Every time I talk with my friend John Kellogg, I learn something new. John’s the vice president of Advanced Strategic Solutions for Xperi, parent company of DTS. That basically means he’s a liaison between Xperi and the music and movie production communities, a position he previously held for Dolby. John spends a lot of time in recording studios, and has a very good one of his own, too, so he’s always up on the latest trends in pro audio. Thus, when he recently told me, “Oh, the loudness war’s over; it’s all LUFS now,” I had yet another of the “Wait . . . what?” moments common to our conversations.

As John explained, the shift toward streaming music platforms and away from CD has increasingly made streaming services the gatekeepers between music producers and music listeners—a role formerly held by CD mastering and production services, and still held, to some degree, by radio stations. Once an artist or record company hands their digital music files to a streaming service, the streaming service then decides how best to present them.

John KellogJohn Kellogg

Except perhaps for niche services catering to audiophiles, music streaming services strive to keep a broad spectrum of customers happy. One way they can do this is by adjusting the volume of individual music files so that playback level remains fairly consistent from tune to tune and album to album, so listeners don’t need to adjust the volume—something that’s especially important for curated playlists featuring multiple artists. Typically, this feature can be defeated—for example, Amazon Music HD and Spotify have an option to turn volume normalization on and off—but the apps usually have volume normalization on as a default.

To implement volume normalization, one measures the average level of a music file, then adjusts the volume for that file so its average volume matches a certain target, such as -15dBFS (i.e., 15 decibels below the maximum possible signal level, or full scale). Volume normalization has typically been done using RMS average level measurements spanning the full audible spectrum. More recently, volume normalization has been done using LUFS measurements, which were standardized in 2011. LUFS stands for “loudness units relative to full scale,” and it’s designed to reflect the fact that humans don’t sense all frequencies of sound equally—a theory embodied in the equal loudness contours, better known as the Fletcher-Munson curves.

LUFS incorporates K-weighting, which combines two filters. One filter compensates for the acoustical effects of the human head; it’s a shelving-type treble-boost filter that starts kicking in at 500Hz and levels off at +4dB above 4kHz. The other is a high-pass (bass roll-off) filter with a -3dB point at 60Hz and a slope of roughly -10dB/octave. By measuring average level with these filters in place, LUFS provides a more accurate assessment of how a human will perceive the average volume than a simple broad-band RMS or peak-level measurement can.

A LUFS value reflects the difference between full-scale (or 0dBFS) and the average weighted loudness. So a LUFS value of -15 is 15dB below full-scale, after the weighting filters are applied.

Back in the days when CD changers, early digital music file players, and radio stations served up our music, record producers and mastering engineers couldn’t be sure how loud the tunes preceding and following theirs on a playlist or radio program would be. But they all wanted their recordings to sound at least as good as everybody else’s—and as anyone who’s conducted controlled listening tests knows, the human ear tends to perceive a louder music source as sounding better. This resulted in the “loudness war” of the last 20 years or so, where mastering engineers (often at the insistence of record companies) would heavily compress a recording’s dynamic range so its average level might be just a few dB below full-scale. Thus, a tune would have an average loudness at least as loud as the previous or next tune—but its dynamic range would be so compressed that the impact of loud transients (encountered in drum hits, bass-guitar slaps, etc.) was lost and the music sounded much less lively and realistic.

LUFS targets for streaming services are typically somewhere in the mid-teens: for example, Amazon is -11, Apple Music is -16, and Spotify is -14. If a file mastered in the peak of the loudness war era for an average level of, say, -5dB RMS, is played through these streaming services with the volume normalization on, the service will bring the volume down to its LUFS standard. The file won’t sound any louder than any other, but it will still suffer the sonic degradation of extreme dynamic-range compression, and its peak loudness will actually be lower than that of a less-compressed file.

Knowing that their files will be played back with the same reasonable average loudness as everyone else’s gives recording and mastering engineers the freedom to incorporate more dynamic range into their productions. They may still want to master with higher LUFS values for CDs, which will necessitate more extreme dynamic-range compression. But considering that the CD is no longer the primary focus for most of the music industry, the record company may not wish to pay the additional mastering costs—and the CD will sound better at the streaming-friendly LUFS value. It’s quite an ironic twist that the popularity of streaming services may end up making CDs sound better, isn’t it?

Reaper Waves

To see the difference among typical LUFS values, I ran a recording through the Waves WLM Plus Loudness Meter plug-in, which is used with Reaper digital audio workstation software, and adjusted the level and dynamic range compression to hit -14 LUFS (the Spotify target) and -6 LUFS (a value that might have been common in the heyday of the loudness war). The recording I chose was “Aprender,” a tune I recorded with saxophonist Ron Cyger for our upcoming album. The reason I chose this one wasn’t purely for shameless self-promotion; it was also because I needed an unprocessed mix that I knew hadn’t gone through dynamic-range compression.

The graphs here show the waveform for the first two minutes of “Aprender” at different LUFS values. The original, uncompressed mix (top graph) had a measured average loudness of -19 LUFS. I then applied a light 4:1 compression ratio so the signal peaks wouldn’t clip (i.e., exceed 0dBFS) and raised the level by 5dB, which gave me a Spotify-optimized average loudness of -14 LUFS (middle graph). Finally, I applied a heavier compression ratio of 9:1 and raised the compressor threshold to achieve average loudness of -6 LUFS (bottom graph), which is more typical of what we’d hear from loudness war-era masters.


You can see how the -14 LUFS version still has a good amount of dynamic range. In fact, the compression I used for that version is comparable to what I use for most of my music productions, and it’s similar to what I used for the soundtrack of the YouTube video in the link above. But the -6 LUFS version has very little dynamic range. I listened to it only briefly, but from a dynamic standpoint, it sounded something like what you hear when you play music from an AM radio station through a high-quality stereo system. Of course, what you see here represents just a quick attempt by someone with limited mastering experience to hit certain LUFS values. A professional mastering engineer would know the right tricks and tools to get better sound at -6 LUFS—but still, it’s probably not going to sound great.

I guess you can sum all this up by saying that the loudness war is over, and audiophiles won!

. . . Brent Butterworth
This email address is being protected from spambots. You need JavaScript enabled to view it.

Say something here...
Log in with ( Sign Up ? )
or post as a guest
People in conversation:
Loading comment... The comment will be refreshed after 00:00.
  • This commment is unpublished.
    Jeremy W · 1 months ago
    @Brent Butterworth I guess I'm looking at it from the perspective of a musician walking into a recording studio and seeing some of the current trends going on. It scares me when I have had to question the choices of a producer, a recording engineer, a mixing engineer and a mastering engineer on why they did things a certain way and the answer I repeatedly get back is that's how they do it. They clearly missed out on the 5 w's and 1 h lesson when it came to applying certain processes and adding tools. Thankfully they all listened to artistic input and we achieved the desired results. So it makes me question more who is responsible for pushing these trends to continue, The artist? The record labels and execs? Why is there such little discussion on preparing an album or a song to conform to loudness standards that are in effect? Again, I am quite thankful for your article as I believe it touches on that point and hope it increases awareness of the subject to everyone. I apologize if I have offended you in any way, my intent was to have a healthy debate on the statement of the Loudness Wars being over. If anything it's just gone stagnant. I appreciate your replies and thank you for your discussion on the matter.
  • This commment is unpublished.
    Brent Butterworth · 1 months ago
    @Jeremy W I agree with your last statement, it's 100% correct. But what makes you believe the recording industry is not educated on the latest technologies? There's considerable discussion of it in pro audio publications, and there has been for a long time -- the article you cite is from 7 years ago. I find it frustrating when an anonymous person with no experience in a field feels the need to tell professionals how to do their job. Show some respect. I know and have worked with a lot of recording pros and they are not the ignoramuses you and so many audiophiles seem to think they are.
  • This commment is unpublished.
    Jeremy W · 1 months ago
    @Brent Butterworth Again, there is nothing wrong with educating ourselves on the latest technologies. If we didn't do this we would still be using a 100% analog chain for all stages of recording. Thank you for providing the link to the document, I overlooked it initially. After reviewing it I still do not see anything new, mainly it shows that the Loudness Wars started in 1989 and continues on to this day. They set 2008 as a year where it finally catches media attention and I would argue that many already knew the Loudness Wars existed well before this timeframe. The increase of social media, web forums & video stream platforms are certainly a good reason why they chose that number. I guess we will have to agree to disagree on the whole debate. I truly hope you are right that mixing and mastering practices change, we're going on 10 years of Streaming services being available and I have seen no change in recording industry practices to adapt and conform to modern stardards. The volume can be lowered on hyper compressed, clipped, limited and distorted music but what's the point when the damage is already done?
  • This commment is unpublished.
    Brent Butterworth · 1 months ago
    @Jeremy W I think when you say an entire industry needs re-education, you do need to present some evidence that you are familiar with standard industry practice. If you read pro audio publications, you'll see that awareness of this issue is very high and that it's a common topic of discussion. Also, I wrote "audio enthusiasts didn't seem to notice that recordings were being heavily compressed until 4 years after the peak of the loudness war"; please read the AES paper I cited. https://www.researchgate.net/publication/269102263_About_Dynamic_Processing_in_Mainstream_Music
  • This commment is unpublished.
    Jeremy W · 1 months ago
    @Brent Butterworth I don't need to provide my credentials in this debate. Technology is ever changing and we seep further and further into a digital realm with new hardware, software, plugins, etc being created every day. Do you not think it is important for everyone to continue educating themselves to adapt to the latest technologies that are out there? If you argue that audio enthusiasts only realized 4 years ago that recordings were heavily compressed then why has the Loudness Wars been a recognized and ongoing issue since 1994 when Oasis released 'Definitely Maybe'? I agree with your statement that many enthusiasts do not have much experience with DRC. But many can hear the effects of it, they just don't know how to describe why it sounds the way it does nor would they understand how that result was achieved. I would argue there are many professionals in the industry that don't fully understand DRC and again, education is the answer to resolve this. I appreciate you posting the original article and replying to me on my comments, again I think it is a great first step for Streaming Platforms to integrate a loudness standard. There's more work to be done on the industry side if we want to truly make the statement that we "Killed the Loudness Wars". Thank you.
  • This commment is unpublished.
    Brent Butterworth · 1 months ago
    @Jeremy W "The recording industry needs to be re-educated." Unless you can share your identity and CV, and establish your authority to make such a statement, I doubt that people who make a living in audio recording will be eager to queue up for your re-education program. As noted in the AES paper cited above, audio enthusiasts didn't seem to notice that recordings were being heavily compressed until 4 years after the peak of the loudness war, when publications started telling them they should be concerned about it. I have to wonder how many of the enthusiasts complaining about DRC have any experience in audio production and even know what compression sounds like?
  • This commment is unpublished.
    Jeremy W · 1 months ago
    @Brent Butterworth I find it highly doubtful that popular music will be remastered again... the recording industry does not want to spend the money to do that. It has been at least 10 years now that we have had streaming platforms in our environment and the recording industry has done little to nothing at all to adapt to loudness standards. The excuse for continuing to follow archaic mixing/mastering processes is always going to boil down to "a purely artistic and musical decision" which is a shame. While streaming platforms have taken a great approach to begin correcting the Loudness Wars, it is only the first step. The recording industry needs to be reeducated and begin to adapt to modern standards that are already a decade old.
  • This commment is unpublished.
    Soundstagenew2021 · 1 months ago
    @Brent Butterworth Yeh so it's obvious the loudness war is not 'over' then. And if you listen to Darko's latest podcast with Mastering Engineer he also acknowledges that it is not over, especially with most popular genre of hip-hop for example. So that's one pro audio guy with a different take to the pro-audio gear in your article. I think it's pre-mature to call the loudness war over. We're headed in the right direction though.
  • This commment is unpublished.
    Brent Butterworth · 1 months ago
    @Soundstagenew2021 That's what I said in the article.
  • This commment is unpublished.
    Soundstagenew2021 · 1 months ago
    @Brent Butterworth "The loudness war had nothing to do with listener behavior (i.e., turning off normalization), it had to do with mastering behavior."

    Normalization is directly influencing mastering behaviour.

    If everything is normalized then the way to 'stand out' on playlists is with dynamic range now. The way to stand out before normalization changed things was louder, to the point of clipping.

    Anyway there's no question everything has been moving in the right direction the last couple years.

    But it's not over yet.

Latest Comments

Your review helped me to make a decision on getting these headphones.

I'm a super picky ...
Brent Butterworth 11 days ago Focal Arche DAC-Headphone Amplifier
@RoyI'm sorry, I am the furthest thing from an iOS expert. I keep an iPad ...
@Brent ButterworthCan you please tell me how to increase the usb sample bit rate when my ...
Brent Butterworth 12 days ago Technics EAH-AZ70W True Wireless Earphones
@Scott HI'd prefer these if you want noise cancelling, and IMO you need noise cancelling only ...
I'm in the market for my first pair of earbuds and I was deciding between ...
Very, very useful and informative article. Thank you so much!
I found SpeakerCompare to be a decent comparative tool.  As a way to increase sales ...
Michel CZARNY 23 days ago Calyx H Headphones
@jerryThe Thinksound On1 was very good indeed and classy too. Only problem was with the ...
Brent Butterworth 24 days ago The New Standard That Killed the Loudness War
@Soundstagenew2021 Saying the loudness war is over doesn't mean no one's going to create highly compressed ...
Soundstagenew2021 25 days ago The New Standard That Killed the Loudness War
@Soundstagenew2021 Correction discussion is for 2017 album 'Colours', so not new. But if you follow that ...