Music Into Noise: The Destructive Use Of Dynamic Range Compression part 2

Not sure if this has ever been linked to, but it's also a good read:

http://www.chicagomasteringservice.com/loudness.html

(Chicago Mastering Service is co-owned by Bob Weston from Shellac)

How long will it take for people's computer hardware (including smartphones) to be of such a quality that Spotify may become all lossless? It would be a very important change because it would mean the kids would get used to more dynamic sound again.

YOU'RE STILL DERANGED

yeah, you can losslessly hear how the recordings sound like ass soup.

i always forget that this is a thing that's meant to exist until these threads are bumped

have never noticed it while actually listening to music

How does lossless/lossy encoding make sound more or less dynamic?

it doesn't

You mean Geir might be wrong sometimes?

What's your usual listening setup lex?

tbh given what lex listens to, it's probably the audio equivalent of those fish that live at the bottom of the sea and can thrive under thousands of pounds of pressure

have never noticed it while actually listening to music

― lex pretend, Wednesday, October 12, 2011 10:27 AM (5 hours ago) Bookmark Suggest Ban Permalink

wake up sheeple.
shit is hella obvious

what is the singular of sheeple?

sheeperson?

what, ewe don't know?

How does lossless/lossy encoding make sound more or less dynamic?

The more compressed, the less difference between the loudest and most silent parts. Because the most silent parts become less silent. Extensive use of dynamic range compression is just like USING CAPS LOCK ALL THE TIME.

data compression is not the same as audio compression

It has much of the same effect on the audio. Today's recordings are being compressed because the kids are used to hearing badly compressed mp3s and expect music to sound that way.

you do not know what you are talking about

Geir, you're confusing low-bitrate-encoded files with brickwall-mastered audio.

They can both sound bad, but neither necessarily so. There's some truth to the idea that kids are being trained to accept poor fidelity by both low bitrates and extreme (distortingly so) audio compression, but this may only be a passing phase in the grand scheme of things. Data compression is just the natural result of low bandwidth and storage limitations, but those are receding by the month. And audio compression has, in part, come from the fact that everybody listens to music through chintzy headphones with high impedances that require a high average level (the "RMS") to give a listenable level. Which should change, too, if more slowly.

Things that are well mastered can still sound pretty good even at 128kbps. Depends on the music involved. And many things sound fantastic when the audio/dynamic range is seriously compressed. There's no single rule to cover this. I'd say that more important than data or audio compression is EQ—whether the piece in question is well balanced across frequencies. If it is, it can be pushed pretty hard, just as a well balanced car can be driven harder than one that's off kilter.

Of course, a lot of music these days is marred by both too much data and audio compression.

What's your usual listening setup lex?

at home, mp3s through laptop plugged into this stereo system (sony ss-cpx333)

http://t1.gstatic.com/images?q=tbn:ANd9GcS7iBmSTn8Dx2qEY2BvOSit-Fp2BSoIIdNHp1rgj_TnCq3MHSZODiQpYmPQ

while out and about, ipod with these headphones (sennheiser px100)

http://www.dansdata.com/images/3senns/px100640.jpg

tbh given what lex listens to, it's probably the audio equivalent of those fish that live at the bottom of the sea and can thrive under thousands of pounds of pressure

what are you saying about sade??

i love how deej can come out with "it's obvious" despite the overwhelming evidence to the contrary

no, audio compression is entirely audible. download the greatest hits version of sugar ray's 'every morning' from amazon, then download the album version. they remastered for the greatest hits. the difference is entirely noticeable

What evidence to the contrary? Once you know what limiting sounds like, it's impossible not to notice it.

One track that always comes to mind for a really really horrible brick-wall job is Girls Aloud "Can't Speak French". Listen to the instruments in the background and hear the way they are constantly pulsing in and out when the drums and vocals push them into the background, only for them to come steaming back in afterwards. I quite like the song, but listening to it just gives me a headache.

download the greatest hits version of sugar ray's 'every morning' from amazon, then download the album version.

why in god's name would I do either of these things much less both

zing!

Ignoring the choice of music, this video gives a good demonstration:

https://www.youtube.com/watch?v=3Gmex_4hreQ

Geir's thoughts are like imagining that changing a richly-formatted Word doc to a plain .txt file also changes it to all uppercase.

man i love old analog word docs, the mids in the punctuation is so creamy

I only listen to mimeographs now.

that'll make sense that you don't hear the effects of highly compressed music. it's designed to bang on consumer hifi stuff. a lot of music relies on that clipping-in-ur-face thing

i don't have a problem with music that's mastered too hot, i just wish there was an alternative

has anyone come across any labels offering 96k 24bit audio files? it'd make sense for techy dance stuff aimed at djs who are going to stick in some software and play it out

xpost

yah my friend just faxed me a bunch of old jug band mimeos from the 20s, great stuff

Ignoring the choice of music, this video gives a good demonstration:

http://www.youtube.com/v/3Gmex_4hreQ&fs=1&hl=en

― Chewshabadoo, Thursday, October 13, 2011 12:27 PM (23 minutes ago) Bookmark Suggest Ban Permalink

uh that video has it wrong...he's just making the track louder, not compressing it to do so, which is missing the point. the dynamics remain intact with what he's doing.

One track that always comes to mind for a really really horrible brick-wall job is Girls Aloud "Can't Speak French". Listen to the instruments in the background and hear the way they are constantly pulsing in and out when the drums and vocals push them into the background, only for them to come steaming back in afterwards. I quite like the song, but listening to it just gives me a headache.

haven't heard the song, but this sounds like it might be intentional use of sidechain compression at the mixing stage, not necessarily the result of the mastering job?

even if it is sidechain compression, it is still a symptom of the need to fill the dynamic range to get your music to stand out, just at the mixing stage not the mastering.

Audio compression = a typed document with less and less spaces between words and no punctuation so that there is very little variation between lines and it becomes difficult for the reader to stop reading once they start.

Data compression = a typed document that uses as little printer ink as possible but still allows the text to be read.

24bit 96k files are becoming more popular as people want something that sounds better than CD and now that buying a few terrabytes of storage is affordable. However it is very much a specialst market with a small range of albums. HDtracks offer the most well known stuff, but I have been hearing that they often take the left and right channels of a surround mix, so you end up with less instruments in much higher quality.

Holy shit on the RMS levels of the 1997 remastered "Search and Destroy" from that Chicago Mastering Service article.

i know! that article was really cool

i don't know why anybody would be proud of not being able to hear this sort of thing. Being tin eared isn't a positive

even before i knew the science behind it, even back in my formative musical days, i just knew that some albums were really hard work to get through. It's a physiological sense as much as a facet of your hearing

I'm not sure if you're trolling or not Jordan, but if you think those two clips sound the same you've either got cloth ears or cloth speakers.

i was mostly watching/skipping around the video before to see what he was doing, but i just went back and listened through the whole thing.

the audio examples are good illustrations, yeah, but i was thrown off by what he's doing with the waveforms. it looks like he's just dragging up on the waveform, which in any DAW just increases the gain (turns up the volume), and he doesn't mention (or visibly apply) compression at all. not the most clear visual representation of what's going on imo.

why in god's name would I do either of these things much less both

― wrestlingisreal420 (crüt), Thursday, October 13, 2011 12:20 PM (6 hours ago) Bookmark Suggest Ban Permalink

song is dope fuiud

Was listening to Braids' "Lammicken" yesterday (dope song fuiud etc.) and was thinking of this thread.

Song is this kind of slow-mo 4/4 pound with female vocals like a less songy version of Glasser's "Mirrorage". Anyway when there are vocals the kick drum recedes and then when the vocals cuts out the kick drum gets massively louder and more intrusive.

I have no idea whether it's a deliberate effect (if you assume it's deliberate it totally works) or if it's just a result of dynamic range compression. Possibly it's even a combination of both, like they noticed what was happening with the kicks and then decided to accentuate it.

What evidence to the contrary? Once you know what limiting sounds like, it's impossible not to notice it.

the fact that only a tiny minority of music fans notice it?

Consciously notice it. I suspect its subconscious affects are pretty widespread.

you don't think people notice that new releases generally sound louder than old releases? I noticed it before this whole thread/topic came about.

I don't tend to notice corrective autotuning but I accept that a lot of people can spot it immediately (and it often bugs the hell out of them).

People listen to and for different things in music.

I don't tend to notice corrective autotuning

jesus

I mean, like, I know people come from different musical perspectives but I didn't realize the human listening/sound-parsing experience was anything but universal

to the extent that people wouldn't notice something like autotune, anyway

If I focus and/or it's pointed out to me, then yeah, but otherwise it doesn't really leap out.

Obv I notice post-"Believe" distortive aututone.