After 35 years of digital audio, I expected higher quality recordings, mixes and masters. One problem is, I am in the wrong music genre. Classical recordings usually get the audiophile treatment, but rock, blues, alt, country, et al., not so much. The reality is that most mainstream recordings, mixes, and masters have had dynamic range compression applied, at multiple stages in the recording, mixing, mastering, re-mastering process, including broadcast. Wimpy loud sound has been going on for over 20 years. In the infamous words of Howard Beal, “I’m mad as hell and I am not going to take this anymore!”
My thesis is that we are so used to overly compressed (read: crushed) sound, people forget what good dynamic sound “sounds like.” So much so, the rebirth of the LP has made a comeback as the vinyl revival. While there is much speculation why vinyl revival, I suggest the underlying reason is folks are tired of overly compressed sound and have fond memories of vinyl not sounding as “bad”, compared to digital audio. But digital audio is not the culprit.
This article is intended to be educational and empowering. Educational in understanding parts of the recording, mixing, and mastering processes that are detrimental to good sound quality and what can be improved. Empowering, if one so chooses, to voice your opinion in this forum or other forums dedicated to overcoming the Loudness War.
State of the Art Analog
I was a pro recording/mixing engineer, and sometimes producer, in the 80’s, having too much fun being a punk, and not appreciating how lucky I had it. Many would say that the late 70’s and early 80’s were the state of the art in analog recording, mixing and mastering. The 80’s was also a time of disruptive technological change in the recording audio industry. It was a time of major technology adoption of digital audio.
From an analog perspective, I was using state of the art (Neve) analog mixing consoles, 2” analog (Studer) 16 and 24 track tape decks and top of the line (Neumann, AKG) microphones, in recording studios with state of the art control room design and monitoring.
While I have recorded minimalistic stereo recordings, I love rock and roll, which usually means multi-track recording. However, I love capturing the live off the floor feel of the band playing together, like they would be playing live. I would get the band to practice up a couple of times, without them knowing I was rolling tape. Those were often the best musical/emotional takes where the band was excited to play in the studio, unencumbered by not knowing they are being recorded. I digress.
To get a good sense of the state of the art analog sound recording and some of the studio gear used behind the scenes, I highly recommend the documentary film Sound City. Check out that drum sound! Also, Muscle Shoals, and The Wrecking Crew, give excellent insight to how these studios created a “sound” that is simply no longer today.
I was an early adopter of digital audio when I purchased a Sony PCM F1 digital audio recorder in 1982:
And by 1986, I was working in all digital studios:
After years of “slammin tape”, what blew me away with digital audio in the studio, even with the Sony PCM F1, was how quiet it was compared to analog tape. No noise at all. Eerie to (not) hear for the first time. People forget, by the time one gets a digital version of an analog recording, it is at least 2nd or 3rd generation tape copy.
Typically, the 2 inch multitrack master is mixed down to stereo tape, and then a safety master or dub is made from the two track master, which is usually the one that is sent in for mastering to LP record or digital audio. The “original” masters are almost always locked up in vaults.
One would be surprised to hear the difference between the original 2” master and the LP that came back from the mastering house as to how different it sounded. Usually much to the disappointment of the band, who heard first hand in the studio how dynamic the master recording sounded.
The other aspect that had me really excited was how much dynamic range could be captured. Maybe it has been too long for folks to remember, but the raison d'etre behind commercializing digital audio is due to the physical limitation of analog tape and tape recorders.
The Digital Audio Technology book pictured at the beginning of the article was first issued in Japanese in 1979 and translated into English in 1982, the year I purchased the book. 35 years later, it is a fascinating read about the history of the commercialization of digital audio. As mentioned, the number one factor is the analog tape and tape machine had been optimized to the point where no further advancements were possible due to the physical limitations of both the tape and the machine. Thus the introduction of Pulse Code Modulation (PCM) and the revolutionizing of the analog tape machine.
There are measurement comparisons between state of the art analog tape and the newly created Sony PCM F1 digital audio recorder from 1982:
Wow, look at that pulse response! So much for accurate sound reproduction from analog tape. Even in 1982, it was clear that the PCM digital audio recorders had significantly better specifications and measurements than state of the art analog tape recorders.
Fast forward 35 years and I would say virtually everyone has one or more state of the art digital audio playback devices, plus a virtually unlimited supply of digital audio recordings, offered in a wide variety of sample rates and file formats.
Further, every Mac since 2002 came with a free digital audio workstation (DAW) called GarageBand, which allowed anyone to digitally record music in their own home or studio. This technology disruption made the “studio” recording experience freely available to the masses, but just like folks found out when the masses tried to be web designers, it still takes training, skill and experience to be good at it.
Which I believe is part of the issue. The training and skills for modern recording, mixing and mastering engineers – i.e. the art in using the science. Where is this being taught? What is being taught? I get the feeling that dynamic range compression, via the many presets in software DAW’s today are simply engaged, and the mixer says, that sounds way louder, cool!
One can hear dynamic range compression first hand in this recent film release of: Oasis: Supersonic. Starting at 44:15, the band is panicking over the sound of their first album – it’s not loud enough. They can’t find their sound, and after a couple of tries, they get Owen Morris to come in and produce. You can hear/see on the video that they were using a box “that made the sound twice as loud, without going into digital distortion.” That folks is a compressor. In fact, one can read the reference to it, “… and heavily compressed the final mix, to an extent he admitted was "more than would normally be considered 'professional'". This was 1994, just reaching the pinnacle of crushed sound in 1995. Quite honestly, not much has improved since, as we will see and hear later in the article. I just wanted to present one example of extreme dynamic range compression of a band that was very popular at the time:
Compared to the dynamic range chart below, from the dynamic range database, this is the epitome of wimpy loud sound, as most of the tunes are DR5 to DR7 = bad.
Digital Audio 23 Years Later Into the Loudness War
Today, as a recording/mixing amateur, in my home studio, I have access to an unlimited number of digital audio recording tracks and effects, as a professional studio in a box goes for about $90. Today’s home speakers and control room monitors can be DSP’d to fit virtually anyone’s frequency response (i.e. tonal) preference., with an unprecedented high degree of accuracy and precision.
I feel folks underestimate and perhaps misunderstand the power of software DSP in 2017.
Remember that Studer 24 track tape machine I was talking about? It is a floor to ceiling, 900 pound physical monster. Here is a picture so one can grasp the magnitude of technological disruption about to unfold.
That’s a Studer A800, released in 1978, which was the technological pinnacle of 24 track analog tape recording.
Now here is the software emulation of the exact sound of this monster for $350.
Here is a demo of the before and after sound with the software modeled A800 in the mix – one can hear the increased loudness with the DSP modeled sound of the analog tape/electronics:
Can you hear the difference in loudness? I sure can, and if I may point out, it does not matter what playback device one is listening to, as one can readily hear the difference whether iPhone and earbuds or ones high resolution home audiophile system. The drums are markedly louder and punchier and the cymbals sound “splashy” but not harsh. If you have not, I urge you to give it a listen. If you have listened to any multi-track rock music over the past +30 years, it is likely to have been recorded on a Studer.
The loudness increase is a compression effect of both the analog tape machine and (mostly) physical tape, usually referred to as tape saturation. One can enhance the effect by hitting the tape a little hotter, (i.e. increase in level), especially on transients and which make the sound “rounder” in a pleasant (albeit distorted) way. If you watch the Oasis Supersonic movie I linked to earlier, there are a few seconds of the film capturing the mix to a two track analog tape recorder and the needles on the VU meters are literally hitting the mechanical stops it is so hot to tape.
That’s slammin’ tape. I am pointing this out as it requires a bit of ear training to hear the different types of compression being applied and what it contributes or takes away from the sound quality. Even a YouTube video can demonstrate audible difference in loudness. However, compression can be artistically used in a way to enhance the emotional effect one feels while listening to music.
Back to my technology disruption point, that 900 pound Studer A800 monster, priced well over 6 figures when it first came out, has been reduced to a $350 software DSP package that plugs into any DAW – no cables required, and achieves the same sonic signature. That is revolutionary technological disruption.
In fact, if you visit http://www.kvraudio.com/allpluginsononepage.php this is literally the marketplace for thousands of software DSP plug-ins. It is amazing what one can find as virtually any analog device and/or effect has been modelled in software DSP.
The simple reality is that we have all heard the sound of this device since 1967 – that’s 50 years ago folks. In fact, if you are listening to mainstream music right now, it is likely that you are also listening to the sound of this device.
All broadcasts typically have this compression/limiting device, or similar, in line with the master outs before transmission. Same with all LP record’s ever cut, special limiters are used in circuit to prevent overloading the cutter head. And heavily used in recording, mixing and mastering scenarios. While the limiting function may not have engaged in all cases, the fact remains this device is in the audio circuit.
I could go on, but the point I am making is that compressors/limiters can unobtrusively do their job as overload protection and/or artistically enhance the music, or can be abused to the point of wimpy loud sound. Having lived both professionally through the technology adoption of digital audio and being on the consumer end, I must say that overly compressed sound is rampant.
And has been going on for more than two decades. I figure if I can help folks understand what overly compressed sound, sounds like and what they are missing, in enough numbers, may help the industry produce more dynamic recordings, mixes and masters with a DR in the “good” zone. I don’t think that is asking too much. We all have access to volume controls and can just turn the music up!
That’s the subject of the next section. We are going to listen to a few examples and I am going to point out what to listen for so you can hear the effect this device has on the sound when abused. Which, unfortunately, has been the norm rather than the exception, for the past +20 years.
I am reminded of the scene in They Live where Rowdy Roddy Piper puts on the special sunglasses for the first time to really see what was going on. Once you see it, you can’t unsee it. Same goes for the next section. Once you hear the difference, with a few examples, you can’t unhear it.
Dynamic Range Compression
Folks, if you only knew… that one can apply dynamic range compression on each of the 24 channels going to the 24 track tape machine while tracking (i.e. recording) and on each individual channel strip during mix down, on subgroups, and on the overall stereo mix. Or even feeding the two channel mix, through another couple of strips to pre-master the mix. Compression/limiting can be applied during mastering. Depending on how the music is delivered, like LP’s or broadcast, may yet have another set of compressors/limiters on the audio signal. The link above to dynamic range compression goes to Mastering Engineer, Bob Katz’s web site. I consider Bob’s discourse on the subject of compression to be the best explanation.
If you were in the studio/control room with me, listening to a live band and what they sounded like mic’d live over large format control room monitors, you would be surprised at how dynamic it sounds. Sounds fantastic, you can feel the band's energy coming across.
As consumers, we are not getting anything remotely close to that when it finally makes it to our stereo systems. More often than not, the dynamics, especially the drums or the band itself controlling its own dynamics, is crushed to death with dynamic range compression.
I feel this two minute video gives an excellent audio/visual explanation of what dynamic range compression sounds and looks like. Even though the sound file is compressed when uploaded to YouTube (ironic), one can easily hear the audible difference, and correlate with the visual difference, between the compressed and uncompressed drums, independent of playback system:
Hear the difference? One can also visually correlate the difference as shown in the video, with the audible difference. “The loudness treatment permanently changes the sound. Wimpy loud sound! When there is no quiet, there can be no loud.”
Think about this, one can hear the difference, no matter what the playback system is. Whether it is on my iPhone using my Bose noise reduction earbuds on a Vancouver SkyTrain or listening to my accurate sound reproduction system using DSP at home, it’s still wimpy loud sound. Folks, wimpy loud sound is independent of the file container it is delivered in. Meaning, just because it says high resolution does not mean it hasn’t been crushed to death with dynamic range compression. More often than not, the worst offenders are the “remastered for audiophile’s”, in high resolution, at premium prices, but have massive dynamic range compression applied and sound worse than the original.
The loudness treatment permanently changes the sound is worth repeating. Regardless of what one does on the playback side, this is simply undoable, even with so called declipper’s.
This right here folks is the fundamental issue. We have had over two decades of wimpy loud sound. As described at the beginning of this article, I feel most folks have either forgotten what good dynamic sound “sounds like” or maybe have not have heard good examples. So let’s look at what I would call a few good examples of dynamic sound.
The Good - Dynamic Sound
Here is a classic, Stevie Ray Vaughn and Double Trouble, Tin Pan Alley from 1984. DR 18:
While I can’t upload the music, perhaps one can locate the CD and give it a listen. Needless to say, I feel it sounds wonderful, even from 1984. The music breathes and pulses up and down dynamically, as Stevie and band take you for an emotional roller coaster ride. Turning the volume way up gives one the sense of being in an intimate club setting. With the right playback system, one can literally feel the bass guitar notes and drum impacts. It is a live off the floor recording, at least the “bed tracks” are, as one can hear the bass cabinet resonating the drum snares, just like it would be live. Fantastic, everything I want in a quality recording, mix and master.
Sure there is some compression being applied, but it is not soul crushing the life out of the music. That’s the point I am making. One can still use compression, judiciously and artistically, to deliver the desired effect, but without killing the musical enjoyment of dynamic sound.
How about Peter Gabriel’s Security:
Or even hard rockers AC/DC, Back In Black:
I am hoping with some of these examples, people have the CD’s and can listen for themselves. Turn it up and enjoy the dynamics.
Or hunt through your music collection, and if using JRiver, add the DR column to the music list and sort on highest DR first, pick a rock song, give it a listen, then sort on the lowest DR and without turning the volume down, try and give it a listen. See how long you can listen without turning the volume down. Rinse and repeat as many times as necessary to hear the dramatic loudness differences between music that sounds more real and dynamic versus the squashed, screechy, made for radio sound. The more accurate the sound reproduction system the greater the differences.
Sometimes it is hard to tell exactly what form of compression is being applied as it is so over the top. This can result in “pumping or breathing”. Another sometimes overused approach that contributes to pumping and breathing is side-chain compression.
What this means is that the compressor is being triggered by an external source, which in turn compresses (i.e. lowers) the music passing through the compressor. In the specific case below, the drums are the trigger to the compressor, even though the drums are also being compressed. This is also referred to as “ducking”, meaning the sound gets lowered (i.e. compressed or squeezed) every time the kick drum hits, the bass guitar “ducks” below the drum sound, so just for that moment in time, the drum stands out and not partially masked by the bass guitar that is occupying the same frequency range.
Here is a YouTube video of two examples of side chain compression. One example is used as an effect and the other example is to clean up the mix by ducking the bass guitar, every time the kick hits. I know it is a 10 minute video, but consider it a crash course on compression as the ratio, attack, and release controls are also being played with and one can easily hear the range of effect when these controls are varied. The range is so great as to make the sound unrecognizable to its original form. That’s because its timbre is being altered.
The Acoustical Society of America’s definition of timbre, “Timbre depends primarily upon the frequency spectrum, although it also depends upon the sound pressure and the temporal characteristics of the sound”.
Looking at the attributes of timbre, one can see the definitions for both spectral and time envelope attributes. Looking at the time envelope, one can see ADSR – attack, decay, sustain and release. Add threshold and compression ratio and voila, these would be the controls on your garden variety dynamic range compressor in any studio. It does exactly that, it alters the time envelope, in terms of rise time, duration, and decay of the sound.
There is a definite art to the science of applying dynamic range compression and what sounds good versus what sounds not so good. During my studio training, one adjusts the range of each of these attributes/controls individually on dynamic range compressors, while listening to the sound, and training one's ears so when I twist this knob, it sounds like this. Overtime, one gets pretty adept at “getting one's sound”. However, given the ubiquity of a studio in the box on everyone’s computer, not everyone has had training. Additionally, the software makes it so easy, literally selecting dozens of factory presets. Often those presets are designed for wimpy loud sound and the cycle continues.
The art of the science is how much compression to apply to still retain the dynamic nature, but also mix to a certain level of loudness to retain an overall good dynamic range, i.e. DR +12. Not an easy task as one could have over compressed a track while tracking and there is not much one can do to restore it. Or the final 2 track mix before mastering is already over compressed.
The Bad and the Ugly
While most people won’t have this tune, I just wanted to point out that it’s bad because it has a DR of 7 and the ugly is that it has pumping compression sound. This unfortunately, is quite common. What possibly could be the reason for crushing the dynamic life out of this tune, and pumping to boot? I really like the band and the tune, but the over the top compression kills it for me. This is the type of tune that would benefit from a DR 15 or greater and would be exciting and emotional to hear it cranked up, like being at the club. Makes one want to get up and dance.
Instead, it is so overly compressed, it does not matter if it is coming across a radio or one’s full range audiophile system, as still sounds like it is coming across on a radio. Really disappointing. It should have the same DR like the Stevie Ray Vaughn tune mentioned earlier. Give it some life, let the music breath, have some natural up and downs in dynamics – the very thing that gives music emotion and movement.
Anyone who is a musician, amateur or pro, and plays an instrument or sings, understands dynamics. I have an acoustic drum kit, acoustic and electric and bass guitars and amps in the same room as my stereo. One can tap lightly on the drum kit or bring it to a crescendo, demonstrating huge dynamic range. As mentioned before, aside from older recordings, pre-1990’s, there is very little of that left in today’s recordings, mixes and masters, especially in the non-classical musical genres. Dynamic range compression sucks the life out of music.
I am not saying that dynamic range is the end all beat all attribute that determines the overall enjoyment of a musical piece, but there is a minimum bar that should be met on any musical piece. Sure, some music genres benefit more than others from dynamic range compression, like pure metal and thrash, but even still that music benefits from some dynamic range movement.
Look at AC/DC’s Back in Black, it has sold an estimated 50 million copies worldwide, making it the second-highest-selling album in history, without the use of overly compressed sound, except for the later remasters! Remember earlier in the article, the average DR of the US release is DR12. Sounds great!
Music lovers and consumers have a voice and can exercise one's right not to purchase wimpy loud sound. Or at the very least, voice your concerns, at one of the many links at the end of this article.
I find it ironic that after 35 years of digital audio, which is at an all-time high technology adoption, where virtually everyone has access to one or more, super high resolution, low noise, accurate digital audio player, the music content is at an all-time low for dynamic sound quality. How the heck did this happen? Even more so if it is the root cause of the vinyl revival.
The way music is overly compressed today, and for the past 23 years, basically wipes out the reason for having the requirement for a wide, dynamic range, low noise, digital audio playback device, software, or system. Crazy.
Also ironic that there is already excellent information for recording, mixing and mastering engineers on how to make better recordings in the 21st century. I hope one day this or similar would be a standard and the recording needs to be DR 12 or greater before it is released for public distribution. There is no technical or artistic reason, why this can’t be. For sure, there are exceptions like EDM, where pumping, breathing are part of the sound design. But what’s good for AC/DC should be good for any rock band.
Say yes to DR:
While I was writing this article, AC/DC (DR 12) was playing and my daughter was goofing around dancing and then I put on Texas T (DR 7) above and she stopped dancing and put her hands over her ears. Enough said.
I hope folks try sorting music by DR number and start by listening to a high DR like 15 to 18 rock tune, if you can find one. Turn up the volume and give it a listen. Then select a DR 7 tune, should be easy as there are lots of those, and without changing the volume, give that a play. Let’s see how long you last before turning down the volume. I like this comparison approach as it points out how much more enjoyment I get out of the tune where there is dynamic movement instead of wimpy loud sound. It also demonstrates the radical difference in loudness between two tunes.
While I try not to sort my music by DR, sometimes it just plains sucks when you really dig a tune, but it has been so overly compressed, you go to the Dynamic Range Database to check to see if there is an earlier pressing before 1995. But hopefully folks see the severity of the problem. That’s over 22 years ago. Does no-one in the audio industry remember what good dynamic range sounds like?
What can be done?
Vote with your wallet. Don’t buy anything less than DR12. Check the Dynamic Range Database. I know, not too realistic as I look at my collection or rock music with at least 80% of it below DR 12.
Share info about CD’s that have the most dynamic range.
With that I hope everyone can still enjoy the music!
Mitch “Mitchco” Barnett.
I love music and audio. I grew up with music around me, as my mom was a piano player (swing) and my dad was an audiophile (jazz). My hobby is building speakers, amps, preamps, etc., and I still DIY today.
I mixed live sound for a variety of bands, which led to an opportunity to work full-time in a 24-track recording studio. Over 10 years, I recorded, mixed, and sometimes produced over 30 albums, +100 jingles, and several audio for video post productions in several recording studios in Western Canada.
I love accurate sound reproduction. For a more detailed perspective, I wrote this 327 page eBook that provides a step by guide to Accurate Sound Reproduction using DSP. Click on Look Inside to review the table of contents and read the first few chapters for free.