Talk:Dynamic range compression/Archive 1

merge with Limiting?
waddaya think?? It is discussed in this article quite thoroughly...either that or provide a link to the main article here...Ab2kgj 03:06, 4 May 2007 (UTC)

a link would be a good idea, as long as it is made clear that Limiting is a form of compression adapted for a specific use and not a Dynamics tool in its own right.

robert parkinson 24/05/2007

DO NOT MERGE WITH SIGNAL COMPRESSION!!!
These are two very different applications of the same concept. They are relevant to two very different groups of people, with relatively infrequent crossover relevance.

Note: Some material in this article was originally from Federal Standard 1037C, which was written in support of MIL-STD-188.


 * Currently, one says "signal compression has the following meanings: In analog (usually audio) systems, reduction of the dynamic range of a signal ..." while the other says "Dynamic range compression ... is a process that manipulates the dynamic range of an audio signal."
 * That certainly sounds as if they are the same thing, and hence should be merged.
 * Why not merge?
 * If signal compression is somehow different from dynamic range compression, shouldn't at least one of the articles mention what, if anything, is the difference?
 * Or at least describe the differences between these 2 different groups of people?
 * --68.0.120.35 23:51, 17 April 2007 (UTC)

Article on signal compression describes really "data compression" of streams. In an audio sense, this is audio compression (data). Dynamic range compression can be achieved with no reduction in data volume. Blouis79 23:31, 30 May 2007 (UTC)


 * No it doesn't. Signal compression is about the exact same thing as Dynamic range compression. — Omegatron 01:17, 31 May 2007 (UTC)

application oriented vs. electronic design oriented
--- hey Omegatron, I've done a bit of work here, by no means complete. It seems to me that you were approaching this article from an electronic engineers perspective. I've made my edits more application oriented, because I believe that more readers will be on this page because they want to use compression, rather than build compressors. I condensed the electronics stuff to it's own section, which could use some expansion (about digital compressors, and probably correcting me a little on optical compressors). I'm still working on how to explain side chaining in full (de-essing, ducking, and more!) as well as expanders/exciters. I'm basing a lot of my information on Bob Katz's book "Mastering Audio", and on experience. This article could get very long, if I'm not carefull. BTW, I'm very confused by your waveform diagrams for compansion. I'm not sure they fit, but it's also not my greatest area of expertise. I also prefer the longer time-frame waveform diagrams for illustrations. Individual wave cycles are often orders of magnitude faster than the amount of time between attack and release. ---RevMax


 * We haven't really gone into the electronics at all yet. :-)  Just trying to explain what the damn thing actually does, which few people seem to know.  They just know how to use them.  (And I know the opposite.)  Anyway, I've learned more and now know the difference between clipping and limiting.  I'm not sure I understand compression exactly.  It's like hard limiting but "softer".  But I can figure it out.
 * It seems there are two types of companding. One is the type I was drawing earlier, which non-linearly distorts the wave and then undistorts it after transmission.  This is effectively what is used in mu-law and A-law telecommunications and exponential quantization.  Then there's the kind like Dolby uses that is actually a standard audio compressor and then a standard audio expander?  I'm not really sure how you can "undo" a compressor, yet.  Still learning. - Omegatron 14:31, May 24, 2005 (UTC)


 * Since you brought up quantization: is the audio level compression on a Nextel iDEN radio using the Direct Connect feature still compression? It occurs as part of the quantization process. They're trying to shoehorn more bits through the system by reducing the possible steps from loud to soft during quantization. Is that still compression? User:David Jordan 3/18/2007.

Additions and details
If the following information is correct, please add it to the article:

Multiband compression
Multiband compressors can selectively compress specified frequency bands. They are primarily a mastering tool, but their inclusion in DAW plug-in sets is increasing their use among mix engineers.

compression vs limiting
needs to show the difference between compression and soft clipping, which always confused me. i'm pretty sure i understand it now, but i bet others are confused too. compression will apply a scaling or attenuation to an entire wavelength when the peaks of that wave get above a certain level. soft limiting will apply an attenuation to only the parts of the wave that actually go above that level, reshaping the waveform. - Omegatron 17:36, Sep 1, 2004 (UTC)

.... the difference between compression and limiting is the ratio. Two commonly accepted definitions are that more than 4:1 or more than 8:1 ratios are limiting, anything less is compressing. They are no different in process. Soft and hard limiting are differences of degree too. "Brick Wall Limitting", the "hardest", means that NOTHING passes the threshold. In practice, this is a ratio of 50:1 or greater. In practice, unless applied to only the shortest, transients, this sounds like shit. It is most appropriate as a safety device in live and broadcast applications.

Parallel Compression
We need something about parallel compression too. This is when you send to a compressor from a pre-fader Aux Send, instead of an insert. You then blend the uncompressed and compressed signals, effectively allowing you to turn up the lows, instead of turn down the transients. The effect is subtler, and crisper than standard compression, because the results of the signal processing have no time based action on transients. Instead, the unaltered signal has it's quieter components phase-amplified when a signal with squashed transients is added to it. Because of this, phase corelation between the two signals is crucial.

companding
Actually, Compansion is pretty simple, and it's mainly used for noise reduction. All that happens is that the signal is compressed before transmission, and expanded before playback. I've never heard of compansion as anything other than this. It can be done in multiple bands, and those bands can be variable. That's how dolby works.

I believe companding is more of a non-linear effect like limiting. for instance, a signal can be exponentially companded which scales the entire waveform according to an exponential curve, so that the quiet parts are slightly less loud than the unchanged loud parts, it is passed over the noisy channel, and then the exponential non-linear function is done backwards, to make the quiet parts quiet again. i don't think it uses thresholds or changes gains anywhere. this, compression, limiting, and other forms of amplitude modification should be clarified. - Omegatron 17:45, Sep 1, 2004 (UTC)

http://www.arts.arizona.edu/studio/Compressors,LimitersExp.html

that site has different definitions for both, it seems. there is a lot of confusion about the specifics. i will try to learn the actual properties and draw some graphs. - Omegatron 17:52, Sep 1, 2004 (UTC)

http://www-ccrma.stanford.edu/~jos/filters/Dynamic_Range_Compression.html

yep. according to http://diana.ee.pucrs.br/~decastro/pdf/spra163a.pdf :


 * The U.S. and Japan use m-law companding. Limiting sample values to 13 magnitude bits, the m-law compression portion of this standard is defined mathematically by the continuous equation:
 * F(x) = sgn(x) ln(1 + &mu; |x|) / ln (1 + &mu;) Equation (1)
 * -1 &le; x &le; 1
 * where &mu; is the compression parameter (&mu;=255 for the U.S. and Japan), and x is the normalized integer to be compressed.

That's not the same as regular audio compression, which "reacts" to a loud signal by turning down the gain. Instead it's an equation that reshapes the signal to a logarithmic/exponential curve. These all need to be clarified. - Omegatron 16:18, Sep 2, 2004 (UTC)

Graphs
My impression is as follows:

Regular waveform:



Compressed waveform:



Hard limited or clipped waveform:



Soft limited waveform: (THIS IS WRONG) Soft-CLIPPED waveform:



First stage of non-linear companding:



it sort of maps all the amplitude values to an exponential curve (this example uses the original function^(1/5)), so the quieter parts are louder, but it is nonlinear distortion at the same time.

although this particular signal gets louder, so i guess it is expansion? kind of confusing.


 * no, it is compression, since it is compressing the dynamic range of the signal. however, it has a higher gain than the first example.  the rule is that compression always decreases dynamic range and expansion increases it. - Omegatron 16:03, Sep 2, 2004 (UTC)

My field is communications rather than music so terminology could differ. I think that graphs showing waveforms are not good for representing compressor action. Compressors are linear; sine waves in will be sine waves out after level adjustment.


 * Linear on short time scales, you mean. The gain is just changed dynamically, right?  The compressed waveform image is supposed to show that, with the compression acting halfway through... - Omegatron

The gain changes are much slower than the waveform.


 * Oh yeah. Good point. - Omegatron

Limiting is an ambiguous term, it could mean a sort of modified clipping as you are representing or a fast-acting high-level linear compressor. Meggar 22:18, 2004 Dec 8 (UTC)


 * Ok. I would still like some images in the article, but I don't know how to show them accurately. - Omegatron 22:29, Dec 8, 2004 (UTC)

Yeh these graphs are definatly not correct. The reaction speed of a compressor is at least one period of the slowest signal present (ie: the bass end) and usualy slower. A better graph would be showing loudness over time. For example the opening movement of Beethoven's 9th starts with a very loud burst of strings and tympany, followed by a much quieter bit. After compression the loud burst would be much quieter, and since the gain is usualy turned up to compensate for the compressor's action, the quieter bit would be louder. I've added an example of TV advertizing, where compression is used liberaly on the sound track to make it sound louder. --Swamp Ig 01:36, 21 Jan 2005 (UTC)




 * Thanks. I still want graphs, though.  So I would use something like this kind of waveform:, and then have sections of the "before compressor" and "after compressor" blown up to show that the wave has not been distorted nonlinearly in the process?  The limiting and soft limiting are correct, right?  I will put them in the appropriate articles.  Not sure about the compander waveform. - Omegatron 14:58, Jan 21, 2005 (UTC)


 * Here's a crude idea for an image. The inset is supposed to show the same waveform simply at a lower gain level.  I'll fool around with audio programs later and maybe make it with those.  Is this a good way to show the relationship?  - Omegatron 22:24, Jan 21, 2005 (UTC)

Illustrating this article
Here are some images of limiting and clipping:

A better idea for illustrating the different concepts could be showing a simple waveform in both a "wide view" to show the change in gain over time, and a "zoomed in view" to show how the waveform is either distorted or not distorted. Here's the example wave I am thinking of, but maybe someone has a better idea:




 * It's a mixture of 100 and 110 Hz sine waves, to get the beating effect (so that it has some structure visible from this "macroscopic" view but still looks like a sine wave up close), then I increased it linearly from left to right, so you can see as the clipping/compression kicks in. I'm sure there's a better way to show this, though, so I'm holding off on making all of the images/sound files for a bit until I get some other opinions. — Omegatron 02:34, 2 February 2006 (UTC)

this article needs sound samples!!
i mean, duh. :-) - Omegatron 21:49, May 25, 2005 (UTC)


 * The trouble is that compression is hard to hear unless it's laid on rather thick, especially for people who aren't audio engineers or musicians. - furrykef (Talk at me) 04:39, 15 Jun 2005 (UTC)


 * That's why it needs sound examples. Oh.  I meant more than one example; before and after compression.  - Omegatron 13:29, Jun 15, 2005 (UTC)


 * I know, but it's still difficult to tell the difference, unless the threshold is pretty low and ratio is pretty high. I suppose we could do that to exaggerate the effect but it wouldn't really make the usual purpose of compression more clear. - furrykef (Talk at me) 02:29, 16 Jun 2005 (UTC)


 * The point of before and after sound samples is to make the usual purpose of compression more clear. I still don't really understand what it does to the sound.  — Omegatron 18:10, 30 September 2005 (UTC)

Compression as effect
Does "compression" mean the same when applied as an effect (e.g. in http://line6.com/tonecore/constrictor.html - with sound samples)? This seems to contradict the previous comment by furrykef - "effect" being the opposite of "hard to hear". What exactly does compression ratio mean in this case? Common Man 17:40, 30 September 2005 (UTC)

Video level compression
Does video signal level compression exist and if so, what is it used for?
 * On a monochrome signal it would be like reducing the contrast, I guess. But I'm not expert. --84.149.197.230 06:43, 26 June 2006 (UTC)

infinity limiting
shouldn't there be something about an infinity:1 ratio for limiting? --24.117.45.211 19:11, 15 January 2006 (UTC)

Compression ratio
My edit was reverted! So sad.

I shall now weakly attempt to defend my wording change. I edited the description of ratio control to say something like "an increase of 4 dB will be needed to increase the output signal level by 1 db over the threshold." It seems that since we're talking about a compression ratio, and since ratios generally come in the form "ratio of foo to bar, that wording is easier to understand. Though perhaps I should have added a statement like "the compression ratio is the ratio of input signal increase to output signal increase", then included my example.

The current wording seems to imply that the compressor will allow the signal to get to 3db above the threshold, then only knock it back once it gets to 4db, which is incorrect. (Perhaps I'm just seeing it that way.)

References: http://mixonline.com/mag/audio_understanding_compressors_compression/index.html

Eh?

A layperson's use: compression for noisy environments
I came to this page hoping to find some pointers on how to do my own dynamic range compression. I enjoy classical music, but am often in a noisy environment (an apartment in New York or driving on the road). Classical music has a notorious dynamic range (especially on CDs), and in these environments, I can't hear the quiet parts, then suddenly the loud parts startle the neighbors and deafen me.

Limiting (such as Sony's AVLS) sounds terrible, and a single-band compressor isn't great either. For what it's worth, I eventually found this Winamp plugin (Sound Solution), which did the trick quite nicely with this preset. For the car, I can output via winamp's DiskWriter then burn or encode the resulting wavs.

So, if anyone is in the same situation as me, hopefully this will work as a solution for you as well.--4.237.241.23 05:20, 28 July 2006 (UTC)


 * Cars are a perfect example of a place where compression is necessary. My stereo has compression built in, but I haven't figured out how to set it correctly yet.  :-) — Omegatron 18:19, 29 July 2006 (UTC)

The article's got a full section on this subject now. I originally came to it via the ReplayGain page and was frustrated to find zero information on how to actually implement it. Incidentally, having done more research, Stereo Tool is now my compressor of choice.--4.231.243.121 22:36, 30 July 2006 (UTC)


 * Interesting. Seems like that's something you would just build into the player, though; analyzing the audio files and playing them all at a psychoacoustic constant volume. — Omegatron 01:29, 1 August 2006 (UTC)

Yup, compression is commonly used for noisy environments. Most every restaurant, shop, or public facility with piped-in music provided by a professional service uses compression to stabilize volume levels and keep quiet sounds audible (by making them not so quiet). I've added a paragraph to the 'common uses' section to this effect.

One thing I didn't add was another very common use: speech and other vocals will always be compressed when amplified in a live PA setting, and typically in any broadcast or recording situation, as well. I'm not sure how best to phrase it for the article, but there are several reasons for this: One is that microphones tend to be disproportionately sensitive; get too close to the mic and you're blowing speakers, get too far away and no one can hear you. Good microphone technique means you'll know to pull back when you raise your voice, and get close in when you're quiet. But even when people do it right, that only goes so far, so engineers still have to use compression. Another reason is that amplification exaggerates (at least, in our brains) the dynamics of speech, which is pretty dynamic as it is; unless you're hearing it right at speaking-voice level, it's less jarring and easier to understand a "flattened" voice coming from a speaker than it is to hear something wildly fluctuating between loud and quiet articulations. —mjb 20:38, 3 October 2006 (UTC)

I'd also like to see at least a section on lay-person's usage, especially for Classical Music, e.g. converting MP3s for replay in a noisy environment. The word "dynamic range compression" was what I typed in, there is no way I would have found this article if it were called "limiting". I hope that helps. - cce 29 May 2007


 * See the "Compressors for software audio players" section. Stereo Tool is perfect; use its presets and output the audio to a file.  If possible, run it on a newly-ripped file (wav) so you don't get generation loss from recompression (although the mp3-to-mp3 quality loss won't be audible in a noisy environment).--Father Goose 03:58, 30 May 2007 (UTC)

Presence of limiting circuitry in consumer hardware, overcompression of modern albums
My personal experience isn't particularly wide, but I do have some... having seen mention that "modern Sony products include AVLS", I'd be interested to hear from other users - DO they? Whilst it's certainly true that quite a few of them did in the 90s - I did, at school, own a very nice, simple Walkman with genuinely such a feature, priced well within reach of a not particularly affluent teenager - more "recently" they seem to be blazing a more cheapskate trail and redefining the "limiting" function more as a switchable resistor or electronic feature that simply lowers the maximum volume setting, either dropping the volume available at a particular setting across the whole range (for analogue controls), or preventing the user from increasing volume beyond a certain point (for digital). Rather sad, as it misses the entire point of the feature - to allow you to pump the volume higher for quiet sections when in a noisy environment, without blowing an eardrum when a forgotten crescendo kicks in (or the volume wheel rubs against something in your pocket), then being able to switch it out for improved dynamics (particularly, removing the increased tape noise!) when listening in a quieter spot. The real limiter on my old (sadly deceased) Walkman was a brilliant and very useful feature, and the sudden discernable loss of impact from loud noises - not quite clipping, but i'd say it's 20:1 if not a whole lot harder - was a good reminder to lower the volume setting; the current variant seems to be little more than Sony doing some legal ass-covering and assuming it's users are numpties who will turn the volume up to ear-damaging levels unless kept in check, but unwilling to include anymore the slightly more expensive circuitry that would allow louder overall volume without the dangerous peaks. Just another example of their corporate insanity I guess (it seems like the entire board of directors are on shrooms some days)

Or is this wrong? Have I just been buying (or in-store playtesting) the wrong products? And do any other manufacturers offer such a feature on their kit? I'd be most interested... :) (if it's available as a software mod or even a built-in feature on iPods, it would be the one killer app that would finally convince me to jump ship to Apple from my currently beloved no-name alternative)

Also, what would the feasibility of feeding ripped CDs through an expander filter to recover some of the dynamics? The somewhat 8-bit-sound-card/mu-law nature of various recent releases starts to grate after a while... I know they're probably intended to let poor-ass teenagers rock out to the latest hits whilst only owning a £24.95 5-watt all-in-one stereo or a pair of powered iPod speakers, but it sounds like crap through a hi-fi with even a modestly decent amplifier and set of cones (1992 Amstrad Micro 1000 driving transplanted Grundig twin-sets, anyone? Truly a ghetto arrangement but sounds miles better than your average 'blaster)... it's not limited to the latest and greatest either, after replacing a physically ruined original CD copy of an early 80s album with a "digitally remastered!" re-release, I noticed it was now sonically ruined - comparison with mp3s taken from the first one (unfortunately too low-quality to burn a replacement from) showed it now to be much more heavily compressed, and certain nuances or depth of mix where many instruments were playing together were lost. Quite what they needed those 20 bits for in their "super-fi" remaster is a mystery.

Having just come up with this idea, I suppose I'd better have a crack and see if it's feasible... once I clear enough hard disc space!

-tahrey 21/4/07 —The preceding unsigned comment was added by 82.46.180.56 (talk) 23:25, 21 April 2007 (UTC).

Need for clearer language
For someone without a lot of technical expertise who's trying to understand these concepts, the language in the article could be improved. Most importantly, this sentence, in the second paragraph of the "Explanation" section: "For example, with a ratio of 4:1, an increase of 4 dB will be needed to increase the output signal level by 1 dB over the threshold." I simply can't figure out the syntax here. The phrase "will be needed," in the passive voice, makes it unclear what precisely is being referred to. —The preceding unsigned comment was added by 65.19.211.29 (talk) 22:30, 1 May 2007 (UTC).

History?
seems like there should be at least a small section about the history of compression. when it was first used, by whom, notable improvements and uses, etc. Powrtoch 15:37, 29 August 2007 (UTC)

Notable compressors
Here's a start of a list of compressors that I think a worthy of a mention (please add more): Iain 06:45, 31 August 2007 (UTC)
 * UREI 1176
 * Teletronix LA 2A
 * Fairchild 670
 * SSL console compressor (I forget which one)

Adverstising?
Is this advertising? (external links): Geir 07:30, 4 September 2007 (UTC)
 * Description of expansion, compression, and limiting in an audio processor

It's about a specific compressor so I'd say it's advertising. Iain 04:20, 8 September 2007 (UTC)