Logic Mastering Workflow

Hello, all -

 

I realize this is a perennial topic, but since some of the discussions are about older versions of Logic, I thought to start a thread on it here.

 

So: What's the optimal workflow for mastering within Logic?

I'm a composer/producer/artist, and still quite "green" about mastering, so please forgive any naiveté on display. Hopefully this is useful starting point for discussion, where others with more experience and wisdom can critique and improve, if they so wish. I've included a few obvious question points in the checklist.

 

LOGIC MASTERING WORKFLOW (Version 0)

 

• MIX

Mix and bounce down each track within Logic.

Settings:

- 24-bit (if mastering within Logic, since Logic apparently still cannot import 32-bit float files!)

- 32-bit float if sending to a mastering engineer, or mastering in a DAW, which can accept 32-bit float files)

 

If 24-bit, use Dither? (POWR-1). This issue still seems controversial, especially since this means there will be 2 dither phases (see below).

From what I can tell, there is no consensus. What does Logic do to the bounce if no dither is applied at this stage? Does it just truncate the internal 32-bit mix to 24-bit, or does it do some more subtle process, whether an automatic dither or something else?

 

Edit: It seems like Logic does not auto-dither when bouncing to 24-bit. See here, for example:

viewtopic.php?f=1&t=63475

 

• IMPORT MIXES INTO A MASTERING PROJECT IN LOGIC

- Start with either a blank project, or a MASTERING template of your own creation.

- Each mix into its own track

- Start the mixes at measure 1? Or space them out so that the entire project plays in album sequence (i.e. each track start time in Logic is after the prior track's end time, with appropriate silence added to hear the album in sequence)?

- Add appropriate mastering plugins to each track.

- Route all tracks into a single Aux (which I call "MASTER"), which routes to to the Stereo Output

- add any metering and analysis plugins to this MASTER aux (no processing plugins here; all processing plugins go on the individual tracks)

(I like Izotope Insight and NuGen's Visualizer and MasterCheck)

- add any REFERENCE plug-ins (I like Mastering the Mix's "Reference") to this MASTER aux also

 

• MASTER

- This is where the magic happens (or where those of us, like myself, who are not mastering engineers, get ourselves in trouble).

- Add/subtract/adjust the mastering plugins, track levels, trims, fades, etc. for each track

- If exporting to 16-bit, best to do the bit-rate reduction and dither in the last plugin (e.g. mastering limiter like ProL2 or Xenon), rather than let Logic do the bit-rate reduction at the export?

- Where to do any volume automation (e.g. fades and rides)? Region or track automation? A gain plugin, automated? Applying fades to the region itself?

- Use meters and analyzers as needed to check target loudness, peak / true-peak, stereo image, etc.

- Use ears to listen to entire album flow, ideally at various volume levels and on different playback systems (in different rooms), including speakers and headphones

(I actually have my studio system wired to the hi-fi in the living room distributed whole-house audio, so I can listen to works in progress in different rooms on different speakers, including in-wall speakers in different locations)

 

• EXPORT

- Best way to export the now mastered tracks?

 

1. Export Regions as Audio Files

- Format: WAV or AIF

- Bit Depth (24-bit if for a hi-res format; 16-bit for CD or other distribution that only accepts 16-bit)

- Problem: No dither options here! Yet we should dither if exporting to 16-bit.

- Bypass Effect Plugins: OFF (else won't apply the mastering plugins)

- Include Audio Tail: OFF (there should be no audio tail at this point, unless you are applying reverb, delay, etc. at the mastering stage!)

- Include Volume/Pan Automation: ON or OFF? If automating volume here (e.g. fades), then this should be on, but I have had very erratic results with this setting on (e.g. the "Files Disappearing" bug)

- Normalize: OFF

 

2. Bounce

- This seems to be the only way to proceed, if you wish to dither...?

- Unfortunately, unlike the Export Regions command, this seems to require a separate Bounce command for each track, after first setting the bounce range appropriately each time

- Format: WAV or AIF

- Bit Depth (24-bit if for a hi-res format; 16-bit for CD or other distribution that only accepts 16-bit)

- Bypass Effect Plugins: OFF (else won't apply the mastering plugins)

- Include Audio Tail: OFF (there should be no audio tail at this point, unless you are applying reverb, delay, etc. at the mastering stage!)

- Include Volume/Pan Automation: ON or OFF? If automating volume here (e.g. fades), then this should be on, but again, I have had very erratic results with this setting on

- Normalize: OFF

 

• FINAL CHECK

Listen to the exported tracks in sequence. If in iTunes, consider listening with both SoundCheck on (which loudness-adjusts the tracks), and off, since you never know which way a given listener has iTunes set up. Perhaps even burn a CD to listen on a CD player. Try exporting as MP3s (high and low resolution) also. You can also use plugins (e.g. MasterCheck) to simulate some of the playback algorithms.

 

 

• ETC

Well, that's all I can think of at the moment. Hopefully others will feel inspired to contribute and improve upon this list.

 

Regards.

Edited August 30, 2019 by Marcus Aurelius

First of all, I wouldn't dither on the mix because Logic goes back to 24-bit fixed on the output to the interface.

So you would add an unnecessary dither process.

 

Second, I wouldn't use an Aux for the master in the mastering session.

I have each song on its own track, that way it's easy to compare them by putting up a cycle.

I use a mastering EQ and maybe some hardware compression and limiting.

I then dither with POW#3 (my favorite) down to 16 bit on the bounce, usually wave.

 

I use the Apple Roundtrip plugin to check for intersample peaks on AAC conversion, but rarely use it since I don't master louder than -11 LUFS.

Thanks to Marcus for starting this thread with great questions that are as useful as the answers to come, and to triplets for very useful responses already. I’ll be following this thread closely.

Since I’m self-mastering (I know, the doctor that diagnoses himself has an idiot for a patient) for sync and sound track I’d be interested in how mastering workflow would change for that scenario. Two submission formats are required, MP3 (256) files to supervisors for audition purposes and 24/48 files for final files and stems. All thoughts and suggestions appreciated.

Two submission formats are required, MP3 (256) files to supervisors for audition purposes and 24/48 files for final files and stems.

 

Definitely check for intersample peaks if you're bouncing to mp3. In those cases I'd lower the Output to -0.3 on the last limiter you have.

Also ask for LUFS levels on your mixes before you submit.

Thanks, triplets, useful info. I just recently saw (probably here) a method for checking intersample peaks. I usually bounce to 13-18 LUFS for MP3’s, nothing red-lining in the mix. It seems the clients are OK with adjusting MP3 playback volume on their end even though those bounces sound a little “quiet” on my end. They’re probably applying their personal “6-Second Rule” so they can move on quickly, but no complaints so far. Recording in the 70’s it was the “8-Second Rule”, I guess we’re running out of time... I’ll definitely ask for LUFS guidelines for the final full res files.

I just recently saw (probably here) a method for checking intersample peaks.

 

https://www.apple.com/itunes/mastered-for-itunes/

 

I use the Roundrip AAC plugin from Apple.

Thanks for the links, I was about to ask where I could find RoundTrip. This is a timely topic given Apple’s recent announcements regarding audio. I’ll be downloading and learning the linked plugins ASAP.

Thanks for the links, I was about to ask where I could find RoundTrip. This is a timely topic given Apple’s recent announcements regarding audio. I’ll be downloading and learning the linked plugins ASAP.

 

What are the announcements about audio? I hadn't "heard" anything...

Apple is implementing a hi-res audio process into iTunes, I’m still trying to determine if it’s for iTunes (soon to disappear?) streaming only or for all audio usage. Several tech news articles mention it, I haven’t read the details yet.

On the apparently contentious question of whether to dither on the 24-bit mix (=pre-mastering bounce) (which effectively will lead to double-dithering, since the mastering process will dither again; I thought double-dithering was to be avoided), I've been down the rabbit hole for a while now reading threads and articles and info on mastering sites and am more confused than ever.

 

Here's something from LANDR's info:

 

"If you’re sending your files for mastering, leave dithering out if you can export 32-bit float files. In this case, the mastering process will take care of dither for you. When you export anything other than 32-bit float, you have to dither. That includes when you bounce files that are the same bit-depth as the ones you recorded."

 

Also an old long thread on GearSlutz on whether to dither to 24-bit before mastering. Bob Olhsson there gives a resounding "Yes!".

https://www.gearslutz.com/board/mastering-forum/134214-dithering-24-bits-mastering.html

 

And another:

https://www.gearslutz.com/board/so-much-gear-so-little-time/530627-why-not-dither-mix-down.html

 

Waves adds this:

"You don’t need to dither when going from 32-bit floating point to 24-bit (because 32-bit floating point doesn’t have a higher bit depth), but you do from 32-bit fixed point to any lower bit depth."

(Since Logic is internally 32-bit floating, then this would imply not to dither Logic's mixdown to 24-bit...)

And: "If you’re preparing 24- or 32-bit data for a mastering engineer, don’t dither. Let your mastering engineer, who’s an expert, handle it."

https://www.waves.com/audio-dithering-what-you-need-to-know

 

Bob Katz also seems to recommend dither at mixdown to 24-bit:

https://www.digido.com/ufaqs/dithering-mixing-using-plugins/

 

It's just one of the many issues involved in this Logic Mastering workflow, but apparently a "hot topic".

I guess I'll have to experiment...

If you use Logic, which like Waves says, no need to dither.

If you use another DAW that does everything at 32-bit fixed and not float, then find out with the mastering house what they want.

Guidelines for submission for film or TV seem to be 24/48 files, no dithering. Fixed or floating point factors don’t seem to be mentioned. If the files go to a library first with no direct contact with the show boss, knowing the dither-answer in advance would be helpful. I always record 24/48 (48 being the video standard) but I was surprised that I could hear a difference between 44.1 and 48 (without dithering), with 48 audibly better on several playback systems.