Mastering Loud But Not Squashed
ฝัง
- เผยแพร่เมื่อ 11 ธ.ค. 2022
- Join the Sage Audio Membership here: www.sageaudio.com
If you’re new to Sage Audio, we’ve been providing industry-leading audio engineering services and education for over two decades and created this channel to help you make professional songs.
Finally someone who touches the many aspects of « loudness » and shows how it is a complex interaction. The next big step would be to show that it’s during mixing that you can prepare a track to be loud and dynamic. Great work as usual. This is one of the few no BS audio engineering channels. Thanks Sage Audio.
This video has convinced me thoroughly to send my mixes into a pro because I cannot hear a damn difference
It helps to have a good monitoring setup! If you're able to, make sure to listen on good reference level open-back mixing headphone or use studio monitors in a treated room.
bro i love how you show what doing it wrong sounds like! what an excellent vireo 10/10!
Awesome! Thanks for watching!
Great video! This helps so much. That tip about separate limiters for left and right really makes a big difference.
so much of high value information, I just want to cry tears of joy
This is some of the best information I've ever seen on mastering! Thanks Sage!
Haven't seen a mastering video this good in a long time. You've covered some things that many others haven't. Thank you :)
5:25 I just did a quick test and if you're not changing the freq response pro-q 3 in linear phase mode does not affect the signal in any way (perfect null)
Ikr what a huge fail
Yep. I've made a couple of these mistakes. Thanks for showing me a better way!
Thanks especially for the Pro L2 tips! I've never tried the Left and right separation trick.
Thank you for giving these incredible knowledge. I love you my teacher. I'm from Rwanda
I wonder if this guy can enjoy music or if he’s constantly picking apart different elements and how they’re mixed
Lol fortunately I can enjoy both the production and song as separate things. Sometimes one is good while the other isn't, but ideally both are. Thanks for watching!
SageAudio.com
@@sageaudio you're a home studio hero!
Once I took the plunge down the never-ending rabbit hole of audio production, mixing and mastering I cannot for the life of me listen to music objectively now. It's akin to finally being able to see the matrix code. Once you see it you cannot unsee it.
@@Bronwyn031 that sounds dramatic
@@Bronwyn031 try listening to the music you grew up on.
Very useful as alway Sage. thank you
amazing dude like legit god level techniques and explanation
Finally, I'm learning how to master properly... THANK YOU!
Awesome! Thanks for watching!
By the time we get to almost at the end, this demo song has lovingly took a good place in my mind. Nice work!
That's awesome! Thanks for watching
that song is mixed so nice
I wish I had better headphones and monitors to hear what I'm doing better. Thank you for making my masters better! 🙏
Excellent video as always!
This channel owns, man.
I’ve been logged in with my GoogleTV account and always watch on my worst TV in the house. If audio sounds good on this POS, something must be right. So when I can hear differences as described in the video, I consider it good advice. 😊
This is a great video! I will be referencing this video more
Thanks so much for this 💗
Thank you so much for your help Steve!
Thank you for watching!
Thank you.. this is really helpful..
I love this video, I am still having trouble getting to the 3.2 lufs without distortion. Ima try these tips out
thank you! VERY helpfull !!
Thank you
This may be a stupid question but many engineers i watch like jaycen joshua and josh gudwin put a compressor on their buses containing a studio D chorus or soundtoys microshift, presumable to even out the effect/make it more consistent (so its not more pronounced on loud vocal passages and less on quiet ones)
Any tips on how to set the compressor for that task?
Gold.
Bravo 👊🏿
I would Iove to see the maag eq in action on vocals how you would use it
Can I ask what's the title of the song on this particualr video that you used as a demo? It sounds so good to me!
I find it funny that i actually still learn from this, even tho i have to do the opposite most of the times with my music lol. I make uptempo/hardcore
Where can I listen to that song it sounds awesome
Thank you!!
Thanks for watching!
SageAudio.com
Please can you use both saturate and fabfilter l2 for mastering
Wow. That was great. Thanks! One question: How do you separate signals on the master bus to insert the Oxford Limiter on both sides? Still a noob here groping in the dark.
👍🌟✨ Thank U so much !
Whats the tune used for A/B examples?
Can i use natural phase or it has similar effects on transients? … I’ve seen luca pretolesi using linearphase mode a lot and when i say a lot i mean A LOT! Great channels i would like to hear some recommendations about the most original and not overused Synth or vst plugins great channels sage u are grear!
Hi! How to Left/Right Limiting in Ableton Live11?
WHAT ABOUT IF I HAVE IZOTOPE OZONE 10 ???
Very good video :)
Thank you!
name of the song please
My master chain goes, EQ, midside EQ, multi band compression, saturation, clipper, limiter
Hello ! So we should avoid linear phase on the pro q on the master ?
Yes. Set it to maximum setting.
I have seen in these videos you talking about using true peak and look ahead and linear phase on eq for mastering. I’m a bit conflicted now as some things in newer videos conflict with old teachings in old videos. Is this just because you have found better ways to master?
Thanks for watching Jay! In short, I think what I'm showing now is better - some of the info I shared in the past, particularly about linear-phase was misguided.
As for lookahead, it can still be a good way to avoid true-peaks, and I'd recommend using it with oversampling instead of using some form of true-peak limiting. That said, I think lowering the output slightly and introducing oversampling is the best option. Sorry for the conflicting info! I've definitely learned a fair amount more from researching new topics/diving deeper into various plugin functions over the last couple of years.
SageAudio.com
@@sageaudio Thank you for your honesty and, of course, awesome content!
Yeah my music still sucks
LMFAOOO 😭 same
I will suck until it wont anymore. Keep going my guy.
Mine still sucks too.
dw jus make the mix as good as u can. the more u record the more better u will get
Time will tell
If a song is reaching out at -11 LUFS as Integrated Value
Like only at the louder section of a song(basically the chorus)
But rest is coming under -14 LUFS(which is the best value for streaming services)
So does the streaming platforms going to compress the song or not?
Is it going to sound as I'm listening or changed?
The streaming service will take the average for the full song - so louder and quieter parts will be measured and combined to create an average loudness for the full track. If the overall average loudness for the track is quieter than -14 LUFS the streaming service will apply clean gain (usually +1dB of gain = +1 LUFS). If it's louder than -14LUFS then the streaming service will simply attenuate it (again usually -1dB of gain = -1 LUFS). So it won't be compressed, just turned up or down.
Wats the name of the song ? It’s hella vibey 🤤
u have a video for mastering for dynamics not loud
Not yet but that's a good suggestion!
@@sageaudio ok let me know when if u do thanks
I have two questions from my experience.
1. What do you recommend on the master bus, soft or hard clipping before limiting?
2. How much of such clipping should be employed? 2 dB or 3 dB?
Pick one clipper that has output gain compensation in real time so you hear it’s effect exactly and don’t get biased from loudness’s! Kclip is a good one on doing this! The signal you put through the clipper will define how much dbs you want to shave off! The more peaky transients will allow for more gain reduction! Sometimes exaggerating clipping becomes part of the sound, as it ads saturation and punch! Experiment and try your self just don’t get caught after “numbers”!
@@TheChromaticz355 Thanks a lot for the advice. I tend to use submission audio's flatline a lot and it does wonders on the master bus.
I thought that true peak limiting is a good thing, as all of our masters will be converted to some format like mp3, aac for streaming. Isnt´ true peak limiting good to avoid concversion errors ? I am a bit confused with the " avoid look ahead" , too .. :)
personally i see a bunch of different opinions on that, but the way i see it for him is that he changed the way he does things vs how he used to teach
It squishes the transients. That's why he reccomends to monitor them (with youlean loudness meter) while turning the peaks down.
No matter how you set your limiter, you can always normalize the file afterward to whatever peak value the service requires. True peak setting will reduce transients and look ahead will cause "distortion", but deciding if those effects help or hurt the song is at your discretion. If you do use look ahead, one compromise is setting the peak hold to the same length as the look ahead, which will lower the amplitude of the distortion.
Hi. So do the Oxford Inflator and Weiss Maximiser perform the same function?
Thanks for watching! They work with the same concept - bringing up quieter details, but how they go about this is different. Weiss imparts a cleaner sound, almost like transparent amplification of quieter signals. The inflator is more colorful - it introduces harmonics that sound similar to tube saturation, but it also amplifies quieter details at the same time.
@@sageaudio aah, right. I get that. Much thanks.
You are a Mastering ghost bro😯😯😯😯😯😯😯😯 loudness can't see you coming
Hahaha thanks for watching!
Why did you use your clipper after the limiter? Or this wasn't in any order perhaps...
How’s “look ahead” a mistake when the limiter presets have it on?
Look Ahead makes the compression from the limiter respond to attack and release snappier (not faster). snap means that the track becomes more forward and slightly louder in perceived loudness. presets can't account for all styles of music and track, so sometimes lookahead, true peak and algorithm/curves in presets can take away from the music. The vid is talking about a loud but not pumping master, so look ahead could introduce slight pumping. look ahead is great though, just not needed all the time.
Please excuse if this is a stupid question.
I'm new to mastering..... or attempting to master I should say.
My question is..... the examples you show..... are they intended to be one or the other per say...... or used in combination?
Thanks for watching - not a dumb question! These tips aren't part of a chain. That said, you can combine some of the techniques when mastering and incorporate them into what you're doing.
why said not to use True Peak limiting but didn't give an alternative. And streaming platforms want out put of -1 db but you had your limiter output at just -.1 . Why?
Can you mix and master a song for me?
does anyone know the name of those graphs so i can search them
Fletcher Munson curve
@@MOSMASTERING thanks man
I don't understand the "avoid using look-ahead limiting". Transients are typically what prevents making a track louder, so anything that helps the limiter react fast enough to transients is a good thing. No?
the transient is not it's loudest at the very start. So when the limiter is looking ahead it is turning down the signal before the transient essentially cutting off the sharpness in the front edge of the hit. The hit gets attenuated, but its shape is altered from sharp edged to round and that takes the punch out of the sound. So it rounds off the edges of every hit, and sounds un-natural. It is better to attenuate the signal after it is actually going above the threshold, because this is transparent: the ears cannot tell that the signal does not go high enough because the peak it just being cut off rather than the rise to the peak. I hope that makes sense.
Thanks for the detailed explanation. :)
Your limiter already reacts fast enough for transients, lookahead usually just helps avoiding unwanted distortion and smoothing dynamics out. But similarly to pre-ringing in linear EQs it messes up the transient definition in the process (although linear mode does it in a different way). For percussive transients a hard clipper typically sounds best as it literally shaves off the extra top, you'd be surprised how clean it can sound before introducing distortion.
I have been Mixing An artist song for a month now and I still can’t achieve the mixing I desire 😢😢😢😢
might be the microphone,, i came to the realization lol
If you guys need more tips for getting masters to -3.0 / -3.5 LUFS hit us up. 🥂🔥🙌🏻
on god tho what’s this song? 👀
Rollerblades by STVN
son g name?
There is so much bad information in this video. It's kind of shocking that it has been put out there. Many times it is touching on things that seem true but when you really know what's going on with these processors you will see that they aren't necessarily an issue.
The bad:
As another commenter mentioned, just switching on linear phase on an eq with no filters enabled will do nothing; it will null. The linear phase part of it is to do with the filters so, no filters? No Linear phase is being applied. The latency comment is just plain ridiculous. All that is going to potentially do is make the playback latent. That's it. However even that won't happen if you have PDC enabled. Really poor.
TP limiting, look ahead and oversampling. This also highlights a lack of knowledge about the plugin and how this would be used in the real world. If you want an actually informative video about the specifics about a limiter like Pro-L2 I would recommend you watch that video on the channel 'Panorama Mixing and Mastering'. I brought up some of the true peak stuff to him in a previous video and then he released this after investigating, which was great to see. 8x oversampling iirc IS the true peak limiting. That is how they do it. So just switching on oversampling isn't a guarantee that you don't "catch more of the transients". The more you increase this, the softer they will get because...you're doing more gain reduction. If you kept oversampling and TP off and just increased the gain reduction you would notice the same thing. This is blatantly obvious. The bigger issue with all this though is application and this applies to the look ahead too. You AREN'T comparing apples to apples. You can't just switch it on and off or increase the look ahead to compare. You NEED to adjust the amount of gain reduction too to make a fair comparison. You may NEED the look ahead because you WANT the limiter to react that way but you may not want the added gain reduction. In a real use case you would simply adjust the amount of gain reduction being applied. We would do this naturally altering any other parameter but for some reason folks don't when they alter these parameters on a limiter. They're reacting differently, just like how an attack control on a compressor would, and no one would say comparing two compressors; one with a slow attack and 1.5dB of compression vs one with a fast attack and 5dB of compression, that it's a fair comparison when judging what attack time is appropriate.
The same things apply to any compression you are using. There are very few absolutes in music production and "don't heavily compress the signal" isn't one of them. It's horses for courses. If you receive a wide open mix and are trying to push the loudness i'm afraid you may have to be slightly more heavy handed. This video is hamstringing people that are mostly beginners into thinking about things in terms of rules instead of listening and doing what is required based on a needs analysis. How much is too much? Exactly...it depends. 4dB of compression could be too much for one track and just right for another. Listen. Assess. Do what you need to do to get the result you want.
WRT unlinked limiting. This probably will result in louder masters in general but like everything, it comes with a cost too. It will have the benefit of potentially greater loudness and also the perception of width but also potentially at the cost of a solid centre image which went unmentioned. Use you ears and don't just automatically use a (fully) unlinked limiter. Using something like Ozone maximiser which allows the unlinking to be used on a continuous scale or even better, Pro-L2 which gives you that control over both the transient and release linking is a much better way to start experimenting with this instead of just blindly setting up a L/R like is shown here.
The Good
Relying solely on a limiter to for loudness isn't recommended, unless the mix allows you to do so. The most valuable piece of info missing from this video is that the loudness potential is in the mix, not the mastering. If you mix with loudness in mind, you will rely way less on forcing it with processing at the mastering stage.
Louder masters still sound louder than quiet ones when normalised, especially at lower volumes. The density etc of the louder one will mean it is perceived louder in a lot of cases. The spectral distribution changes from limiting and other processing at the mastering stage are also at play here and there is an upper and lower boundary, along with how well the loudness is executed to get this effect but it is reality. That is why no mastering engineer worth their salt is mastering to -14LUFSi. All major releases are at least -10LUFSi but usually closer to -8/-6LUFSi and it's for a reason.
Thank you so much for this comment!
i dont know, i prefer the unmastered every single time... it sounds more natural, and breathing. all the other techniques makes it actually squashed, less 3D and more harsh
Thanks for sharing your thoughts!
you have -0.7db on that limiter
Do you mean the Pro L2? The -0.7 at the bottom is referring to LUFS as it relates to -9LUFS. So it's -0.7dB below -9LUFS or roughly -9.7LUFS.
some of what you said isnt true tho, if mastering for streaming services you pretty much have to master at -1 true peak
I agree that it's a good idea to have a lower dBTP level. That said, most encoding won't cause gain changes as great as 1dB. Unless the track is super loud, the change should be closer to 0.5dB at most.
Ok now ppl, ready to have CRAZY phase issues in your songs? 😂
Its not loud enough
Thanks for watching and sharing your thoughts! The track gets up to roughly -9.5LUFS but I understand that some engineers like to push it louder
I didn't hear any difference with that pro q3 in lin phase mode?
The difference is incredibly small, so don't worry if you didn't hear a difference.
just wanted to tell you man, you have really been a big impact on my learning process. the way you explain things is just perfect. keep doing you❤️🩹
Why can't I hear any difference in most of the examples shown above? 🥲 (using Sony XM4 Bluetooth and Audio Technica ATH-M50X)