Thanks for your honest opinion. 🙂 Off topic: have you heard, tested or any kind of experience or opinion on the Kali Audio LP-6 Monitor Speakers? I need some new budget monitors and would like to know what you think of them. If possible. 🙂 Have a blessed day.
David when you did your loudness test, was it on Desktop or mobile? I also did some tests not too long ago, but found most of the loudness normalization isn’t active on desktop (even though they claim it’s there)
i shed a tear for all young engineers trying to mix/master with anything BUT the actual music/program in mind. meters don't measure quality . Be bold, have vision , commit. my 2c
David, your videos are the bomb! Here's my experience with loudness and "penalties" when it comes to streaming: There ARE no penalties. At least not on Spotify. There is ONLY end-user normalization. If the listener selects Settings > Music Quality > Normalize Volume, then Spotify will play "all songs" at "the same volume level." Try testing it with and without the normalization setting selected. There is indeed SOME audible normalization when selected, but most end-users are unaware of this setting. Therefore, most of the world hears tracks streamed on Spotify pretty much at the loudness at which they are uploaded (hence your observation that you're not hearing any "penalty"). For years I've been taking the advice of Ian Shepherd and the folks at Meter Plugs telling us that there is an "automatic" volume reduction of uploaded tracks if they exceed LUFS values which are "standard" for a given platform. These "standards" are referenced in the plugin you mentioned in this video (the plugin is boldly called "Loudness Penalty" by Meter Plugs). Unfortunately for me and 1,000's of other producers, it seems we've been told an outright lie. As for iTunes there appears to be more consistency in loudness across tracks, especially for those tracks specifically "mastered for iTunes." But unlike what folks like Ian Shepherd are telling us, tracks across TH-cam vary WILDLY in their loudness. I can't believe that this is merely because Skrillex, for example, knows how to push perceived loudness with midrange saturation, pumped up transients, and harmonic distortion. His tracks are simply LOUDER. I would really appreciate a Part 2 to this awesome video where you do a bit of research on this important topic and get back with us on the points I've mentioned here. Thanks again for the amazing education, my friend.
Isn't the Normalize Volume option on spotify on by default though? At least I remember it being so for me. And I think you're right that most listeners won't even check their settings, so whatever is the default is probably the one used by the vast majority. On the other hand I'm totally ok with people lying about it if it gets dynamics back into music. I'm just so tired of great records getting ruined by overcompressed and fatigue-inducing production. I get that a competitive loudness is important for radio and dance music but I wish producers would go for sounding good over just sounding loud even if some can do both. Pretty much every audio player has a volume control but I haven't seen any with one for bringing back dynamics, so I know what I think is the more important variable to aim for.
@nightdew I think David has it right when he says if the mix holds up, it can also be loud. Yes please, let’s strive for powerful, clear mixes with dynamic ranges that are genre-appropriate. My issue is, some genres require loudness as part of their “vibe.” Aggressive electronic genres for example. These not only have to be louder than other genres (classic rock for example) for “vibe” but also to be competitive. Following the faulty advice given by Ian Shepherd, my Spotify masters are weaker compared to genre-matched tracks. This is because I assumed there would be an automatic “penalty” if I uploaded too hot. When played alongside louder tracks in a playlist, my listeners have to turn up the volume on my tracks which makes them less competitive. On Spotify, I believe normalization isn’t on by default. End-user playback is only “penalized” if the listener knows to toggle normalization on in Settings. In the end, I love dynamics. I also need to have my tracks be competitive with what’s out there for my genre. At some point I’ll have to remaster and re-release. Glad David ignores the “penalty” hype and I’m sure his mixes and masters hold up, loud or otherwise. Ty for replying and keeping the conversation going!
@@mixbustv Yeah, I'm pretty sure you're right, I've definitely met some musicians who are convinced loudness is the only thing that matters in mastering and labels are probably even worse, but it doesn't really matter if producers and mastering engineers are going loud for themselves or due to external pressures, the end result is the same. Agreed loud isn't a problem, but stupid loud actually seems like it's the norm. So I think normalizing and penalties are probably the only way we have to change the development and it seems that it has been working somewhat since I've actually seen some improvement in dynamic ranges just in the last two years or so. Also I think it might become a big problem for the development of amateurs, bedroom producers and up and coming producers, since they rely a lot on reference tracks, and bad habits are hard to shake. Btw, thanks a lot for your videos! They're definitely among the best on youtube. Also big props for you actually being active in the comments!
@@Dr-Agonfly Yep, agreed, but I've heard way too many albums that are absolutely squashed to not think the loudness war is a big problem right now. And yeah I get that music aimed for radio or the dance floor have other considerations than other genres, so I get that it's really frustrating when your tracks sound weak in comparison. It's really a hard problem to get at. I guess one solution would be for services to let you upload a different master for every setting, but that doesn't sound too likely. I think penalizing is the only way to end the loudness war, but as long as it looks different on different platforms and even different settings it's not going to be easy for producers. A quick googling seems to suggest that the normalizing is on by default on spotify, but maybe that has changed over time? Thank you too! There's certainly arguments for both sides(if no one normalized the playing field would of course be more even) and since I mainly listen to and produce metal and rock I probably am quite a bit biased.
I have been breaking my head over this for months now, so basically TH-cam has found an algorithm for perceived loudness which you can’t beat. What I do is I have a separate master for TH-cam which has 3-4dB lesser gain on my mastering chain so that TH-cam doesn’t reduce the volume too much on my final print. There are many plugins which show you how much TH-cam will reduce the volume by so you can adjust accordingly. But even after doing all of this even the most popular tracks on TH-cam have 5-7dB of normalisation so I don’t know if those engineers are oblivious or it’s on purpose.
I like the sound of TH-cam. I don't like the sound of Spotify, because an analog compressor doesn't translate well on Spotify. It is not the loudness penalty that makes the difference. It is the data compression that makes a difference. You should data compress before you upload to TH-cam, but do not data compress before you upload to Spotify. Spotify claims to have the most advanced data compression, and I agree that it is advanced, but I think it needs improvement. I think TH-cam was smart not to try to get "HD" before the technology had been perfected. Anyway, I see it trending that everyone is asking about loudness, but I think you are asking the wrong questions. I agree with David when he says it doesn't matter what level you record at. Everyone on Spotify will have the same general loudness as everyone else on Spotify. Everyone on TH-cam will have the same general loudness as everyone else on TH-cam. Does Spotify sound different than TH-cam? Yes, but we have no control over that.
Dave, i think you didn't fully get the point here, or maybe i didn't have understand your reply… In your answer to Eri you said that never happened that your tracks were quieter than others, but it will never happen, because if you mix very loud, streaming services wil push down your mix to target level. So it's pratically impossible that your mixes will be quiter than another one, because you start from an higher volume. The point is that your tracks will sound exactly as loud as one mixed much lower than yours. But the other one, having LUFS in mind, will be more dynamic than yours, because of a superior crest factor due to less use of compression and limiting. So, IMHO, the Eri's question is: why do people still mix loud (sacrificing a lot of dynamic range) if their volume will be the same (not less) than others?
I made the mistake of releasing a single of mine mixed -14ish - I was under the impression that they were taking the K14 concept to mind. Yeah, it was annoying.
I master sometimes it’ll go over -14LUFS but not everyone listens on Spotify. Some people download tracks from Apple or Bandcamp and you don’t get the “loudness penalty” and even on Spotify you can turn it off. I don’t try and go crazy loud but I do weird music so these songs aren’t competing with pop music.
Hi David, my 2 cents about this topic. All the streaming platforms have their standard loudness reference measured in LUFS. LUFS uses the K-weight curve that refers to the equal-loudness contour (Fletcher-Munson curves) based on the human ear. Furthermore, LUFS is an average loudness of the entire program, so it's not measured for example like RMS (which measure the level in a momentary short range). So theoretically, all the normalized tracks on the streaming platforms should sound at the same perceived loudness (relative to the human perception) bearing in mind the LUFS scale. If you compress the music too much, it will be normalized and it will sounds "little" compared to another track that was mastered in the required range and it has the -14LUFS standard that allows you to keep the dynamics, the punch and the transients intact. So the track that wasn't normalized will sound with more "impact", because was mastered for that purpose. In fact many engineers do several mastering for different kind of application (CD, Streaming etc...). So even if your track is louder than the others, it will be normalized taking account of the human perceived loudness measured with the LUFS scale. The intent is to stop the Loudness War, bring more life to the music and provide a good listen experience to the audience. Hope to hear back from you on what do you think about it! Cheers
i don’t think the streaming platforms do the normalisation to stop the loudness war (though it will stop it I thinks), I think they do it to give the listener the best experience (meaning have same perceived loudness from track to track).
@@paszTube Of course one of the reason behind the normalization is to give the best audible experience to the listeners so there isn't jump in volume when you listen several programs but the LUFS legislation is regulated by EBU (European Broadcasting Union) in the R 128 recommendation which contains all the references for different kind of applications and the assumption is stop the Loudness War also as well as the noise pollution.
People are overthinking and overcomplicating this. Some people are doing like 5 different masters for each of the streaming services, redicilous. You are one of many proffesionals i heard lately that don’t care for it at all, and rightfully so. Streaky mastering on youtube has a video about it as well, sharing similar opinion.
Hello! Could you tell us please if analog vintage mixers/consoles(soundcraft,tascam etc.) are suitable for analog summing? is it better for summing audio something like Dangerous music boxes? Also could you suggest some low budget summing mixers/boxes that could do the job? Thanks!
I was penalized with quality ! Once in the past my clients uploaded 3 songs, from which 1 sounded terribly bad on YT ( like below 128 kbps mp3) we tried to upload it in a different ways/formats, but nothing had changed! Just that one track , the rest was normal.
With all of the online collaborative aspects of the music industry now days, do you attach metadata to the master before it is delivered to the client? Or do you leave that for the client when they upload to their aggregate of choice?
What I know it's this, latest foals track the runner has a music video on TH-cam, sounds better then the audio version. I had a look at the wave form, totally different masters. The video is better becouse it isn't crushed by a limiter to -5 in fact it looks like a soft clipper is how to get clear and loud on TH-cam.
@@mixbustv no the big guys are doing different mixes, you can't add back transients and dynamics from uploading to youtube, the cd versions are crashed where as audio on youtube in many cases is superior. It seems to me top mastering houses are givng more then one version of the mix back to the client. I've seen this on a few accessions now. One loud cd master and one less loud that retains transients that is used for for TV and video streaming.
5:55 "that's why at some point in history we went overboard and everything was super distorted because the industry figured out that if it's distorted then that will give our brain the impression that it's so loud" What year was that? When the industry figured that out and went overboard with it?
We learned a lot thanks to the entire team behind Metallica's Death Magnetic, released in '08. I always remember having to scroll down the volume on my iPod because it was loud enough to make my new apple earbuds distort at like 75% volume. The digital realm has changed the game a lot, too, forcing us to really take in mind harmonic content and overall balance of the arrangement.
It can take months or even years for the algorithm to penalize you. Can you go double check some of your old mixes and right click on the video and look at stats for nerds?
Thanks for your honest opinion. 🙂 Off topic: have you heard, tested or any kind of experience or opinion on the Kali Audio LP-6 Monitor Speakers? I need some new budget monitors and would like to know what you think of them. If possible. 🙂 Have a blessed day.
@@mixbustv Wow. How cool is that? Is this what they call synchronicity? 😀 Can't wait for the video. Thanks in advance.
David when you did your loudness test, was it on Desktop or mobile? I also did some tests not too long ago, but found most of the loudness normalization isn’t active on desktop (even though they claim it’s there)
i shed a tear for all young engineers trying to mix/master with anything BUT the actual music/program in mind. meters don't measure quality . Be bold, have vision , commit. my 2c
David, your videos are the bomb! Here's my experience with loudness and "penalties" when it comes to streaming: There ARE no penalties. At least not on Spotify. There is ONLY end-user normalization. If the listener selects Settings > Music Quality > Normalize Volume, then Spotify will play "all songs" at "the same volume level." Try testing it with and without the normalization setting selected. There is indeed SOME audible normalization when selected, but most end-users are unaware of this setting. Therefore, most of the world hears tracks streamed on Spotify pretty much at the loudness at which they are uploaded (hence your observation that you're not hearing any "penalty"). For years I've been taking the advice of Ian Shepherd and the folks at Meter Plugs telling us that there is an "automatic" volume reduction of uploaded tracks if they exceed LUFS values which are "standard" for a given platform. These "standards" are referenced in the plugin you mentioned in this video (the plugin is boldly called "Loudness Penalty" by Meter Plugs). Unfortunately for me and 1,000's of other producers, it seems we've been told an outright lie. As for iTunes there appears to be more consistency in loudness across tracks, especially for those tracks specifically "mastered for iTunes." But unlike what folks like Ian Shepherd are telling us, tracks across TH-cam vary WILDLY in their loudness. I can't believe that this is merely because Skrillex, for example, knows how to push perceived loudness with midrange saturation, pumped up transients, and harmonic distortion. His tracks are simply LOUDER. I would really appreciate a Part 2 to this awesome video where you do a bit of research on this important topic and get back with us on the points I've mentioned here. Thanks again for the amazing education, my friend.
Isn't the Normalize Volume option on spotify on by default though? At least I remember it being so for me. And I think you're right that most listeners won't even check their settings, so whatever is the default is probably the one used by the vast majority. On the other hand I'm totally ok with people lying about it if it gets dynamics back into music. I'm just so tired of great records getting ruined by overcompressed and fatigue-inducing production. I get that a competitive loudness is important for radio and dance music but I wish producers would go for sounding good over just sounding loud even if some can do both. Pretty much every audio player has a volume control but I haven't seen any with one for bringing back dynamics, so I know what I think is the more important variable to aim for.
@nightdew I think David has it right when he says if the mix holds up, it can also be loud. Yes please, let’s strive for powerful, clear mixes with dynamic ranges that are genre-appropriate. My issue is, some genres require loudness as part of their “vibe.” Aggressive electronic genres for example. These not only have to be louder than other genres (classic rock for example) for “vibe” but also to be competitive. Following the faulty advice given by Ian Shepherd, my Spotify masters are weaker compared to genre-matched tracks. This is because I assumed there would be an automatic “penalty” if I uploaded too hot. When played alongside louder tracks in a playlist, my listeners have to turn up the volume on my tracks which makes them less competitive. On Spotify, I believe normalization isn’t on by default. End-user playback is only “penalized” if the listener knows to toggle normalization on in Settings. In the end, I love dynamics. I also need to have my tracks be competitive with what’s out there for my genre. At some point I’ll have to remaster and re-release. Glad David ignores the “penalty” hype and I’m sure his mixes and masters hold up, loud or otherwise. Ty for replying and keeping the conversation going!
@@mixbustv Yeah, I'm pretty sure you're right, I've definitely met some musicians who are convinced loudness is the only thing that matters in mastering and labels are probably even worse, but it doesn't really matter if producers and mastering engineers are going loud for themselves or due to external pressures, the end result is the same. Agreed loud isn't a problem, but stupid loud actually seems like it's the norm. So I think normalizing and penalties are probably the only way we have to change the development and it seems that it has been working somewhat since I've actually seen some improvement in dynamic ranges just in the last two years or so. Also I think it might become a big problem for the development of amateurs, bedroom producers and up and coming producers, since they rely a lot on reference tracks, and bad habits are hard to shake. Btw, thanks a lot for your videos! They're definitely among the best on youtube. Also big props for you actually being active in the comments!
@@Dr-Agonfly Yep, agreed, but I've heard way too many albums that are absolutely squashed to not think the loudness war is a big problem right now. And yeah I get that music aimed for radio or the dance floor have other considerations than other genres, so I get that it's really frustrating when your tracks sound weak in comparison. It's really a hard problem to get at. I guess one solution would be for services to let you upload a different master for every setting, but that doesn't sound too likely. I think penalizing is the only way to end the loudness war, but as long as it looks different on different platforms and even different settings it's not going to be easy for producers. A quick googling seems to suggest that the normalizing is on by default on spotify, but maybe that has changed over time? Thank you too! There's certainly arguments for both sides(if no one normalized the playing field would of course be more even) and since I mainly listen to and produce metal and rock I probably am quite a bit biased.
I have been breaking my head over this for months now, so basically TH-cam has found an algorithm for perceived loudness which you can’t beat. What I do is I have a separate master for TH-cam which has 3-4dB lesser gain on my mastering chain so that TH-cam doesn’t reduce the volume too much on my final print. There are many plugins which show you how much TH-cam will reduce the volume by so you can adjust accordingly. But even after doing all of this even the most popular tracks on TH-cam have 5-7dB of normalisation so I don’t know if those engineers are oblivious or it’s on purpose.
This is interesting info.
I like the sound of TH-cam. I don't like the sound of Spotify, because an analog compressor doesn't translate well on Spotify. It is not the loudness penalty that makes the difference. It is the data compression that makes a difference. You should data compress before you upload to TH-cam, but do not data compress before you upload to Spotify. Spotify claims to have the most advanced data compression, and I agree that it is advanced, but I think it needs improvement. I think TH-cam was smart not to try to get "HD" before the technology had been perfected.
Anyway, I see it trending that everyone is asking about loudness, but I think you are asking the wrong questions. I agree with David when he says it doesn't matter what level you record at. Everyone on Spotify will have the same general loudness as everyone else on Spotify. Everyone on TH-cam will have the same general loudness as everyone else on TH-cam. Does Spotify sound different than TH-cam? Yes, but we have no control over that.
@@danielmankiller5098 At least Spotify has an option to disable Audio Normalisation.
@@Abhishek-kg3je true. Spotify has a lot of good qualities.
David can you do a video about Barricade Limiter/clipper plugIn? Its free But I need some clarification on some of the features. Thanks in advance
Dave, i think you didn't fully get the point here, or maybe i didn't have understand your reply… In your answer to Eri you said that never happened that your tracks were quieter than others, but it will never happen, because if you mix very loud, streaming services wil push down your mix to target level. So it's pratically impossible that your mixes will be quiter than another one, because you start from an higher volume. The point is that your tracks will sound exactly as loud as one mixed much lower than yours. But the other one, having LUFS in mind, will be more dynamic than yours, because of a superior crest factor due to less use of compression and limiting.
So, IMHO, the Eri's question is: why do people still mix loud (sacrificing a lot of dynamic range) if their volume will be the same (not less) than others?
I made the mistake of releasing a single of mine mixed -14ish - I was under the impression that they were taking the K14 concept to mind. Yeah, it was annoying.
I master sometimes it’ll go over -14LUFS but not everyone listens on Spotify. Some people download tracks from Apple or Bandcamp and you don’t get the “loudness penalty” and even on Spotify you can turn it off. I don’t try and go crazy loud but I do weird music so these songs aren’t competing with pop music.
Hi David, my 2 cents about this topic.
All the streaming platforms have their standard loudness reference measured in LUFS.
LUFS uses the K-weight curve that refers to the equal-loudness contour (Fletcher-Munson curves) based on the human ear. Furthermore, LUFS is an average loudness of the entire program, so it's not measured for example like RMS (which measure the level in a momentary short range).
So theoretically, all the normalized tracks on the streaming platforms should sound at the same perceived loudness (relative to the human perception) bearing in mind the LUFS scale.
If you compress the music too much, it will be normalized and it will sounds "little" compared to another track that was mastered in the required range and it has the -14LUFS standard that allows you to keep the dynamics, the punch and the transients intact.
So the track that wasn't normalized will sound with more "impact", because was mastered for that purpose. In fact many engineers do several mastering for different kind of application (CD, Streaming etc...).
So even if your track is louder than the others, it will be normalized taking account of the human perceived loudness measured with the LUFS scale. The intent is to stop the Loudness War, bring more life to the music and provide a good listen experience to the audience.
Hope to hear back from you on what do you think about it!
Cheers
i don’t think the streaming platforms do the normalisation to stop the loudness war (though it will stop it I thinks), I think they do it to give the listener the best experience (meaning have same perceived loudness from track to track).
@@paszTube Of course one of the reason behind the normalization is to give the best audible experience to the listeners so there isn't jump in volume when you listen several programs but the LUFS legislation is regulated by EBU (European Broadcasting Union) in the R 128 recommendation which contains all the references for different kind of applications and the assumption is stop the Loudness War also as well as the noise pollution.
People are overthinking and overcomplicating this. Some people are doing like 5 different masters for each of the streaming services, redicilous. You are one of many proffesionals i heard lately that don’t care for it at all, and rightfully so. Streaky mastering on youtube has a video about it as well, sharing similar opinion.
Hello! Could you tell us please if analog vintage mixers/consoles(soundcraft,tascam etc.) are suitable for analog summing? is it better for summing audio something like Dangerous music boxes? Also could you suggest some low budget summing mixers/boxes that could do the job? Thanks!
MixbusTV Great ! Thanks !
I was penalized with quality ! Once in the past my clients uploaded 3 songs, from which 1 sounded terribly bad on YT ( like below 128 kbps mp3) we tried to upload it in a different ways/formats, but nothing had changed! Just that one track , the rest was normal.
Thank David!
With all of the online collaborative aspects of the music industry now days, do you attach metadata to the master before it is delivered to the client? Or do you leave that for the client when they upload to their aggregate of choice?
@@mixbustv Thank you!
❤️ AMAZING !!! THX
What I know it's this, latest foals track the runner has a music video on TH-cam, sounds better then the audio version. I had a look at the wave form, totally different masters. The video is better becouse it isn't crushed by a limiter to -5 in fact it looks like a soft clipper is how to get clear and loud on TH-cam.
@@mixbustv no the big guys are doing different mixes, you can't add back transients and dynamics from uploading to youtube, the cd versions are crashed where as audio on youtube in many cases is superior. It seems to me top mastering houses are givng more then one version of the mix back to the client. I've seen this on a few accessions now. One loud cd master and one less loud that retains transients that is used for for TV and video streaming.
@@mixbustv yes it's not that common but it's happening more and more, and I'm glad as now we have access to better mastered tracks.
I found out that its better to upload mp3s that looseless format, sound quality-wise.
5:55 "that's why at some point in history we went overboard and everything was super distorted because the industry figured out that if it's distorted then that will give our brain the impression that it's so loud"
What year was that? When the industry figured that out and went overboard with it?
We learned a lot thanks to the entire team behind Metallica's Death Magnetic, released in '08. I always remember having to scroll down the volume on my iPod because it was loud enough to make my new apple earbuds distort at like 75% volume. The digital realm has changed the game a lot, too, forcing us to really take in mind harmonic content and overall balance of the arrangement.
Late answer but Californication is a great example of shitty loud mastering with soooo much distortion.
@@PACKTdotSPACE Everyone always cites that example but when i listened to that as a 12 year old i loved it, did not even think about it
It can take months or even years for the algorithm to penalize you. Can you go double check some of your old mixes and right click on the video and look at stats for nerds?
it happens as soon as the video is processed , almost instantly
-it might've taken a while back in the day, couple years ago, but not anymore
Love that shirt bro🤘🤘
You should have been wearing that shirt when you got together with Mr. Gear God. :)
Nice tee man:-)