What are some of the specifics for mastering dance music to 45 rpm vinyl? What are some of the challenges faced in that particular scenario? Do you have any personal favourite vinyl masters in that genre?
I master professionally and also have worked along side other top grammy-winning mastering engineers. NONE of them master at -14. You just need to make your master sound awesome and competitive. That usually means between -7 and -10 LUFS depending on the the artist and genre. In my experience -8 LUFS is really a sweet spot IF you can get there intelligently. Intelligently usually means using a combination of high quality saturation, light compression, EQ, selective peak editing, gentle stages of limiting, and discreet multi band compression or dynamic eq. Give yourself a little breathing room for the peaks...Around -0.5 on the output works just fine. One of the top mastering guys I have worked with outputs at 0.0 and told me he NEVER goes below -0.3 on the output setting. Once your master sounds amazing, print and send to the client. It will sound amazing everywhere. If not, that's someone else's fault.
Anything over the streaming service's loudness normalization doesn't matter. For example, unless you turn the setting off, every song will average -14LUFS on spotify.
I agree with this EXCEPT for the suggested range of -10 or higher. In my experience -10 LUFS is plenty loud enough even in the most aggressive genres, and will sound great online too. Balance everything else musically relative to that and you’ll be in great shape. Pushing up to -8 or higher makes it much harder to get a good result, and doesn’t really give any benefits, especially not after lossy encoding, as Jonathan describes.
Wouldn’t a track at that loudness get turned down anyways? Then why push it so loud? Wouldn’t it be better to keep the dynamics with a little less limiting?
I've actually experimented with loudness on streaming services particularly Spotify. I released two tracks one mastered at -14 LUFS and other at -9 LUFS and I found that the -14 LUFS track DID NOT get through normalisation and sounded very soft in comparison to the second track. Whereas the -9 LUFS was slightly normalised but still sounded great! So if you're going -14 or -12 it is probably NOT going to sound good or loud!
That is a great perspective. I am prepping an album for release for streaming and I exported copies of the tracks at -14LUFS, does anyone else have insight on their experience with this as I know it could be genre specific however if I can push another db or 1.5db out of my tracks going to -11 even -9 I would prefer to do that. Thanks.
@@Silentnarcotic make it louder Scott. you won't regret it. unless you make classical music. 😉 if you don't trust me, just listen to what Streaky got to say on this topic: th-cam.com/video/MIQuIrbTm5A/w-d-xo.html . and Streaky knows a few things about mastering.
The target that streaming services use is INTEGRATED LUFS. if you're just setting the target to -14 LUFS on the chorus, you'll likely have much more room to push the limiter. Integrated LUFS is the average over the whole track
I think what would benefit a lot of folks is you to take songs you have mastered from different genres and show them using insight 2. Explain what your fundamental thinking was for your level and treatment. That way we're not talking about general recommendations for levels etc. but a process that would certainly help a lot of us. What caused one tune to be "louder" than another...what was your general thinking on how much to limit or compress certain styles of songs...
I've noticed a trend in loudness and mastering for different genres, and I'm curious if you could dive into the topic more. The traditional role of a mastering engineer is to be transparent and do no harm, eg: when the ME has done their job well, you (or the audience) won't "know we exist". However, with the rise of EDM I'm especially noticing the desire of edm artists and producers treating mastering as a final sound design/mixing element. I've noticed this particular approach bleed into other genres like post-rock and metal with certain acts/groups, and am curious what your take is on this newer approach (though I can point back to records as far back as the late 90's that were pushing loudness practices that were utilizing similar approaches).
Also if your song is played as part of the album, Spotify turns off the normalization (or maybe normalizes the whole album as a unit), but does use it when played as a single with various other tracks.
I use an aggregator, CD Baby, to release my music for streaming and for purchase. I can only submit one file, 44100Khz - 16 Bit wave file, for each release. I use Ozone 8 for mastering and use the master assistant set to CD High, I then set the maximizer threshold to -1.0 dB and the loudness to -10 LUFS. It seems to work for all the services that I can check (I don't have accounts on all the streaming services) but I'm wondering if there is a better way to make one file work for all!
Is there any known test on Spotify or other streaming service where we can hear -14 vs -8 and be able to compare to itself and other songs. Ideally multiple songs and genres. If not - there’s a project!
We've been conducting tests lately and noticed how it is true that platforms will, for a fact, 'turn down' your music. It is definitely there. I guess the whole point is not killing/squashing the soul out of your music if you're exclusively going to stream it. -11 through -9 seem to be working pretty good, even if you're doing heavy rock or similar genres that will sound compressed per se. This whole loudness normalization has taken it's time to settle but to be completely honest, I like it. Somehow it's trying to bring a bit of life back to the tracks. The experience of listening through a playlist feels consistent.
Hello Jonathan, great series of tutorials. We all appreciate the clear instructions and well-organized thought process behind everything you say. My question is: I am really struggling with getting good competitive results during "mastering" for tracks that are utilizing a lot of not "so great" virtual instruments. I do understand what's going on here and the nature of the problem (sound source), but does your approach to low end treatment change between virtual instruments vs actual acoustic instrument recordings? Also, in this kind of scenarios how to you talk to the client and make them understand that there are limitations to what you can actually improve? Looking for emotional, physical, and technical points of view if you could please take a moment to share your thoughts on the point of discussion. Thanks in advance and all the best, Ioannis
Great video..!! .i did a master beofre with a -9.00 LUFS CD level and Spotify of course loweered the volume....my question is: Do I always have to semd it -13 dbm for streaming?? I dod it before amd still lowerr the volume so how can we put a song that sounds realy good in Spotify???? Thanks 👍🔥🔥🔥
Thanks for the video. What about a video to cover why and when you would use a dynamic EQ over a multiband compressor? The pro's and con's of each situation? Cheers
Thanks for this amazingly free content iZotope! I'd like to ask if it's necessary to use the dither option on the Maximizer as part of the final plugin in our chain in order to export to 24 bit. Are there any benefits? Thanks!
Your DAW most likely process the audio at 32 or 64 bit floating point internally (regardless of the bit depth of the files), so yes it's always recommended to dither when exporting to something lower than that.
Hi Jonathan, at 6.00 you adjusted the threshold to 8.5dB, why? Surely that will affect the final level and you made no mention of it...... just trying to understand. Thx
Hi guys, The thing is that I measure loudness from youtube music through Youlean for some major songs, every single one is a lot above -14 lufs integrated, with true peak well above -1. Usually I measured around -9 lufs and +0.5 (and above to +1) true peak. How does this explained according to spotify/youtube loudness policies? I'm pretty sure about my measurements that I did with different methods (streaming, downloading,...) on major hits. Cheers C.
This might get lost in the many comments, but my question is, what happens with "back catalog" and older recordings? Are they all being remastered for streaming, or just encoded as-is? Anyone know?
One more question about dither. I have Ozone 8. If I do not check off the 'enable dither' box when I export, does that mean any settings I have in the dither section are ignored? And also, if I DO check off the box, does it use the settings from that dither section. I'm assuming that's the way it works, but wanted to make sure. Based on your last answer, I now understand why dither is not saved when you save a preset!
Check out our dithering guide and help article here on our site: support.izotope.com/hc/en-us/articles/360023817574-Ozone-dithering If you still have questions, please contact our support team!
Another great, well explained video.. Thanks for doing these, sure helps out with the struggle of loudness...mostly being consistant with other completed mixes.
May find some way to trick the level tampering Bots used by streaming service into not tampering with the masters. Some kind of front end or decoy in a plug in or app maybe?
How come (apart from legacy) Ozone controls/meters are in dB not LUFS ? Necessitates (downstream) use of separate meters and/or Loudness-enforcing modules (as in RX) ? Would it be better to have Ozone mode (metric-space) selectable eg various Loudness standards as well as Legacy?
One thing you didn't mention. Always master from a 32-bit file. It keeps the high end much more intact when converting to MP3 than equivalent 16-bit file
@@Willigrow As I explain above, i discovered accidentally that creating an mp3 from a 32-bit file provides a much better result with headroom and high end. Take '32-bit' mp3s that looked completely compressed to f**k -- if you use software to reduce the gain by a few dB you will usually then observe fluctuations in the peaks which suggest there was substantial headroom retained and not just squared off waveform peaks. The high-end is retaind much better with mp3s made from a 32-bit file, regardless of bits used during recording
@@annother3350 So if I understand correctly, you export your master in 32bit float regardless of the format you record/mix in, and you convert that file to mp3? I'll have to try, I just don't see how helpful can that be when we know you can't create new data by just by exporting a lower resolution file in higher resolution.
@@Willigrow It's something to do with 32-bit files converting to mp3 better than 24-bit files. You're not creating new data but the mp3 seems to retain more data from a 32-bit file than 24-bit or 16-bit conversion. My experiments are in no way scientific but I've had some suprising results -- I have been using Adobe Audition for the conversions
@wili grow you dont necessary need to record in 32-bit but at least mix it in 32-bit because of all the high end that you might create with some efx (reverb, etc) and of course master it with the same bit depth and only when exporting it in wav or mp3 you will see the difference @ann other is talking about
Vinyl is its own beast and quite complex. You have to worry about things like bass notes hitting too quick that might make the needle jump. You got some serious limitations that are very specific for vinyl. Get a professional service to do that part I'd say.
What should we consider about the loudness normalization for all the songs in a full length album? (Album with both high energy songs and soft ballads)
Super late reply, but Spotify will normalize an album as a whole, preserving the relative loudness between the songs. Release the songs separately as singles and they will be normalized individually.
Is there any way to access these in full resolution? Ironically I think the TH-cam audio encoding is compressing the audio so much I'm having trouble hearing encoding artifact differences.
Bro, I always say the same thing! It’s like saying TV is bad, but then putting your ad on TV. Lol... I love the series though. Always gives me something to think about and go explore on my own. Keep up the great work @IZT
Is there a right way to measure dynamic range? There are some plugins out there like DynaMeter which is looking at the relationship between Peak and RMS. It presents a target dynamic range based on averages from streaming services. Of course, adjustments should be made for each song, but what are your thoughts on a target dynamic range meter?
The artists can inform the audience. It does not have to be that we just assume the audience will always prefer louder. We could easily shift the way people talk away from loudness being so important. We just have to stop the tendency to overly compress and make tracks louder and talk about the benefits to dynamic range and not overly compressing so that it is hard to listen to.
Additional factor (2) when deciding what headroom to leave: Historically (at least), some encoders have behaved badly (eg giving giving horrible kinds of clip) when headroom is near the maximum. I’ve heard elsewhere that a number of plugins “work at their best” within a certain levels range (and for streaming, our maximiser by definition won’t be the “final link in the chain”). Any thoughts on this?
When we are talking about loudness and the RMS number there would be a constant number for streaming platforms such as spotify. What is the exact number? And even what number is the worlds standard?
They all have different numbers and different measurments. TH-cam and Tidal use LUFS for example, Spotify uses ReplayGain, and iTunes has their own thing called Soundcheck.
The average is about -14, which is what Spotify (responsible for probably 80% of streams for most artists) uses. But they vary by a decibel or two either way.
Hopefully you are still monitoring for questions...about dither. I find myself doing it at multiple times. At the mixing stage to prepare for mastering and at the mastering stage. Is this bad? Should I dither only once?
Hey there, there might be some helpful information for you here in our free Dithering guide: support.izotope.com/hc/en-us/articles/360034054274-Free-Guide-Dithering-with-Ozone
Hello Jonathan! I just made a test using a LFUS loudness meter of 2 songs from Spotify, one of them a song I just mastered and the other was "American Idiot" from Green day, both song were on - 14LFUS but American Idiot sounds much more louder than mine even being at the same LFUS... how do I can get more loudness from my -14 LUFS masters??? help please! Thanks!!!
Maybe American Idiot is more loud than your track, and loudness penalty in this particular case has given another result. I think all this LUFS is pain in asssss
Great series. While I appreciate the “every track is different” approach would you be able to suggest a LUFF’s setting for Spotify as a general target? I’ve heard -8,-9. and-10 from different sources. Is there a level that you could recommend that both sounds good as well as not incurring significant level change ?
When I listen to Spotify I get a lot of different overall loudness levels and that is all professionally produced music. How would I or you or anyone pick what to target when all throughout history, there has never been a certain RMS level to target? There is just as big a difference between various tracks released last year as there are between newer and older tracks...
Jonathan, would you agree that if i have a reference track on x LUFS integrated, and it sounds louder than my track at the same LUFS integrated, the first thing to do is check my tonal balance? Maybe the focus of energy in my track is different as opposed to the reference?
Hi. What about mono compatible ? For streaming my max peak is - 3 db. So if any one for some reason are listning in mono there won´t be any distortion. Michael S
My tracks are mastered at a good volume (about-9 dB LUFS) but the streaming services such as Spotify keeps turning the volume down and it sounds very soft compared to other tracks on Spotify. I really don't know what to do ! I master my songs with Ozone 8 and I'm a beginner in mastering
try thinking about RMS on individual tracks as you mix. I've been looking into this and if you're not using a nice console - you should consider saturation and light limiting on your audio buses to help get more consistent levels pre master. just my .02
you might be crushing/over compressing your songs, at -9db LUFS, you might be killing your transients/dynamic range compared to other songs that don't push the limiter as hard. maybe try mastering at a lower LUF for more dynamic range as a suggestion
Question: What levels in Wav or Mp3 would you recommend for playback in a standard computer with windows 64 bit Op-system (in Windows media player). I send files to the Police in Sweden and they claim not to hear what is being said. They do not have speakers and just rely on the integrated speaker in the monitor. They won't buy decent speakers/soundcard and refuse to listen to audio with me present or in my studio since that is not their standard op-procedure. Any tips? Best Regards.
Is the "gain" using a gain plugin different and reacts differently to the limiter as compared to the input gain on the limiter itself? if yes then what is the best approach for a clean and loud master?
Each piece of software or hardware would work a little different in your context. If the limiter has a great gain stage then maybe use that or else use the gain of the main mixer if that is better. A lot goes into a clean and loud master and what you use as a gain is not enough to cover it all. Less limiting / compression / loudness will generally leave your music cleaner sounding but might sound less loud as compared to a lot of popular music that is made to be louder rather than great sounding. Seriously, 1980s production and mastering is still the best-sounding by far. We have only gotten more loud and more messy since then, at least with anything remotely popular.
Thanks for a great video. I have a question about headroom. Is it true that one should leave a couple of dB's of headroom on the 2-track for mastering? If so, why? Can't the mastering engineer just gain the signal up or down as needed, before applying processing? Thank you.
How do you address creating an album with each track having a different perceived loudness level? Is this okay? Then people have to be ready to adjust volume all the time though? Thanks!
im struggling with that as i type. for this album each song has different dynamics, different loudness factors, different density and all i could do keep playing them in my car and go back to the files and re adjust as necessary. its painstaking, but this is what i have to do
@@mrscratchhook My challenge is I am trying to create the best sounding song, and one song on my album sounds best mastered to 9 LUF and the other 14 LUF and I dont want to change either of them, they sound perfect. So how do I get gain continuity? Are you saying you are trying to bring them all close together in level? Is that best for the song? Thankfully I won't be making a physical CD, so on streaming services the levels will be balanced for me thankfully . Cheers
@@chakradocta8883 well yes for me, because im mixing and mastering hiphop/boom bap, so its less stringent on the rules, as long as they are close its ok...but if 9 and 14 sound good respectively i would just runn with that..as ling as its not to far off you should be ok...as mixers and engineers sometimes we worry and over think the most simple things...just let it be once it sounds good..there is no perfect solution, make it the best you can and let it fly!!!!...
If the music is played as an album, rather than playing each track separately, the entire album is normalized as a unit. So your dynamic contrasts between tracks will be preserved. Furthermore, from what I hear it may be that tracks quieter than -14 are not actually raised in level, at least no more than it takes to get the peaks to 0db.
I think you just want it to sound good. I wouldn’t do much different from mastering for CD. I have heard some indications that in fact quieter music is NOT increased in volume (at least not past getting the peaks to 0db). I think they are being smart and lowering loud music to -14, but leaving quieter and more dynamic music alone. So I would ignore all this and just make a good-sounding master.
I'd recommend using the new Dialogue Intelligence Meter in Insight 2. This way, using instances of Relay throughout your mix, you can meter the loudness of the dialogue compared to the loudness of the rest of the mix to see if anything is getting buried or won't be heard at the right level. www.izotope.com/en/products/insight/features/intelligibility-meter.html . There are many standard broadcast settings and a gated dialogue loudness spec preset for Netflix in Insight 2. Feel free to download and try out the demo on our site: www.izotope.com/en/products/downloads/insight.html
Don't. Movies have been going the opposite way of music. In music, the sound has been getting more and more compressed and louder RMS over the years. In Movies the trend is to use as wide a dynamic range as possible so a whisper is as soft as a whisper while and explosion is as loud as an explosion. Movies used to be more compressed and have been getting less compressed while music was less compressed and has been becoming more compressed over the years.
@@adamb8657 There was a good comparison in this video: th-cam.com/video/EiRMYoqU3ys/w-d-xo.html I have not tried this to a lot of examples and I am not a pro by the way. On your specific observation, maybe it has something to do with the mix, etc.
Can i master the qlimax with different lufs (ex: -7 or -6) and make big headroom for the other elements? How streaming services will translate the audio? Is this a correct way to master a track? Thank really enjoying your advices!
TheBlackhat97 checkout the free loudness plug called Youlean loudness meter. It allows you to easily understand which sites stream at which volume. Then you can decide how you want to master your loudness.
tons of questions here with no aswers,now let me join. Iv searched all over the internet but didnt find anything about Facbook''s LUFS. Tried to publish music video with -12 to -16 Luf but on playback on mobile phones specialy iphones. it sounds so intense and distorted at full volume. however, on somebody adverstisement music production they dont distort on full volume the sweatness still maintained. Why is that?
@Izotope I would love to do some review's of the new plug-ins, I don't have a big following but trying to get product sponsor's I run a small business & pay out of pocket for all my music tool's.
Luckily none of the artists I've produced, mixed and mastered these days are big enough to worry about "competitive loudness on streaming services". They just make the music they want and the finished songs sound the way they want them to sound. When it comes over the local radio stations they've got no complaints. I guess being the little guy alleviates a lot of the BS and hoops the big shots have to deal with. That makes me happy.
I agree that mastering for spesific medium isn't that great of an idea. Think ten years forward; will we still use Spotify? If not, your master could sound held back or otherwisely out of place Will we talk about "Spotify-sound of the 2010"?
You would have one master that only gets uploaded to Spotify. You make another one for vinyl. Another for radio... Those would not be the main master which you would keep at 192Khz, 24Bit, uncompressed. Especially for vinyl you have to make a specific version as there are hard limitations on vinyl that do not apply to any other format.
When you say nerd.. Makes me laughing bcs remembering me when do right click on youtube screen after upload the video just for see content loudness . What happen when loosless audio compression not exist anymore, technology always up.. 5G, 6G, 7G, 10G??.. and probably next 2 years 50TB on single SSD. When you download your own content from your own youtube account the audio comeback normal, not reduce/compress like when you playback on youtube.
What is the preferred loudness for Spotify? I have heard that setting your reference at -11 LUFS and going according to that. So its somewhere in the -12 -13 range or ?
He mate, really good video, thanks! I just realized on my DT1770 that the gating on this video is pretty weird in terms of the Mic noise. Maybe on regular speakers or phones you dont realize, but it`s kind of annoying that the noise goes whaap whaap whaaawhapwhap. Thanks anyway, great stuff! SAiZ
@@nascimentol835 there must be a reason why my track with same numbers "on paper" of another track sounds much more quieter and smeared than that, but apparently no one wants to tell us yet. fucking conspiracies in mix / mastering. ridiculous
you most likely have too much low end in your mix. control it and suddenly you can turn up your maximizer way more and your master appears way louder.✌️
@@ArthurGobber that's basic trick - we all do that. still tracks are quiet compared to other ones. and i'm not only talking about levels but also sound definition.
If you compress your sound a lot then the streaming service's loudness normalizer will make it sound more quiet. If you start with something less compressed then the normalization will make it sound louder. The normalization is trying to make all songs have similar RMS volume and they are not aiming for the maximum compression obviously so if your track has such crazy RMS loudness then it will get normalized downward. I'd say to aim at what big artists do without any normalization and you will get similar results.
Streaming services are actually ending the stupid loudness wars, were going back to more dynamics and natural sound again. Thank u streaming, over compressed distorted pop is finally going to take a back seat for once.
Seeing all of the commercial releases still being mastered at -9 to even -6 LUFS I don't see the loudness war ending soon. And those who try to follow guidelines (-14LUFS for example) usually find their tracks are noticeably lower in volume compared to commercial releases.
SKRM Visuals ...and this is why artists who would rather stick to working on their art prefer to send their music to Mastering Engineers! Let us nerds worry about all these details!
Just make music brother. With a great song, it can be recorded with a tin can and string and it will not matter at all. All the tricks in the world won't fix a poor song. Spend your time playing and writing. When it comes time to record - things will fall into place.
in 3 steps, just cut anything below 20hz with a 24db hi-pass filter, be sure to check your mix is maxing about -12db RMS (with a VU meter) and finally master the file -0.6db max and -9 or -10 lufs.
Have a question for Jonathan? Reply below and we’ll address it in a future episode of Are You Listening!
What is your mastering rate?
I have a LUFS meter to measure my levels what are the ideal short term & long term levles?
Can we axpect any thought\tips\video about stem mastering as well?
What is a good strategy to mitigate for the high frequency "pixelation" when mastering for a streaming service?
What are some of the specifics for mastering dance music to 45 rpm vinyl? What are some of the challenges faced in that particular scenario? Do you have any personal favourite vinyl masters in that genre?
I master professionally and also have worked along side other top grammy-winning mastering engineers. NONE of them master at -14. You just need to make your master sound awesome and competitive. That usually means between -7 and -10 LUFS depending on the the artist and genre. In my experience -8 LUFS is really a sweet spot IF you can get there intelligently. Intelligently usually means using a combination of high quality saturation, light compression, EQ, selective peak editing, gentle stages of limiting, and discreet multi band compression or dynamic eq. Give yourself a little breathing room for the peaks...Around -0.5 on the output works just fine. One of the top mastering guys I have worked with outputs at 0.0 and told me he NEVER goes below -0.3 on the output setting. Once your master sounds amazing, print and send to the client. It will sound amazing everywhere. If not, that's someone else's fault.
Thanks a lot, Gregg!
Anything over the streaming service's loudness normalization doesn't matter. For example, unless you turn the setting off, every song will average -14LUFS on spotify.
I agree with this EXCEPT for the suggested range of -10 or higher.
In my experience -10 LUFS is plenty loud enough even in the most aggressive genres, and will sound great online too. Balance everything else musically relative to that and you’ll be in great shape.
Pushing up to -8 or higher makes it much harder to get a good result, and doesn’t really give any benefits, especially not after lossy encoding, as Jonathan describes.
Wouldn’t a track at that loudness get turned down anyways? Then why push it so loud? Wouldn’t it be better to keep the dynamics with a little less limiting?
@@bloood9 It is off by default. Hopefully no one uses that feature.
I've actually experimented with loudness on streaming services particularly Spotify. I released two tracks one mastered at -14 LUFS and other at -9 LUFS and I found that the -14 LUFS track DID NOT get through normalisation and sounded very soft in comparison to the second track. Whereas the -9 LUFS was slightly normalised but still sounded great! So if you're going -14 or -12 it is probably NOT going to sound good or loud!
That is a great perspective. I am prepping an album for release for streaming and I exported copies of the tracks at -14LUFS, does anyone else have insight on their experience with this as I know it could be genre specific however if I can push another db or 1.5db out of my tracks going to -11 even -9 I would prefer to do that. Thanks.
@@Silentnarcotic make it louder Scott. you won't regret it. unless you make classical music. 😉
if you don't trust me, just listen to what Streaky got to say on this topic: th-cam.com/video/MIQuIrbTm5A/w-d-xo.html . and Streaky knows a few things about mastering.
@@LaPlant_Mastering Thank you.
The target that streaming services use is INTEGRATED LUFS. if you're just setting the target to -14 LUFS on the chorus, you'll likely have much more room to push the limiter. Integrated LUFS is the average over the whole track
had this same issue, v disappointing when you follow guidelines that result in a soft track on Spotify...
I think what would benefit a lot of folks is you to take songs you have mastered from different genres and show them using insight 2. Explain what your fundamental thinking was for your level and treatment. That way we're not talking about general recommendations for levels etc. but a process that would certainly help a lot of us. What caused one tune to be "louder" than another...what was your general thinking on how much to limit or compress certain styles of songs...
Love that they used the White Reaper tune as the audio!
Cant get enough of this show!! Bring on the knowledge Johnathan!
I've noticed a trend in loudness and mastering for different genres, and I'm curious if you could dive into the topic more. The traditional role of a mastering engineer is to be transparent and do no harm, eg: when the ME has done their job well, you (or the audience) won't "know we exist".
However, with the rise of EDM I'm especially noticing the desire of edm artists and producers treating mastering as a final sound design/mixing element. I've noticed this particular approach bleed into other genres like post-rock and metal with certain acts/groups, and am curious what your take is on this newer approach (though I can point back to records as far back as the late 90's that were pushing loudness practices that were utilizing similar approaches).
Man this guy is a good dude. Respects the music first and foremost. Making a record is not always an exact science! Thanks for the video, bud.
Also if your song is played as part of the album, Spotify turns off the normalization (or maybe normalizes the whole album as a unit), but does use it when played as a single with various other tracks.
I use an aggregator, CD Baby, to release my music for streaming and for purchase. I can only submit one file, 44100Khz - 16 Bit wave file, for each release. I use Ozone 8 for mastering and use the master assistant set to CD High, I then set the maximizer threshold to -1.0 dB and the loudness to -10 LUFS. It seems to work for all the services that I can check (I don't have accounts on all the streaming services) but I'm wondering if there is a better way to make one file work for all!
You can send 24 bit thru Distrokid 🔥👍😃
Is there any known test on Spotify or other streaming service where we can hear -14 vs -8 and be able to compare to itself and other songs. Ideally multiple songs and genres. If not - there’s a project!
We've been conducting tests lately and noticed how it is true that platforms will, for a fact, 'turn down' your music. It is definitely there. I guess the whole point is not killing/squashing the soul out of your music if you're exclusively going to stream it. -11 through -9 seem to be working pretty good, even if you're doing heavy rock or similar genres that will sound compressed per se. This whole loudness normalization has taken it's time to settle but to be completely honest, I like it. Somehow it's trying to bring a bit of life back to the tracks. The experience of listening through a playlist feels consistent.
Hello Jonathan, great series of tutorials. We all appreciate the clear instructions and well-organized thought process behind everything you say.
My question is:
I am really struggling with getting good competitive results during "mastering" for tracks that are utilizing a lot of not "so great" virtual instruments. I do understand what's going on here and the nature of the problem (sound source), but does your approach to low end treatment change between virtual instruments vs actual acoustic instrument recordings? Also, in this kind of scenarios how to you talk to the client and make them understand that there are limitations to what you can actually improve? Looking for emotional, physical, and technical points of view if you could please take a moment to share your thoughts on the point of discussion.
Thanks in advance and all the best,
Ioannis
Great video..!! .i did a master beofre with a -9.00 LUFS CD level and Spotify of course loweered the volume....my question is: Do I always have to semd it -13 dbm for streaming?? I dod it before amd still lowerr the volume so how can we put a song that sounds realy good in Spotify???? Thanks 👍🔥🔥🔥
Thanks for the video. What about a video to cover why and when you would use a dynamic EQ over a multiband compressor? The pro's and con's of each situation? Cheers
Thanks for this amazingly free content iZotope! I'd like to ask if it's necessary to use the dither option on the Maximizer as part of the final plugin in our chain in order to export to 24 bit. Are there any benefits? Thanks!
Your DAW most likely process the audio at 32 or 64 bit floating point internally (regardless of the bit depth of the files), so yes it's always recommended to dither when exporting to something lower than that.
Hi Jonathan, at 6.00 you adjusted the threshold to 8.5dB, why? Surely that will affect the final level and you made no mention of it...... just trying to understand. Thx
9:22 Thanks Jonathan, this is very well put!
Hi guys,
The thing is that I measure loudness from youtube music through Youlean for some major songs, every single one is a lot above -14 lufs integrated, with true peak well above -1. Usually I measured around -9 lufs and +0.5 (and above to +1) true peak. How does this explained according to spotify/youtube loudness policies?
I'm pretty sure about my measurements that I did with different methods (streaming, downloading,...) on major hits.
Cheers
C.
love these classes
Thanks for the great and easy to understand explanation...cheers
okay what song is used there, its so good
This might get lost in the many comments, but my question is, what happens with "back catalog" and older recordings? Are they all being remastered for streaming, or just encoded as-is? Anyone know?
Amazing presentation And comments
What program/plugin and presets are being shown at the 0:24 mark? These would be great in Ozone 9 (if they're not already there)
Wow, this is fantastic information Jonathan. Thank you so much for doing this.
Great video!
One more question about dither. I have Ozone 8. If I do not check off the 'enable dither' box when I export, does that mean any settings I have in the dither section are ignored? And also, if I DO check off the box, does it use the settings from that dither section. I'm assuming that's the way it works, but wanted to make sure. Based on your last answer, I now understand why dither is not saved when you save a preset!
Check out our dithering guide and help article here on our site:
support.izotope.com/hc/en-us/articles/360023817574-Ozone-dithering
If you still have questions, please contact our support team!
Great video answers the question why streamed music sounds lifeless if played normally usinf natuve dac of device
Another great, well explained video.. Thanks for doing these, sure helps out with the struggle of loudness...mostly being consistant with other completed mixes.
can we get a general idea of how to set our masters instead of working with a specific plugin?
This is the one video I have been waiting for!
Master for all pllatforms.And make it sound good on phones.
May find some way to trick the level tampering Bots used by streaming service into not tampering with the masters. Some kind of front end or decoy in a plug in or app maybe?
How come (apart from legacy) Ozone controls/meters are in dB not LUFS ? Necessitates (downstream) use of separate meters and/or Loudness-enforcing modules (as in RX) ? Would it be better to have Ozone mode (metric-space) selectable eg various Loudness standards as well as Legacy?
I love this show
One thing you didn't mention. Always master from a 32-bit file. It keeps the high end much more intact when converting to MP3 than equivalent 16-bit file
Makes sense if you recorded and in mixed 32-bit, most record at 24 so I don't see why would you export 32bit master.
@@Willigrow As I explain above, i discovered accidentally that creating an mp3 from a 32-bit file provides a much better result with headroom and high end. Take '32-bit' mp3s that looked completely compressed to f**k -- if you use software to reduce the gain by a few dB you will usually then observe fluctuations in the peaks which suggest there was substantial headroom retained and not just squared off waveform peaks. The high-end is retaind much better with mp3s made from a 32-bit file, regardless of bits used during recording
@@annother3350 So if I understand correctly, you export your master in 32bit float regardless of the format you record/mix in, and you convert that file to mp3? I'll have to try, I just don't see how helpful can that be when we know you can't create new data by just by exporting a lower resolution file in higher resolution.
@@Willigrow It's something to do with 32-bit files converting to mp3 better than 24-bit files. You're not creating new data but the mp3 seems to retain more data from a 32-bit file than 24-bit or 16-bit conversion. My experiments are in no way scientific but I've had some suprising results -- I have been using Adobe Audition for the conversions
@wili grow you dont necessary need to record in 32-bit but at least mix it in 32-bit because of all the high end that you might create with some efx (reverb, etc) and of course master it with the same bit depth and only when exporting it in wav or mp3 you will see the difference @ann other is talking about
thank you for these videos, I was wondering about what we should be sending to vinyl pressing plants as opposed to cd or streaming sites as well
Vinyl is its own beast and quite complex. You have to worry about things like bass notes hitting too quick that might make the needle jump. You got some serious limitations that are very specific for vinyl. Get a professional service to do that part I'd say.
What should we consider about the loudness normalization for all the songs in a full length album? (Album with both high energy songs and soft ballads)
Super late reply, but Spotify will normalize an album as a whole, preserving the relative loudness between the songs. Release the songs separately as singles and they will be normalized individually.
@@nambams Or even if they’re played separate from the album. I think it has to do with the context of playback, not how you upload.
Is there any way to access these in full resolution? Ironically I think the TH-cam audio encoding is compressing the audio so much I'm having trouble hearing encoding artifact differences.
TH-cam has the worst sound compression out there. Must be on purpose to keep their big company content producers a little happier.
Bro, I always say the same thing! It’s like saying TV is bad, but then putting your ad on TV. Lol... I love the series though. Always gives me something to think about and go explore on my own. Keep up the great work @IZT
Is there a right way to measure dynamic range? There are some plugins out there like DynaMeter which is looking at the relationship between Peak and RMS. It presents a target dynamic range based on averages from streaming services. Of course, adjustments should be made for each song, but what are your thoughts on a target dynamic range meter?
The artists can inform the audience. It does not have to be that we just assume the audience will always prefer louder. We could easily shift the way people talk away from loudness being so important. We just have to stop the tendency to overly compress and make tracks louder and talk about the benefits to dynamic range and not overly compressing so that it is hard to listen to.
HareKrishna&GodBless❤️🙏
Additional factor (2) when deciding what headroom to leave: Historically (at least), some encoders have behaved badly (eg giving giving horrible kinds of clip) when headroom is near the maximum. I’ve heard elsewhere that a number of plugins “work at their best” within a certain levels range (and for streaming, our maximiser by definition won’t be the “final link in the chain”). Any thoughts on this?
When we are talking about loudness and the RMS number there would be a constant number for streaming platforms such as spotify. What is the exact number? And even what number is the worlds standard?
They all have different numbers and different measurments. TH-cam and Tidal use LUFS for example, Spotify uses ReplayGain, and iTunes has their own thing called Soundcheck.
The average is about -14, which is what Spotify (responsible for probably 80% of streams for most artists) uses. But they vary by a decibel or two either way.
Hopefully you are still monitoring for questions...about dither. I find myself doing it at multiple times. At the mixing stage to prepare for mastering and at the mastering stage. Is this bad? Should I dither only once?
Hey there, there might be some helpful information for you here in our free Dithering guide: support.izotope.com/hc/en-us/articles/360034054274-Free-Guide-Dithering-with-Ozone
Hello Jonathan! I just made a test using a LFUS loudness meter of 2 songs from Spotify, one of them a song I just mastered and the other was "American Idiot" from Green day, both song were on - 14LFUS but American Idiot sounds much more louder than mine even being at the same LFUS... how do I can get more loudness from my -14 LUFS masters??? help please! Thanks!!!
That's because loudness is about power of music, it depends on the song and the production. Not really on the numbers of LFUS.
Maybe American Idiot is more loud than your track, and loudness penalty in this particular case has given another result. I think all this LUFS is pain in asssss
Hi Jonathan,
are you planing to make any mastering seminar in europe any time soon..?
Great. Very interesting! 👍🏻👏🏻
Great series. While I appreciate the “every track is different” approach would you be able to suggest a LUFF’s setting for Spotify as a general target? I’ve heard -8,-9. and-10 from different sources. Is there a level that you could recommend that both sounds good as well as not incurring significant level change ?
When I listen to Spotify I get a lot of different overall loudness levels and that is all professionally produced music. How would I or you or anyone pick what to target when all throughout history, there has never been a certain RMS level to target? There is just as big a difference between various tracks released last year as there are between newer and older tracks...
Does the ozone 9 master assistant streaming configuration works well for mastering songs for streaming services? Thanks...
Jonathan, would you agree that if i have a reference track on x LUFS integrated, and it sounds louder than my track at the same LUFS integrated, the first thing to do is check my tonal balance? Maybe the focus of energy in my track is different as opposed to the reference?
Sweet
Thank you so much for these videos!
TH-cam ...Master at Compression ! Make my master's at Boombox , PA , Old and New Stereos , Car Radio Amplitude & Frequency modulations , best can !
Hi. What about mono compatible ? For streaming my max peak is - 3 db. So if any one for some reason are listning in mono there won´t be any distortion. Michael S
My tracks are mastered at a good volume (about-9 dB LUFS) but the streaming services such as Spotify keeps turning the volume down and it sounds very soft compared to other tracks on Spotify. I really don't know what to do ! I master my songs with Ozone 8 and I'm a beginner in mastering
try thinking about RMS on individual tracks as you mix. I've been looking into this and if you're not using a nice console - you should consider saturation and light limiting on your audio buses to help get more consistent levels pre master. just my .02
@@CoolBreezeMusic7 thx :)
you might be crushing/over compressing your songs, at -9db LUFS, you might be killing your transients/dynamic range compared to other songs that don't push the limiter as hard. maybe try mastering at a lower LUF for more dynamic range as a suggestion
Where's the Cat Shirt ? :D Another fantastic episode, thank you Jonathan and Izotope !
Question: What levels in Wav or Mp3 would you recommend for playback in a standard computer with windows 64 bit Op-system (in Windows media player). I send files to the Police in Sweden and they claim not to hear what is being said. They do not have speakers and just rely on the integrated speaker in the monitor. They won't buy decent speakers/soundcard and refuse to listen to audio with me present or in my studio since that is not their standard op-procedure. Any tips? Best Regards.
Is the "gain" using a gain plugin different and reacts differently to the limiter as compared to the input gain on the limiter itself? if yes then what is the best approach for a clean and loud master?
Each piece of software or hardware would work a little different in your context. If the limiter has a great gain stage then maybe use that or else use the gain of the main mixer if that is better. A lot goes into a clean and loud master and what you use as a gain is not enough to cover it all. Less limiting / compression / loudness will generally leave your music cleaner sounding but might sound less loud as compared to a lot of popular music that is made to be louder rather than great sounding. Seriously, 1980s production and mastering is still the best-sounding by far. We have only gotten more loud and more messy since then, at least with anything remotely popular.
How to chahge the System sound of my PC though the Ozone? like APO Equalizer
Thanks for a great video. I have a question about headroom. Is it true that one should leave a couple of dB's of headroom on the 2-track for mastering? If so, why? Can't the mastering engineer just gain the signal up or down as needed, before applying processing? Thank you.
Yes, he can just gain the signal up or down. Do not make him waste time... Be smart and send it with -6db of headroom.
How do you address creating an album with each track having a different perceived loudness level? Is this okay? Then people have to be ready to adjust volume all the time though? Thanks!
im struggling with that as i type. for this album each song has different dynamics, different loudness factors, different density and all i could do keep playing them in my car and go back to the files and re adjust as necessary. its painstaking, but this is what i have to do
@@mrscratchhook My challenge is I am trying to create the best sounding song, and one song on my album sounds best mastered to 9 LUF and the other 14 LUF and I dont want to change either of them, they sound perfect. So how do I get gain continuity? Are you saying you are trying to bring them all close together in level? Is that best for the song? Thankfully I won't be making a physical CD, so on streaming services the levels will be balanced for me thankfully . Cheers
@@chakradocta8883 well yes for me, because im mixing and mastering hiphop/boom bap, so its less stringent on the rules, as long as they are close its ok...but if 9 and 14 sound good respectively i would just runn with that..as ling as its not to far off you should be ok...as mixers and engineers sometimes we worry and over think the most simple things...just let it be once it sounds good..there is no perfect solution, make it the best you can and let it fly!!!!...
If the music is played as an album, rather than playing each track separately, the entire album is normalized as a unit. So your dynamic contrasts between tracks will be preserved. Furthermore, from what I hear it may be that tracks quieter than -14 are not actually raised in level, at least no more than it takes to get the peaks to 0db.
Wow - I can actually follow you. Maybe I can actually learn to master my own music...
Question: What is a good master loudness for cinematic and orchestral music? Also, any good tips for mastering that type of music? Thanks!
I think you just want it to sound good. I wouldn’t do much different from mastering for CD. I have heard some indications that in fact quieter music is NOT increased in volume (at least not past getting the peaks to 0db). I think they are being smart and lowering loud music to -14, but leaving quieter and more dynamic music alone. So I would ignore all this and just make a good-sounding master.
4:18 preparing audio for streaming service
How can I use this for a movie track filled with dialogue and music
I'd recommend using the new Dialogue Intelligence Meter in Insight 2. This way, using instances of Relay throughout your mix, you can meter the loudness of the dialogue compared to the loudness of the rest of the mix to see if anything is getting buried or won't be heard at the right level. www.izotope.com/en/products/insight/features/intelligibility-meter.html .
There are many standard broadcast settings and a gated dialogue loudness spec preset for Netflix in Insight 2.
Feel free to download and try out the demo on our site: www.izotope.com/en/products/downloads/insight.html
Don't. Movies have been going the opposite way of music. In music, the sound has been getting more and more compressed and louder RMS over the years. In Movies the trend is to use as wide a dynamic range as possible so a whisper is as soft as a whisper while and explosion is as loud as an explosion. Movies used to be more compressed and have been getting less compressed while music was less compressed and has been becoming more compressed over the years.
This is awesome!
I trust you.
Skrillex is the master of loudness - 2 LUFS 😯
What modules are those in your bottom left.
Why does all the stuff I reference during master meter WAAAY over -14. It really seems wimpy and soft for rock and metal
Even the Indie Rock Songs I analyzed where over -10.
Because they were mastered for CD originally.
@@Fonkemman those masters don't sound soft on Spotify so, what to make of this
@@adamb8657 There was a good comparison in this video: th-cam.com/video/EiRMYoqU3ys/w-d-xo.html
I have not tried this to a lot of examples and I am not a pro by the way. On your specific observation, maybe it has something to do with the mix, etc.
Because of the tonal balance and peak control of the mix and master
Can i master the qlimax with different lufs (ex: -7 or -6) and make big headroom for the other elements? How streaming services will translate the audio? Is this a correct way to master a track? Thank really enjoying your advices!
TheBlackhat97 checkout the free loudness plug called Youlean loudness meter. It allows you to easily understand which sites stream at which volume. Then you can decide how you want to master your loudness.
Thank you !
tons of questions here with no aswers,now let me join. Iv searched all over the internet but didnt find anything about Facbook''s LUFS. Tried to publish music video with -12 to -16 Luf but on playback on mobile phones specialy iphones. it sounds so intense and distorted at full volume. however, on somebody adverstisement music production they dont distort on full volume the sweatness still maintained. Why is that?
Bro i noticed that my Beats Are Sounding better on SoundCloud than on TH-cam u think its Because of the codec or lufs setting? I use ozone 8
Prod By T.R.E. Oh ok Thank should I still Follow some rules for settings of lufc or the codec Settings
How can o master with high peak rms at -16lufs for Podcasts? I’m baffled
great guy!
@Izotope I would love to do some review's of the new plug-ins, I don't have a big following but trying to get product sponsor's I run a small business & pay out of pocket for all my music tool's.
How to achieve -24 LUFS on -18 DBFS without cutting any dynamics
Luckily none of the artists I've produced, mixed and mastered these days are big enough to worry about "competitive loudness on streaming services". They just make the music they want and the finished songs sound the way they want them to sound. When it comes over the local radio stations they've got no complaints. I guess being the little guy alleviates a lot of the BS and hoops the big shots have to deal with. That makes me happy.
I'm really missing the cat shirts... 🙂
Just wait 🤫
I agree that mastering for spesific medium isn't that great of an idea. Think ten years forward; will we still use Spotify? If not, your master could sound held back or otherwisely out of place
Will we talk about "Spotify-sound of the 2010"?
You would have one master that only gets uploaded to Spotify. You make another one for vinyl. Another for radio... Those would not be the main master which you would keep at 192Khz, 24Bit, uncompressed. Especially for vinyl you have to make a specific version as there are hard limitations on vinyl that do not apply to any other format.
When you say nerd.. Makes me laughing bcs remembering me when do right click on youtube screen after upload the video just for see content loudness .
What happen when loosless audio compression not exist anymore, technology always up.. 5G, 6G, 7G, 10G??.. and probably next 2 years 50TB on single SSD.
When you download your own content from your own youtube account the audio comeback normal, not reduce/compress like when you playback on youtube.
still seems like the most confusing thing in music production but maybe it's just me
song title??
0:16 video is sped up and he looks like when a cat is trying to catch something 😆
relevant website for this topic: www.loudnesspenalty.com/
What is the preferred loudness for Spotify? I have heard that setting your reference at -11 LUFS and going according to that. So its somewhere in the -12 -13 range or ?
That's what they say, but most commercial tracks are probably at 6-9 LUFS
GangstarapMadeMeDoIt Yea most commercial tracks are never that low they are always around that -8 -6 range
He mate, really good video, thanks!
I just realized on my DT1770 that the gating on this video is pretty weird in terms of the Mic noise. Maybe on regular speakers or phones you dont realize, but it`s kind of annoying that the noise goes whaap whaap whaaawhapwhap. Thanks anyway, great stuff! SAiZ
What Mic / Gate do you use? I just realized the Mic must be pretty far away so high noise level.
👍
still my super loud track (-5 lufs / -3 rms) sounds quiet on soundcloud lol
Blue Lagoon I know the feel bro
@@nascimentol835 there must be a reason why my track with same numbers "on paper" of another track sounds much more quieter and smeared than that, but apparently no one wants to tell us yet. fucking conspiracies in mix / mastering. ridiculous
you most likely have too much low end in your mix. control it and suddenly you can turn up your maximizer way more and your master appears way louder.✌️
@@ArthurGobber that's basic trick - we all do that. still tracks are quiet compared to other ones. and i'm not only talking about levels but also sound definition.
If you compress your sound a lot then the streaming service's loudness normalizer will make it sound more quiet. If you start with something less compressed then the normalization will make it sound louder. The normalization is trying to make all songs have similar RMS volume and they are not aiming for the maximum compression obviously so if your track has such crazy RMS loudness then it will get normalized downward. I'd say to aim at what big artists do without any normalization and you will get similar results.
🎶🔥
👍🏾
Streaming services are actually ending the stupid loudness wars, were going back to more dynamics and natural sound again.
Thank u streaming, over compressed distorted pop is finally going to take a back seat for once.
Seeing all of the commercial releases still being mastered at -9 to even -6 LUFS I don't see the loudness war ending soon. And those who try to follow guidelines (-14LUFS for example) usually find their tracks are noticeably lower in volume compared to commercial releases.
The idea of a lamp pointing the viewer is completely nonsense... disturbs the main theme.
I hope u vell explained in explain
This is all so confusing.
SKRM Visuals ...and this is why artists who would rather stick to working on their art prefer to send their music to Mastering Engineers! Let us nerds worry about all these details!
Just make music brother. With a great song, it can be recorded with a tin can and string and it will not matter at all. All the tricks in the world won't fix a poor song. Spend your time playing and writing. When it comes time to record - things will fall into place.
It’s confusing to me as well, but I come to this video(and anything I’m trying to learn) at least once a day and then I start to get it.
in 3 steps, just cut anything below 20hz with a 24db hi-pass filter, be sure to check your mix is maxing about -12db RMS (with a VU meter) and finally master the file -0.6db max and -9 or -10 lufs.
Tbh! withouth these guys most of the rappers would sound like tikashi 69 raw vocals
😲👏🏾