>[on YT] Everything plays at a similar loudness, regardless of how it was mastered. And no-one has noticed.
None of us noticed; it seems like the labels and sound engineers may have; take this image comparing the dynamic range of the CD and youtube releases of Pharrell's "Happy"[1]- I thought at the time it might just have been a mishap or random occurance, but now it looks like it was a very deliberate move.
So, rather than ending the loudness wars on CDs and radio, we might end up long-term with the ludicrous situation where CD's sound worse than technically inferior Youtube videos. Youtube is the new vinyl.
-
E: I added links to the YT release[2] and (presumaby) a CD-sourced video[3] for anyone who wants to listen to the difference. Pay attention to the claps and drums (but I think it's obvious even on my laptop speakers).
I noticed! I am a bass player who now happens to write software in the audio industry (I work for a company that makes hardware). But I noticed before coming here.
Some of the worst CDs in my record collection are:
1. Paul McCartney - Memory Almost Full (this is the worst album I own - considering Sir Paul left The Beatles over the mix of The Long and Winding Road, he should leave his own band over his master of this album.... I wonder if Phil Spector bribed Bob Ludwig to do that in time-delayed revenge???)
2. Rush - Vapor Trails (not helped by distortion during recording; it's one BIG square wave)
3. The Fratellis - Costello Music (distortion all the way through the album)
4. Red Hot Chili Peppers - Stadium Arcadium (there's no possible way to make the first track sound quiet, try it!); apparently the vinyl has a different master and sounds infinitely better.
5. Foo Fighters - One By One (boxy drum sound and non-compliant CD format aside [chock full of copy protection so you won't see "Compact Disc" on the CD at all....], it's loud loud loud).
I could go on.
Considering the effort musicians go to get the best sound engineers, recording equipment, A to D converters, studio time and instruments, to throw it all away on a LOUD CD seems stupid to me. It means the album won't stand the test of time, and will sound as abysmal in the future as the day it was released; it is very short-sighted to release a LOUD CD for commercial gain (radio??? does anyone master just for radio still?) in the short term.
It was the fact that Youtube was normalising sound volumes that the author said nobody noticed, not that the loudness wars exists.. for which there are so many examples it might be quicker to list the counter-examples.
It's noticeable on almost every track basically as soon as the drums kick in. (About ten seconds into "A Proper Story", for example). It's depressing that this has occurred even on a piece of indie content for which so much care has otherwise been taken— Darren Korb was the sound designer for the game and gave a terrific presentation at GDC 2012.
Intentional or otherwise, the waveform certainly clips— you can see it in any audio editor, eg: http://imgur.com/a/WacOy
Since the drums sounds all started life as hi-fi samples, the detail is certainly there in the original Logic sessions. It would be an interesting experiment to do a listening test comparing the CD versions of the tracks against versions mastered with more headroom, either less loudly or to a 64-bit format.
It may well be that a certain degree of the grunge is artistic versus technical, but I suspect that in a blind test I would still prefer a version with slightly less muddy drums.
Have you listened to the remixed Vapor Trails? It's phenomenal IMHO. There are tons of little quirks in the songs that you just couldn't hear before. Plus they might have added a few tracks to each song.
I'd love to hear what somebody who actually knows what they're talking about compare the original with the remix. I'm a huge Rush fan and could discuss them for hours.
No, I will consider buying it again at some point perhaps, but it is sad they didn't do it right the first time, particularly as they're a 3-piece and there is so much space for them each :-(
It's odd how much difference a mix makes: listen to Deep Purple's Roger Glover remixes of earlier DP albums (subtle), or Dream Theater's "official bootlegs" of Scenes From A Memory (massively different, like a different album).
Unbelievably, some "mastering engineers" have been known to deliberately clip the track in pursuit of loudness, often at an ADC stage, and then turn the level down a fraction so it doesnt clip on playback. unfortunately, lots of DACs will produce inter-sample peaks under these conditions, which leads to that awful sound.
> Pretty sure that first MGMT album clips all over the place.
I haven't heard their first album but I'm surprised at this. In my experience, most indie acts have greater creative control over mastering and just about every indie act I listen to (with the exception of some of the heavier stuff) is very light on the "loudness". By contrast, some of the more "pop" stuff I like, even songs that should be clean and distortion free, are unnecessarily jacked up way past the redline. A great example is John Mayer's "Heavier Things" album, with clipping and distortion on the lightest, cleanest tracks, and it gets worse with each of his albums after. It's frustrating because I know he's a brilliant musician who knows better, but I imagine it's the label who has the final creative control over what sound is heard on the CD.
No I didn't. I thought that Rush had taken my money in the first place for the album, and should have done a good job on that so didn't deserve me buying it again!
It would be interesting to hear it though. The bass drum is completely lost in the first one (no punch at all).
Their most recent album was also full of reverberation and distortion, which made listening to it in the car very very difficult. It fared better in a quieter environment I thought.
How long have you been playing bass? In a band or anything? All good fun!
I've been a huge Rush fan since I was a little kid (thanks to my parents for that). I loved Vapor Trails when it came out even with the poor sound mixing. The remixed version is even better in my opinion. To me, it's just a fuller sound that's much more enjoyable.
Really interesting post thanks. Even as an audio novice I've noticed the terrible loudness trend in music, particularly in Foo Fighters releases. It literally makes my ears tired.
The Foo Fighters are mastered loud a lot of the time. One trend they have is to have a massively gained snare drum shoved through a compressor so it chops the peaks of the snare off and has no crack like a snare should, so it sounds like this: PFFFFFT
Older recordings from the 70s (perhaps early Kansas) have a different sound (not such harsh compression), so the snare drum has a sharp crack on them and they sound like this: CRACK
To get that "sound" Foo Fighters layer distorted guitar on top of more distorted guitar (they use Vox AC30s apparently!). Doing this does make it sound thick but you'll get tired listening to an album of distortion. The bass is also distorted (a tube/valve amp in use). Then Dave Grohl will resort to shouting all the way through the song (In Your Honor was the worst), so his added edgy raspy voice will be added to the distortion soup for maximum distortion.
Listen to some Deep Purple or Led Zeppelin and compare the snare sound to the Foos. Led Zeppelin is a good example because even though Led Zep 1's drums were only recorded with 2 (or 3) microphones, there is real dynamics in the drumming (listen to how the drums come in on Good Times Bad Times). You can tell how few microphones were used because the panning on the drum fill for the toms at the end of How Many More Times pans the entire drumkit, not just the toms. Modern setups would perhaps have close-miked drums panned around the stereo image so as the drummer does a fill on the toms, it doesn't take the entire kit with it (8 microphones for a kit, one for each drum + hi hat and overheads).
Wait, the audio in YouTube doesn't come from the CD? I never suspected that. I thought the article referred to that they took the audio, sourced from the CD, and lowered the volume if it was too loud (which obviously can't bring the dynamic range back).
That's what they're doing to the music that people upload, yes. Nobody's literally sticking a CD into a YouTube server, so what's on the CD is kind of beside the point.
The parent poster's observation is that labels are uploading different, less-compressed mixes to YouTube now than the mix they release on CD, so that they sound better given YouTube's normalization.
The second video you link is 240p which is always going to have awful quality. You can't really use that for a comparison (unless I guess you force the first one to 240p too).
Both of your links ([2] and [3]) are to YouTube; if they are normalizing the audio (as claimed by the author), then you shouldn't be able to hear a difference, right?
The examples you've cited look to be mastered (or at least sourced) differently. True loudness normalization should not affect the short term dynamic range of the material. It simply looks at the average level and lowers or raises it to some preset point. YT could be doing inter-song normalization, i.e. adjusting parts within the song to the same overall level - but this is not a good thing as it would affect properly mastered tracks too.
Your example seems to show a completely clipped sample (in the CD) and a more reasonably mastered copy (in the YT clip) with greater dynamic range. Once the dynamic range has been destroyed through compression (not digital compression, I'm speaking of gain compression http://en.wikipedia.org/wiki/Gain_compression) it would difficult or nearly impossible to restore without affecting the quality.
My guess - in this instance, I think the CD and YT clips come from different masters.
Good. I absolutely hate videos that are so low volume that I have to turn Youtube up to 100%, and then adjust volume in Windows too, to hear anything. And then imagine finding one of the louder videos next.
I've tried to give people advice on volume levels, but it feels like no one is willing to do it. Hopefully Twitch will do something similar soon, too.
I think that is exactly the motivation: avoiding the random shifts in volume between videos. Pushing back against the loudness war was probably just a bonus.
It's like watching a series on TV and then getting blasted with EXTREMELY LOUD COMMERCIAL BREAKS (and they do that shit on purpose too).
That's what this article means by "loudness". They're talking about compressing the dynamic range so that the average power is higher, not increasing the peak power.
Loudness of course isn't new to commercials, but recently I noticed that some TV shows are also compressed into oblivion, resulting in a very noticeable pumping effect between dialogue and the ambient noise in-between.
They passed a similar law here in the US (http://www.fcc.gov/encyclopedia/loud-commercials) but unfortunately, it did not help very much because sound engineers use tricks that make the commercials meet the technical requirements while still being loud to the human ear. For example, they mix a loud opening and a quieter closing so the overall decibel level averages out but you still get startled when the commercial starts.
Related to the sibling comment here on the CALM act - who thinks up these names? Is there a committee in some basement somewhere that gets told the gist of each act, and has to think up a suitable acronym for them?
I hear you. In one way it's weird they only started doing it now: any somewhat decent audio player has this functionality built-in or at least available via plugins.
Yet I feel it would be more fair to all listeners and all people creating the music if everyone at least had the choice to hear the music as it was made to be heard originally.
I don't get it. Loudness normalization past the mastering stage (i.e. at the audio track uploaded by the label) will only make all audio louder or quiter, it can't increase the actual dynamic range of the track. Furthermore, the DR meter can easily produce bogus results if the equalization changes or with non-lossless joint-stereo encoding algorithms.
Humans like loud music, so louder music sounds better and sell more. So, how do you make your music MAXIMUM LOUD? Well, our digital formats have a setting for how loud any given sound is, but you can't just set it to the maximum value because then all sounds in your track will be equally loud. Or can you? Not all parts of a cymbal "phish" is equally loud, so setting it all to max will distort the sound (sacrifice dynamic range), but will consumers really care?
Youtube, by adjusting all tracks to have the same average loudness, is basically saying "to those of you who would give up essential dynamic range, to purchase a little temporary loudness, you deserve neither dynamic range nor loudness"
The hope is that eventually labels will quit sacrificing dynamic range for loudness if all our digital music sources set the average loudness to be the same for all tracks.
Applying Replaygain or EBU R128 loudness equalization will not change the perception of dynamic range and, all other things being equal will actually slightly decrease the DR rating. And TFA is not talking about Replaygain-like algorithms from the description.
Yes, that's true. But in the long run, with YouTube being such an important player, the results should be positive.
The record company's mixers have been escalating a loudness war for years, in an effort to make the most impression on listeners. And the result of that is poor sound quality.
Now YouTube comes along and says "we're resetting you all back to a standard overall volume". Suddenly all that escalating loudness hasn't accomplished anything (or very little; probably there are ways to game the system a little), at least when listened to through YouTube, which as we said is hugely important.
If they're no longer able to compete in loudness, and the vain attempt to do so damages the quality of the recording, then in the long run the mixers ought to quit doing that. It's not going to save Death Magnetic, it's too late for that, but maybe we can reclaim fidelity in future recordings.
Although the tracks have been normalised to have the same average loudness, the more aggressively compressed tracks that have less dynamic range will still sound louder at this lower level. I don't see how what YouTube is doing is going to help.
If the lower dynamic range tracks sound louder even after normalization then the normalization algorithm is flawed. ReplayGain weights the energy by frequency to better match perceived loudness, and it does a reasonably good job. Other algorithms might do an even better job at matching human perception.
Those two things have a pretty similar loudness(A-weighted RMS) and yet the compressed one sounds louder than the measured difference - ~3dB A-weighted which is not obvious to perceive untrained.
As a test - download them both, amplify the Uncompressed one at -2.7dB and listen to them again. They have the same A-weighted RMS yet the Uncompressed one sounds louder.
As I understand it, tracks with more dynamic range will have louder loud's and quieter quiet's, while those with low dynamic range will have less of either.
That's true but most people are used to a smaller DR(up to a point of course but that point is way out) and still prefer the lower DR even after normalization.
So, actually, there's not much evidence that points to this. We see albums with good dynamic range selling just as well as ones with shitty dynamic range that is louder.
The FM radio has long normalized volume for songs - so the only places this mattered was on purchased CDs.
It's some weird feedback loop. The initial input was true - to a point, humans think louder things sound better - but that idea has now been as distorted as the audio tracks they're putting out.
I believe what makes the author excited is the fact that Youtube makes this normalization in the first place, means that a properly mastered track and a track mastered for maximum loudness, will play at roughly the same volume. The benefit is that tracks where the dynamic range have not been compressed, will sound far better than the loudness tracks.
In other words, there is no reason for the studios to master for loudness (at least for Youtube audio).
Dynamic range compression (DRC) is destructive so of course the normalisation cannot repair the damage. But the point is there is no longer any reason to use DRC simply to try to increase the playback level because Youtube will undo any attempt. DRC now simply reduces the dynamic range, making music sound worse, not louder.
I came here to say the same thing. This move probably has everything to do with making sure people don't have to constantly change the volume when shifting between music videos and random amateur cat videos. The simple fact is YouTube can't change the average loudness of a given audio track without applying some form of 'destructive' processing, either compression or expansion. A more dynamic track will always be on average, quieter than a less dynamic track. Re-processing a bunch of audio tracks with just normalization does not change their average loudness, unless of course they are selectively applying compression to audio they find in need of a boost.
Indeed. Perceived loudness is a complex mix of signal level, dynamic range, and frequency content. Perfectly undoing what was done in the mastering studio is not only difficult but maybe impossible.
The loudness war isn't dead until people stop brick-walling their mixes at the mastering stage.
You can adjust playback volume between tracks all you want, but if they all have like 6dB of dynamic range, they're going to sound like garbage.
You're not really eliminating the loudness wars unless you find a way to Command-Z all of that ridiculous limiting (or convince people not to do it in the first place).
I wonder if this is about ending the "loudness war", or if it's more about helping amateur content with nearly inaudible levels.
All the good reasons for compressing music or audio in general certainly apply to YouTube. The average viewer will probably listen through laptop speakers in a noisy environment, and a large dynamic range is not helpful to them.
Also, does this mean they actually lower the volume of high RMS/LUFS videos?
> The average viewer will probably listen through laptop speakers in a noisy environment, and a large dynamic range is not helpful to them.
If you use compression on the assumption that a chunk of your audience will be listening on shitty speakers you ensure that anyone not listening on shitty speakers also gets a shitty experience. Perhaps Google would like YouTube to be a non-shitty experience for people with nice speakers (or headphones or whatever).
This is of course what the "war" is about at its core - do we prefer to to sound OK everywhere, but not great anywhere, or do we prefer to sound great for some people, but be inaudible to others.
The right answer depends on the content, context and audience. The choice is with the content creator, not with YouTube, and for many content creators optimizing for people with bad speakers is the right choice. Lowering the volume is not going to make "shitty" compression unshitty.
If they can offer several different tiers of video quality, they could do the same with the audio. Imagine a set of options that not only offers audio versions optimized for great headphones or optimized for phone speakers in a public space, but educates the (interested) user about the difference!
> anyone not listening on shitty speakers also gets a shitty experience
This hyperbole serves no one. Modern limiters are tremendously transparent, and only people who are listening for it will notice the increase in compression.
That said, there is a case to be made for listener fatigue. Personally, I'm not on youtube for extended periods where that would be a problem, but I can see how it might be.
Further, as a musician and engineer, I have long had to live with the fact that none of the streaming services (not even iTunes the app) are transparent. After having spent days on a mix and master, it's quite a shame I don't get to decide how the exact final product is sonically presented.
> This hyperbole serves no one. Modern limiters are tremendously transparent, and only people who are listening for it will notice the increase in compression.
Agree. Though, I'd add a qualifier - "...are tremendously transparent when used correctly..."
The one thing even audiophiles with the best speakers on the planet need to realize is that no speakers can approach the dynamic range of what modern digital audio have to offer us. Therefore, some tasteful compression (or limiting) will be needed, or the signal will launch the speaker cones out of their baskets.
Google's intent doesn't really matter (I think you're right, I don't see it as a targeted offensive against the loudness war), it's the response from labels that does.
Compressing music (in the loudness sense) at the source is never a good thing for the consumer; if it is required for a specific end user case then it should be done at that point rather than ruin music for every use case.
I don't buy for a second that it's necessary or helpful with the music that it is done to; which rarely has quiet passages. Compare the two masterings of Pharrell's "happy" I posted, the less compressed version sounds far better even on my 11" laptop's speakers, which should be pretty much 'as bad as it gets'.
It's probably about making their autoplay work. If you listen to a playlist, or just another video, you shouldn't need to adjust the volume constantly or risk blowing out your speakers.
I had similar thoughts, but rather if this will shift musical taste in general. Hyperbolically spoken will the best (popular) musician and not the best sound engineer suceed.
I don't mind the actual tracks having various mixing/mastering. I do mind when a commercial is compressed/mastered with the intent to be perceived louder to stand out from the actual music/video that you are watching.
Stuff like that bugs me too, and is one of the reasons I have no compunction about pirating content.
VLC supports a dynamic range audio compressor, meaning you can configure it to make the quiet parts loud, and the loud parts quiet.
If they want to play those bullshit games with the volume of commercials, then I'll play dirty too by using pirated content that has every single commercial removed, and will employ the dynamic range compressor as well to make sure the quiet parts of shows are audible and the loud parts aren't irritating. I just feel sorry for the poor saps who have to tolerate these bullshit shenanigans when even the regulators have been calling out inexplicably loud commercials.
Then again with how cheap Raspberry Pis are, maybe we're closer to everyone using pirated content than we are to get Hollywood and producers of commercials to stop the BS.
In theory, yes, in practice, that just means the commercials are compressed to the single loudest part of the TV show they accompany. Net effect: Nil. The commercial still sounds louder than everything else.
The sole reason I started using adblock, was when I lived in Istanbul for couple of months, their ads were out of hands. Imagine having to control the volume, every time ad plays, in playlists or in front of videos.
It's been suggested that this change guarantees that the ads can be louder than the content. After all, from Google's point of view the ads are the important thing that make money.
Now if only they would add an option to do this to more than just music videos. There's still millions upon millions of videos on Youtube with audio volumes all over the place. It seems like iMovie produced content is especially loud but that's probably anecdotal.
I am not sure it's the final nail. I think at some point they may start playing the commercials louder. I had to skip a lot of Czech commercials simply because they were too loud.
If YouTube compensates for this (lowers the volume of tracks that have been mastered loud), then it removes the original motivation for doing so.
This is why the 'Happy' YouTube mix is less loud than the CD version linked to - the author asserts that the studio has done this because the incentive to apply dynamic compression on YT is now gone...
This makes me happy. My daughter watches a lot of youtube and it's a pain to constantly jump up to turn up the volume or turn down the volume after every video change. Especially if you go to sleep to a video and the next video in the playlist blows you out of your slumber because it's mix is hot compared to the previous one.
This is a good move. Sorry music guys, but on the whole, this is a win.
Volume is one thing but quality should be also important for music videos, and even HD VEVO videos has noticeable worse quality than a 256kbps MP3. Probably due to the re-encoding, and the lower quality settings at YouTube. So YouTube is still not good for music listening if you are sensitive to the quality.
Does anyone actually have a source that shows YouTube has actually implemented technology that is doing this? Contrary to popular belief music videos are still encoded and mastered for delivery to VEVO by a few companies. Universal Music also has one of the largest catalog of popular artists. VEVO transcodes them and publishes to their platform. If they let Google transcode them or not I do not know. Depends on the partnership.
Semi-related: Has anyone else noticed how HD channels have their volume levels set higher than non-HD? This is quite noticeable on the BSKYB service in the UK. I wondered if it applied for other suppliers.
Higher signal to noise ratio. Dr. Barry Blesser suggests that "raising the loudness of music, like a double shot of whisky, elevates the intensity of the experience". Listeners undergo significant, measurable changes in mind-body states and Blesser reckons that "loud music is simply a stronger stimulant than soft music".
And this article argues in depth, that the relationship between actual/perceived dynamic range is ultimately an artistic choice, yielding different feels. What people who complain about loudness are really objecting to is the improper use of these techniques:
http://www.soundonsound.com/sos/sep11/articles/loudness.htm
On the other hand, Youtube plays commercials louder than the videos (this is pretty jarring if you use Youtube playlists for background music, as many people do). So I fear this is less due to an altruistic desire to improve the music landscape, and more in order to give themselves headroom to outshout the music...
I'm not sure what he's alluding to. It may be that they've done simple loudness normalisation instead of one of the more preferable, advanced techniques, like ReplayGain 1.0, R128, or BS.1770 (/ReplayGain 2.0)?
Some of these techniques appear to (optionally) take the entire album into account when conducting normalization.
Maybe that's it. Due to its nature, YouTube tends to lend itself to album fragmentation. It's not uncommon for people to have an entire album on their playlist where no two tracks were uploaded together, let alone even share the same uploader.
In this scenario, barring some sort of mitigation strategy, any seamless transitions between tracks would most likely be destroyed. Each track would have been normalized individually.
I don't know what he's alluding to, even though I'm very familiar with the issue and quite familiar with the ITU loudness standard. I posted mainly because I had noticed the change myself over the last few months.
Maybe he means the fact that if you don't have an adblocker YouTube ads seem to be as loud as the ones on TV used to be. It might be that this will now fall under the aegis of the FCC :) but I'm not really sure about that. Video ads on Youtube are a betrayal of the Google Adwords approach anyway, and the people who gave them the green light should be shot, but with very small calibre bullets so we can keep shooting them over and over until they get the message about how annoying advertising in temporal media is.
I got really lazy recently and uploaded a demo video of an amplifier I built without adjusting the audio level, and it made for a very quiet video. I wonder if it has been normalized for playback by youtube yet.
I've never understood the point of the loudness war. Aren't the best-selling albums of all time all reasonably uncompressed? If everybody's so afraid of failure, why do they ignore success?
Perhaps it is just that they are music videos and not CD releases? They must be mastered differently anyway as they often have extra sounds or a different length, right?
Youtube can do this becuase they have the whole video before they transcode it, and by then they know its average loudness. TV sets receive the content as they play it, so they can only do this if they play the content with some delay, since they otherwise can't know in advance what the average loudness is going to be. They can of cause cap it at some maximal loudness, but that would also cap the loudness of the loudest section of movies as well, and by that reducing the dynamic range.
Not sure if it's still a thing but I used to have a TV that did this. Basically if the overall volume increased more than a certain amount it would drop it a little bit. I believe you could set it by the input you were using so it applied to cable TV but not your A/V input (so it didn't mess with movies on DVD for example).
Not sure why this is downvoted and at the very bottom of the thread.
If the article is true, youtube/google is using their market power in one sector to limit artistic and creative decisions in another sector. If mastering houses simply make a youtube mix next to their CD mix than the title is wrong.
None of us noticed; it seems like the labels and sound engineers may have; take this image comparing the dynamic range of the CD and youtube releases of Pharrell's "Happy"[1]- I thought at the time it might just have been a mishap or random occurance, but now it looks like it was a very deliberate move.
So, rather than ending the loudness wars on CDs and radio, we might end up long-term with the ludicrous situation where CD's sound worse than technically inferior Youtube videos. Youtube is the new vinyl.
-
E: I added links to the YT release[2] and (presumaby) a CD-sourced video[3] for anyone who wants to listen to the difference. Pay attention to the claps and drums (but I think it's obvious even on my laptop speakers).
1. https://imgur.com/KRj5DDS
2. https://youtu.be/ZbZSe6N_BXs
3. https://youtu.be/nVH6RcLHqMo