What I did:
- Imported a rendered final mix into a new project
- Added the plugins I was using to the mix track
- Ordered them with the EQs first, then compressor, then limiter
Your mastering process in Tracktion?
-
- KVRist
- 479 posts since 9 Jan, 2023
Just started implementing this into my song-creation process (yes, I'm a total noob when it comes to engineering). I will say, it seemed immediately clear how impactful it can be on a final mix. That being said, I want to make sure I'm taking advantage of all the elements of the process I can whilst not making things worse.
What I did:
What I did:
-
- KVRAF
- 1599 posts since 9 Jan, 2018
Hey Tony (by the way, loving your work),
My process is definitely a little different than most people because of the way I set up my projects. I don't do much with my master bus, but instead route all my "recordable" tracks to a single track, and then bounce that track to a second track.
Why?
* It keeps plugins off my master bus, which Waveform historically limited you to 4; I understand the number is higher, but by not directly using it, I'm all set with whatever I want to do.
* I don't seem to suffer all the weirdness that rendering gives so many users, and I never have to worry about automation on the master bus. I can automate fades on the single, regular track and this gets bounced easily to my bounce track.
* I can use all sorts of clicks and metronomes on non-routed tracks without every worrying about those being recorded.
* I can adjust the level of the master bus without worrying how this affects a final recording.
In that respect, by setting up my projects this way, I'm already starting the mastering process very early in the project so there's less work later on.
Anyway, all that said, my process is typically:
1. Finalize the mix, using Dsoniq to ensure my headphones are hearing things more correctly.
2. Listen to it a bunch more times and realize how much I need to improve things.
3. Listen to it in mono, using Waveform's Mono plugin to see if gets muddy. If it's fine, I remove it. If not, I try to figure out where I need to remix. Obviously, problems will be on stereo tracks.
4. Run the mix through SPAN on the overall recording track to see if I have any mid-side muddiness. This is a great utility for this: if you set it up right, anything crappy highlights in pink. If I see pink, I can move SPAN from the overall recording track to individual tracks and hunt down the sources. Popping an EQ on any muddy tracks will eliminate the pink quickly. Put it back on the overall recording track, and make sure I've gotten rid of any recurring pink sources. I remove it when it's done.
5. Put Ozone on the overall recording track, and let it do its thing. Generally, it does a great job 90% of the time. I almost always hear dramatic improvement, but the next step really helps.
6. Put Tonal Balance Control on the overall recording track. This tool is amazing. If you're not familiar with it, it plots out a sort-of heat map for your project across the common frequency bands based on hundreds of songs in your preferred genre (there's a drop down list of genres). If you're dropping low in a band, you either need to re-EQ it or you're over-compressing. If you're too hot in a band, you either need compression or need to re-EQ it. Point is, once you center your project's audio in these bands, you've got a pretty good mix for everything from ear buds, car stereos, home audio, CDs, etc., based on how it compares to hundreds of other recordings. When I need to re-EQ a band, I can do it right in Ozone and watch, in real time, as the heat map readjusts to a better mix. So when Ozone doesn't quite hit the mark for me, Tonal Balance Control tells me where I need to readjust Ozone's EQ and the result is obvious.
7. Okay, assuming Tonal Balance Control is happy, I remove it as Ozone is now doing all the real work.
8. I then bounce the overall track to the bounce track, open up my Recorded Items folder for that project, and look for the bounced audio track. I'll drag that out, and listen to it on a few different systems over a few days to make sure Tonal Balance Control didn't lie to me (it really doesn't, but every so often something that sounds great in the car sounds tinny on a tablet or through ear buds and needs a minor adjustment--back to step 6 and retool the Ozone EQ). Sometimes what sounds great on a Monday sounds bad on a Tuesday, so I like to give it few days of active listening.
9. Sometimes there's little or nothing left to change--and I've got a digital copy ready to publish.
A lot of your steps are done on individual tracks, too: I'll compress and EQ individual tracks and color them up, but my goal is always to listen in context. An individual track can sounds great, but when put against the rest of the material, mmm, maybe not so good. But once the whole project is mixed, I really trust Ozone to do the heavy lifting on mastering. It generally does great work, and when it messes up, it's not hard to pinpoint the problem and clean it up in other ways.
But you'll note I don't do much with stems. Once stemmed, I'm kind of stuck with the result. And it's a lot easier to go back and add or remove something from an individual track if it sounds bad in the final, final mix when tested on different systems.
My use of SPAN, Tonal Balance Control, and dSoniq just look for hidden issues or problems.
Once in a rare while, I'll put delay or reverb or saturation on an overall project just to glue it together for a really big sound, but I don't do that too often as it gets to be a cliché for me, personally. But I certainly have done it and absolutely will do it again if I need to glue a performance to make everything sit just right.
My process is definitely a little different than most people because of the way I set up my projects. I don't do much with my master bus, but instead route all my "recordable" tracks to a single track, and then bounce that track to a second track.
Why?
* It keeps plugins off my master bus, which Waveform historically limited you to 4; I understand the number is higher, but by not directly using it, I'm all set with whatever I want to do.
* I don't seem to suffer all the weirdness that rendering gives so many users, and I never have to worry about automation on the master bus. I can automate fades on the single, regular track and this gets bounced easily to my bounce track.
* I can use all sorts of clicks and metronomes on non-routed tracks without every worrying about those being recorded.
* I can adjust the level of the master bus without worrying how this affects a final recording.
In that respect, by setting up my projects this way, I'm already starting the mastering process very early in the project so there's less work later on.
Anyway, all that said, my process is typically:
1. Finalize the mix, using Dsoniq to ensure my headphones are hearing things more correctly.
2. Listen to it a bunch more times and realize how much I need to improve things.
3. Listen to it in mono, using Waveform's Mono plugin to see if gets muddy. If it's fine, I remove it. If not, I try to figure out where I need to remix. Obviously, problems will be on stereo tracks.
4. Run the mix through SPAN on the overall recording track to see if I have any mid-side muddiness. This is a great utility for this: if you set it up right, anything crappy highlights in pink. If I see pink, I can move SPAN from the overall recording track to individual tracks and hunt down the sources. Popping an EQ on any muddy tracks will eliminate the pink quickly. Put it back on the overall recording track, and make sure I've gotten rid of any recurring pink sources. I remove it when it's done.
5. Put Ozone on the overall recording track, and let it do its thing. Generally, it does a great job 90% of the time. I almost always hear dramatic improvement, but the next step really helps.
6. Put Tonal Balance Control on the overall recording track. This tool is amazing. If you're not familiar with it, it plots out a sort-of heat map for your project across the common frequency bands based on hundreds of songs in your preferred genre (there's a drop down list of genres). If you're dropping low in a band, you either need to re-EQ it or you're over-compressing. If you're too hot in a band, you either need compression or need to re-EQ it. Point is, once you center your project's audio in these bands, you've got a pretty good mix for everything from ear buds, car stereos, home audio, CDs, etc., based on how it compares to hundreds of other recordings. When I need to re-EQ a band, I can do it right in Ozone and watch, in real time, as the heat map readjusts to a better mix. So when Ozone doesn't quite hit the mark for me, Tonal Balance Control tells me where I need to readjust Ozone's EQ and the result is obvious.
7. Okay, assuming Tonal Balance Control is happy, I remove it as Ozone is now doing all the real work.
8. I then bounce the overall track to the bounce track, open up my Recorded Items folder for that project, and look for the bounced audio track. I'll drag that out, and listen to it on a few different systems over a few days to make sure Tonal Balance Control didn't lie to me (it really doesn't, but every so often something that sounds great in the car sounds tinny on a tablet or through ear buds and needs a minor adjustment--back to step 6 and retool the Ozone EQ). Sometimes what sounds great on a Monday sounds bad on a Tuesday, so I like to give it few days of active listening.
9. Sometimes there's little or nothing left to change--and I've got a digital copy ready to publish.
A lot of your steps are done on individual tracks, too: I'll compress and EQ individual tracks and color them up, but my goal is always to listen in context. An individual track can sounds great, but when put against the rest of the material, mmm, maybe not so good. But once the whole project is mixed, I really trust Ozone to do the heavy lifting on mastering. It generally does great work, and when it messes up, it's not hard to pinpoint the problem and clean it up in other ways.
But you'll note I don't do much with stems. Once stemmed, I'm kind of stuck with the result. And it's a lot easier to go back and add or remove something from an individual track if it sounds bad in the final, final mix when tested on different systems.
My use of SPAN, Tonal Balance Control, and dSoniq just look for hidden issues or problems.
Once in a rare while, I'll put delay or reverb or saturation on an overall project just to glue it together for a really big sound, but I don't do that too often as it gets to be a cliché for me, personally. But I certainly have done it and absolutely will do it again if I need to glue a performance to make everything sit just right.
Spotify, Apple Music, YouTube, and even Deezer, whatever the hell Deezer is.
More fun at Twitter @watchfulactual
More fun at Twitter @watchfulactual
-
- KVRAF
- 2456 posts since 9 Oct, 2008 from UK
The latest issue of Computer Music magazine (no 325, October 2023) has its main article on mastering. Mastering is done on a finished mix, so time is needed to get that as good as possible before considering mastering I recall watching some interesting YouTube videos on the topic.
[W10-64, T5/6/7/W8/9/10/11/12/13, 32(to W8)&64 all, Spike],[W7-32, T5/6/7/W8, Gina16] everything underused.
-
- KVRist
- Topic Starter
- 479 posts since 9 Jan, 2023
Wow! Thank you so much for all of that info. Think it will take me a while to digest it all, but I'm reading some very interesting stuff here. I started using SPAN, and I've found it to be quite useful. However, I'm not sure I set it up in the way you referred to. The areas you mentioned (mid-side) are exactly the main trouble areas I was noticing. I was using MSED and QRange together to try and ferret out and correct that area where the sides and mid meet. But SPAN didn't show any varying colors in its representation of the analyzer.Watchful wrote: Wed Aug 09, 2023 7:18 pm Hey Tony (by the way, loving your work)...
... Run the mix through SPAN on the overall recording track to see if I have any mid-side muddiness. This is a great utility for this: if you set it up right, anything crappy highlights in pink. If I see pink, I can move SPAN from the overall recording track to individual tracks and hunt down the sources. Popping an EQ on any muddy tracks will eliminate the pink quickly. Put it back on the overall recording track, and make sure I've gotten rid of any recurring pink sources. I remove it when it's done.
5. Put Ozone on the overall recording track, and let it do its thing. Generally, it does a great job 90% of the time. I almost always hear dramatic improvement, but the next step really helps.
6. Put Tonal Balance Control on the overall recording track. This tool is amazing. If you're not familiar with it, it plots out a sort-of heat map for your project across the common frequency bands based on hundreds of songs in your preferred genre (there's a drop down list of genres). If you're dropping low in a band, you either need to re-EQ it or you're over-compressing. If you're too hot in a band, you either need compression or need to re-EQ it. Point is, once you center your project's audio in these bands, you've got a pretty good mix for everything from ear buds, car stereos, home audio, CDs, etc., based on how it compares to hundreds of other recordings. When I need to re-EQ a band, I can do it right in Ozone and watch, in real time, as the heat map readjusts to a better mix. So when Ozone doesn't quite hit the mark for me, Tonal Balance Control tells me where I need to readjust Ozone's EQ and the result is obvious.
7. Okay, assuming Tonal Balance Control is happy, I remove it as Ozone is now doing all the real work. ...
...My use of SPAN, Tonal Balance Control, and dSoniq just look for hidden issues or problems.
...
Ozone, TBC, and dSoniq I'm not familiar with, but I'll take everything and anything that helps me to find and correct trouble areas. I don't mind doing the work by ear, but the hard reality is, my ears become brutally fatigued after a while, and I usually end up doing more harm than good at that point.
But again, thank you so much! I'll try and find out more about those plugins.
-
- KVRAF
- 1599 posts since 9 Jan, 2018
Here's how to use SPAN to help find any mid-side trouble areas.
1. Put a copy of SPAN on a stereo track (obviously, you wouldn't be having many mid-side issues on a mono track!).
2. Look for the down arrow menu next to Routing on the menu bar.
3. Click the down arrow and select Mid-Side Stereo
4. Next to the down arrow, click MID (this button appears after step 2), if not already selected
5. Click Underlay (below the routing button) and select SIDE.
When you play your track, you'll see two colors splashing across the display. Light green is your side information, and dark green is your mid information.
If you see pink, that's a conflict between the two. Basically, you're seeing the muddiness you're hearing.
A splash of pink here and there is NOT a problem. I've put a bunch of professional recordings through SPAN, and you'll see a lot of pink in many of them.
But if you're seeing big batches of pink, or sustained pink in certain frequencies, you've got some mid-side issues. Pop an EQ onto the track and start trimming back until the pink goes away for the most part. Some EQs are designed to work especially with sides, so feel free to use one of those, but any EQ can do it, really.
Remember that you can have no side issues on individual tracks, but see them on buses or on groups or on masters as you combine tracks together.
That's why I put SPAN on the final output--if I don't see any mid-side issues there, nothing to worry about. If I do, it's time to go down to the submix and group levels to see where the issues are adding up.
If you don't see any issues on the track level, I can just EQ the master output to resolve them.
There's a reason for this: the best analogy I heard is that a bunch of tracks are like pints of beer.
You can fill a pint glass 1/4 of the way, and think you'll never spill. But when you dump five glasses like this into one empty pint glass, it overflows. Likewise, if you trim your audio on individual tracks to perfection, but then put 'em all through one output, it overflows. Compression, EQ, and mid-side issues can pop up seemingly out of nowhere on tracks that are otherwise perfect to you. SPAN is a great tool to pinpoint where the mid-side issues are starting.
Hope this helps!
1. Put a copy of SPAN on a stereo track (obviously, you wouldn't be having many mid-side issues on a mono track!).
2. Look for the down arrow menu next to Routing on the menu bar.
3. Click the down arrow and select Mid-Side Stereo
4. Next to the down arrow, click MID (this button appears after step 2), if not already selected
5. Click Underlay (below the routing button) and select SIDE.
When you play your track, you'll see two colors splashing across the display. Light green is your side information, and dark green is your mid information.
If you see pink, that's a conflict between the two. Basically, you're seeing the muddiness you're hearing.
A splash of pink here and there is NOT a problem. I've put a bunch of professional recordings through SPAN, and you'll see a lot of pink in many of them.
But if you're seeing big batches of pink, or sustained pink in certain frequencies, you've got some mid-side issues. Pop an EQ onto the track and start trimming back until the pink goes away for the most part. Some EQs are designed to work especially with sides, so feel free to use one of those, but any EQ can do it, really.
Remember that you can have no side issues on individual tracks, but see them on buses or on groups or on masters as you combine tracks together.
That's why I put SPAN on the final output--if I don't see any mid-side issues there, nothing to worry about. If I do, it's time to go down to the submix and group levels to see where the issues are adding up.
If you don't see any issues on the track level, I can just EQ the master output to resolve them.
There's a reason for this: the best analogy I heard is that a bunch of tracks are like pints of beer.
You can fill a pint glass 1/4 of the way, and think you'll never spill. But when you dump five glasses like this into one empty pint glass, it overflows. Likewise, if you trim your audio on individual tracks to perfection, but then put 'em all through one output, it overflows. Compression, EQ, and mid-side issues can pop up seemingly out of nowhere on tracks that are otherwise perfect to you. SPAN is a great tool to pinpoint where the mid-side issues are starting.
Hope this helps!
Spotify, Apple Music, YouTube, and even Deezer, whatever the hell Deezer is.
More fun at Twitter @watchfulactual
More fun at Twitter @watchfulactual
-
- KVRist
- Topic Starter
- 479 posts since 9 Jan, 2023
That helps a lot! I know a lot of folks always offer the easy answer of "use your ears," but ears fatigue after hours and hours with the same song, and it's very useful to utilize other tools that can more finely help you needle out where the problem areas are, especially when it comes to sound design, as once you remove or add one thing, it changes the balance of another. But if you can learn the vicinity of your problem using a technique like this, it sure helps.Watchful wrote: Thu Aug 10, 2023 6:10 pm Here's how to use SPAN to help find any mid-side trouble areas.
1. Put a copy of SPAN on a stereo track (obviously, you wouldn't be having many mid-side issues on a mono track!).
2. Look for the down arrow menu next to Routing on the menu bar.
3. Click the down arrow and select Mid-Side Stereo
4. Next to the down arrow, click MID (this button appears after step 2), if not already selected
5. Click Underlay (below the routing button) and select SIDE.
When you play your track, you'll see two colors splashing across the display. Light green is your side information, and dark green is your mid information.
If you see pink, that's a conflict between the two. Basically, you're seeing the muddiness you're hearing.
A splash of pink here and there is NOT a problem. I've put a bunch of professional recordings through SPAN, and you'll see a lot of pink in many of them.
But if you're seeing big batches of pink, or sustained pink in certain frequencies, you've got some mid-side issues. Pop an EQ onto the track and start trimming back until the pink goes away for the most part. Some EQs are designed to work especially with sides, so feel free to use one of those, but any EQ can do it, really.
Remember that you can have no side issues on individual tracks, but see them on buses or on groups or on masters as you combine tracks together.
That's why I put SPAN on the final output--if I don't see any mid-side issues there, nothing to worry about. If I do, it's time to go down to the submix and group levels to see where the issues are adding up.
If you don't see any issues on the track level, I can just EQ the master output to resolve them.
There's a reason for this: the best analogy I heard is that a bunch of tracks are like pints of beer.
You can fill a pint glass 1/4 of the way, and think you'll never spill. But when you dump five glasses like this into one empty pint glass, it overflows. Likewise, if you trim your audio on individual tracks to perfection, but then put 'em all through one output, it overflows. Compression, EQ, and mid-side issues can pop up seemingly out of nowhere on tracks that are otherwise perfect to you. SPAN is a great tool to pinpoint where the mid-side issues are starting.
Hope this helps!
Man, thank you very much!
-
- KVRAF
- 1599 posts since 9 Jan, 2018
You're always welcome, but I'm a bit concerned you're working too hard. Be sure to take frequent breaks. Seriously, don't mix for more than 20-30 minutes at a stretch without doing something else. Recording, arranging, cleaning up audio...that's one thing, but mixing + fatigue is a bad idea. Mix for a while, but then switch projects or do something else on the project to give your ears a rest, then go back to mixing after you've had a chance to reset.
Spotify, Apple Music, YouTube, and even Deezer, whatever the hell Deezer is.
More fun at Twitter @watchfulactual
More fun at Twitter @watchfulactual
-
- KVRist
- Topic Starter
- 479 posts since 9 Jan, 2023
Indeed. Good advice, and it's definitely something I'm guilty of.Watchful wrote: Fri Aug 11, 2023 3:36 am You're always welcome, but I'm a bit concerned you're working too hard. Be sure to take frequent breaks. Seriously, don't mix for more than 20-30 minutes at a stretch without doing something else. Recording, arranging, cleaning up audio...that's one thing, but mixing + fatigue is a bad idea. Mix for a while, but then switch projects or do something else on the project to give your ears a rest, then go back to mixing after you've had a chance to reset.