Audiobus: Use your music apps together.
What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.
Download on the App StoreAudiobus is the app that makes the rest of your setup better.
Comments
I did not. I have bad habits and tend to mix when I master. So I either have the mastered tracks, or stems with no EQing or effects.
It’s all good. I was thinking about offering some free mastering to people on this forum later this year anyway, so maybe later if you’re still interested.
Yeah I’d most certainly be interested in seeing what you could do with one of my older songs. Luckily I save the AUM project files so I can just strip off the master chain.
Actually wait, I do have individual tracks with effects, EQ, and mid-side balance. Would that work? Or do you need without the EQ and mid-side balance?
For mastering all I need is the single stereo mix down file without any master effects, I would not want someone to send me all the individual tracks. That would be mixing not mastering. 😉
that's great, Tarekith, would love to give that a try (planning my first release for Jan or Feb '22)
Mastering chain:
You have an tape echo / flanger in your mastering chain? 😳
Also, why two compressors? I mean, barricade is a limiter which means, it is an compressor too. Just for the typical magic death eye sound? Would that not be part of the mixing?
Just curious...
Reelbus is a tape sim more than it is an echo/flanger. I also like to have it on the master bus, it has a dedicated "master" mode that adds a very subtle amount of tape saturation.
It's also pretty common to run a compressor before the final limiter. The limiter is just there for peak control (to make sure no peaks clip) rather than for bus compression.
@richardyot said:
Ah, interesting - I don't have this App and just looked up the description in the App Store.
Thanks!
Here you are kind sir. https://drive.google.com/file/d/1Hh91Z0nSO7jxQx09rMvVp9W7SwrpVUdL/view?usp=drivesdk
Also there if anyone else wants to take a crack and play around with it.
Downloaded as well 😅
It seems to be limited straight up to about 0 dB which may already lead to some automatic changes by Spotify.
The Integrated LUFS per Pro-L2 is -11.9 which also is above the Spotify recommendation - with such files, Spotify even recommends a peek of -2dB instead of -1dB.
The WAV sounds way better to me than the version from Spotify.
I would just reduce gain to stay at -14 LUFS and have true peek limiting at -1 dB.
Also, did you dither the final master?
@tja you only need to dither for CD, you should never dither for streaming since the variable bit rate encoding will discard the dithering anyway. Dithering is for fixed bit rates (such as 16bit CD encoding).
As far as I know, you should always dither the master, when you go down to 16 Bit from 24 or 32 Bit.
This file is already 16 Bit, so it should have been dithered if the project was in 24 Bit or higher.
Some links that say the same as I had remembered:
https://www.waves.com/audio-dithering-what-you-need-to-know
https://blog.landr.com/what-is-dither-when-to-use
Don't dither, when there is no need to reduce the bit depth, but as compressed files mostly are created from a 16 Bit master, you should dither when going down to 16 Bit.
ReelBus is the most legit sounding tape simulator I’ve ever heard.
Barricade has an actual compressor between the clipper and limiter.
I’m actually compressing in more stages than you think, but very gently on each stage.
TBEQ
MagicDeathEyeStereo
ReelBus
Barricade
Ok, here we go:
https://tarekith.com/downloads/DawnStereo-Master.wav
A few caveats. This was quick and dirty since I'm doing it for free
Had this been a paying job I would have asked the artist if it was possible to make a few changes before I started. Namely lowering the bassline and snare a couple dB since those are both pretty loud in the mix. Also, as mentioned by tja, the file is clipping a little bit it sounds like, so lowering the master fader in the DAW by 3-6dB before exporting would help that too.
And yes, I applied 16bit dither too, though just triangular with no noise-shaping since this was already a 16bit file. If anyone can hear a difference in this song with or without dithering in ANY normal listening circumstance, I will give you my entire mastering studio.
I can.
sounds a whole lot better to my ears, very transparent, though still a little bit too sharp in the highs I think
One of the downsides of the snare being too loud, there's not a lot of wiggle room to address any other issues in the high end.
That sounds very good and looks very good in AudioShare!
I tried to find an easy App that mentions the peek limiting, but neither Twisted Wave nor Hokusai 2 seem to offer this. On PC I used Audacity, IIRC.
Could you mention the steps you did, @Tarekith?
I would normally just try a bit EQ, reduce the overall volume while limiting to a peek of -1 dB and try to aim for -14 LUFS.
Pretty simple set up for this one, EQ (some bands in mid-side mode too), some overall mid-side volume adjustments to bring a little more focus back to the center of the mix (a lot of sounds are panned pretty wide here), and then limiting. Here's screenshots for you:
As I mentioned earlier, I don't master to any streaming loudness "standards", it just doesn't work the way people think it's supposed to. Your song will almost always sound quieter than you want in comparison to other more popular songs.
Many thanks, @Tarekith
I understood that you don't follow the streaming standards...
I would like to see what Spotify would do to your version (-0.1 dB TP and probably high LUFS) , compared to the original, or to what I would do, as recommend by Spotify (-1dB TP, -14dB LUFS).
Of course, I cannot do this, but @DukeWonder could upload your version and then compare what get's streamed from Spotify.
Advice from Reid Stefan.”Spotify is a limiter. So, if you usually have a final limiter, remove it, and let Spotify do the limiting.”
Spotify wouldn't do anything to your version since you are aiming for their 'default' setting, it would playback the same as you submitted. My version is -9.1LUFS (just checked) so it would be turned down 4.9dB (which would also lower TP below -1 obviously).
Easy enough to do that on your end to hear the difference between the two versions.
Of course all this assumes that people have loudness normalization turned on Spotify in the first place, otherwise mine would play back at the current volume too.
Sort of.
Spotify will only apply limiting if your song is quieter than -14LUFS, the listener has a premium subscription, and they have the "loud" playback option selected in their preferences (which boosts the volume to -11LUFS not -14LUFS). Then they will apply a limiter with 5ms of attack and 100ms of release.
Otherwise Spotify will only turn up the track as loud as they can while still maintaining -1TB. So if your track is say -22LUFS with a peak of -3TP, they will only turn up the song 2dB resulting in a volume of -20LUFS.
All of these details are on Spotify's website as linked to in page 1 of this thread:
https://artists.spotify.com/help/article/loudness-normalization
Contrast that with a song that has a loudness measurement of say -9LUFS. For non-premium listers or those that stick with the standard loudness setting in Spotify, they will just turn it down 5dB (premium with Loud setting will be turned down 2dB). Simple gain change, no limiter needed, not affecting your audio at all in terms of quality.
Thank you. I think this gives me a lot to work with with future songs and beats.
Let me put it another way. Spotify is MORE likely to apply processing you have no control over (limiting) if you submit a song at -14LUFS/-1TP. Because a premium user can select the loud setting and then they boost the song to -11LUFS with a limiter in place using catch all settings.
For almost all other masters submited at loudness levels comparable to recent years, Spotify will just turn the song down as needed and not do any other processing on it.
Or if your song is quieter than -14LUFS, they will turn it up as much as they can until it either hits -1TP or -14LUFS, which will likely be pretty quiet compared to other songs (if you even care about that, some don't).
Why is it then that they recommend that you leave -2dB TP headroom instead of -1 if your track is louder? I’ve never understood that part.
I don't know enough about their transcoding process to say for sure, but there's potential I suppose for intersample overs as part of it as you get loud material near 0dBFS. -2TP seems REALLY conservative to me, but I guess they are just giving you a larger safety net.
FWIW most labels I work with ask for -1TP if anything, and I've never had one come back complaining about distortion or anything like that.
Awesome, thanks again for the insight.