lost.

But to the original poster, what accounts for all the talk about it nowadays (and for at least the last 10 years) is the resurgence of a bad trend that has run its cycle on occasion over the years, and that trend is based on a belief that the louder music sounds, the better people will respond to it, especially side-by-side such as on the radio or in a carousel player. And that is probably true and was followed by some companies even in the early days of records, some recorded so hot the needle would jump the groove, but with digital technology such as CD there is a hard ceiling to the maximum level you can master it at. So what is being done almost universally now is to boost the quiet parts and reduce the loud parts until they are almost the same, which gives you room to then boost the whole thing up in level, and thus make it much louder. And this has gotten much more sophisticated over the years with newer dynamic compressors that work psycho-acoustically, even going so far as to recognize things like when the singer stops so that parts of the background music can be boosted more. And oftentimes that isn't enough, so the "engineer" will boost the level beyond what the compressor settings will allow (without highly audible pumping effects), resulting in a lot of clipping or hard limiting of whatever peaks are left in the music. Needless to say, this whole process removes any naturalness the music had to begin with. In fact, that process was often used in the past on parts of music for certain unnatural effects, such as giving a drum track a punchy or distorted sound. And most singers have always used various compressors (along with preferred mic types and reverb settings) to shape their recorded voice. But this current loudness trend has run severely amuk and is ruining for posterity most of the rock music being produced today. Only small pockets of resistance left in the industry, like Albini that Jar mentions, and a few others. But even good guys like Albini sometimes lose control of their nice dynamic recordings when the artist or label takes possession unless the whole mastering process is supervised too. There was a recent Albini thread with many suggestions of nice dynamic recordings.
 
Apr 21, 2006 at 5:13 PM Post #5 of 15
Quote:

Originally Posted by Masonjar

<snip>

The way it should be - the relationship between the "loud" and "soft" is just as it was in the studio.

-jar


THAT makes sense. I've noticed before that music - especially modern recordings - has a sense of "on or off" in that you have no sound or all sound. IOW, there is little if any difference in levels between a splash cymbal being struck and a muted guitar chord. It also makes sense that this would sound better on lower end gear and at the same time be very annoying on higher end gear.

Bravo to you guys explaining something that is easy to hear but somewhat difficult to explain. Thanks.
 
Apr 22, 2006 at 4:44 AM Post #7 of 15
Quote:

Originally Posted by seantx
THAT makes sense. I've noticed before that music - especially modern recordings - has a sense of "on or off" in that you have no sound or all sound. IOW, there is little if any difference in levels between a splash cymbal being struck and a muted guitar chord. It also makes sense that this would sound better on lower end gear and at the same time be very annoying on higher end gear.


Exactly... the worst effect is probably on drum beats, which no longer have any "thump" but melt into the mix... the whole thing becomes one big wall of noise, very un-musical (except on car stereos perhaps).
 
Apr 22, 2006 at 4:47 AM Post #8 of 15
yo system be like, "oh shizzat dat biotch be too big!" then taht file be like, "damn son, why you frontin'??" then yo system gets all up in that grille and be like, "boy you best be knowin yo place 'for I bust a cap!" then the music gets all scared and crap, and then be like, "aight man chill yo" and it just shrinks all up in that mug.
 
Apr 22, 2006 at 4:47 AM Post #9 of 15
Jet - Get Born

Sounds bad even on bad equipment. It reminds me of 64kbps WMA
rolleyes.gif
 
Apr 22, 2006 at 4:47 AM Post #10 of 15
My first double-post in the history of the internet.
 
Apr 22, 2006 at 5:13 AM Post #11 of 15
Quote:

Originally Posted by Masonjar
Compressed recording: At the Drive In's RELATIONSHIP OF COMMAND. It's loud as hell, but the dynamic range is compressed. So if you look at a graph of the sound, you get one big fat bar, instead of lots of little zigs and zags.

Non-compressed recording: Steely Dan's AJA. Quiet parts are quiet, giving the louder parts a chance to really speak. The way it should be - the relationship between the "loud" and "soft" is just as it was in the studio.

For a modern example, probably Nirvana's IN UTERO, recorded by Steve Albini. He's known for not compressing things and leaving the dynamic range in tact. Though I'm sure in the mastering of the album, some of this effect was changed.

-jar


what do you use to look at a graph of the sound?
 
Apr 22, 2006 at 6:03 AM Post #12 of 15
Apr 22, 2006 at 6:07 AM Post #13 of 15
The last 3 flaming lips albums, Soft Bulletin, Yoshimi, and Mystics are all very compressed, super loud, and knocked way up and you can hear what it does to the top of the sound, it fractures and distorts the sound. Red Hot Chili Peppers too, Californication is so bad that I could hear it on my high school car stereo (which wasn't all that bad, but still...) It's everywhere, and once you've trained yourself to notice it, it ruins a lot of music. It's quite a shame. Most people don't care though...
mad.gif

Slade
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top