In this pro mastering tutorial for simply a better sounding master, I will share some pro mastering techniques that I think probably most will find useful, or hopefully at least many.
The most important stage of mastering is actually not the final stage of mastering, it is what is called the post-mix stage. The objective of the post-mix stage is to ensure the mix has been optimally ensured for a great master. I will now try to describe what such a post-mix should be and what you then do with such a mix during mastering to turn it into a killer sounding final master.
In my own opinion the ideal post-mix to receive for mastering, is one that sounds great already, that has left me with awesome options for gain staging and broad stroke fx processing and that provides me as much access as possible to things inside the mix. I want balance in terms of a wide range of dimensions, I want of course a completely clipping free signal and I want great gain staging present inside of the mix so that I have a mix that sings, that vibrates with nice harmonics and that has an already awesome stereo image. I want the chorus to peak somewhere at around -11.5 LUFS, giving me somewhere around 5 dB of headroom before the master starts to break down. I want slow attack times on the compressors as an average in the mix and I want minimal amounts of sound noise and background noise. So essentially that means I demand great balance on the most important low noise sound sources and for the more noisy sound sources I prefer them to be too quiet rather than too loud. I want key dimensions like height and weight to be near final, so a mix with way too much or too little bass is an example of a mix I send back for mixing. But a mix that is a bit quiet in the high end, that's OK. I also don't allow broad stroke pass filtering on the mix, that's for me to decide. Finally, the post-mix must have an awesome sounding side component, that's one of the first things I check before I start to master, because if the side component is noisy it means the mix is not ready. I also validate the integrated LUFS, RMS and average frequency on both chorus, verse, mid, side and individual frequency bands, just to learn whether it is where I want it or not all in all. This gives me a good indication on what kind of mastering sound profiles I will be able to successfully fit with it.
During mastering, I usually move quite quickly to the gain staging process. The gain staging process is focused mostly on the combination of the output volume faders, the compressor attack and release times on the specific tracks (most I do in mastering is in zero phase) and the pan knobs. The reason is simple, when I later apply frequency band level gain staging to compress and expand various frequency bands as well as apply broad stroke compression and post-eq, I want to be able to do so as mildly as possible - light touches everywhere.
When I apply multi-band compression, broad stroke compression and post-eq, I do so iteratively on the post-mix group level in that order.
The multi-band compression goes most deeply into the content, depending on the quality of each band I then target them so that they get the right loudness and relaxation level according to a specific sound profile I am going for on that iteration. Then the broad stroke compression is applied, which levels the resulting dynamics and then the post-eq that finalizes the frequency balance on that iteration/sound profile.
The sound profiles are designed so that when combined they smoothly shape the sound of the mix from type A to type B, 4 sound profiles per print, 2 prints. The first two sound profiles are designed to prepare for the next two that are the main ones. 4 combined are a print and I usually for simplicity aim for 2 final prints. On the master bus I have a stereo widener that relaxes the width of the master, an EQ (pass filtered), a tape saturator, a brickwall peak limiter, a dynamic processor and a clipper, each designed to gain stage the next. (except the stereo widener) The clipper is finally setting the ceiling/desired true peak. These are mostly just kissing the stereo signal towards the final perceived loudness sweet spot. The dynamic processor ensures that the transients are optimal when they are fed into the clipper to boost the performance of the clipper. This is to positively compensate for the peak numbness naturally introduced by the ceiling and is controlled using the dynamic processor's attack time. The stereo widener I typically enable when the rest of the master chain is ready, because at that point the stereo widener has the most dramatic effect, I can run it a bit more mildly and I can more instantly notice when it is at an optimal setting. After that I might do some gradual gain stage fine tuning...
Translation is typically not an issue at all, but I reference check when I start as well as when I gain stage the master bus effects. In between all of that I'm more focused on other things...
Although this might seem as a strictly technical process, in practice it is not. As a mastering engineer you are constantly focused on the vibe and emotion of the song. Sometimes you get it right on print 3 or 4. Quite a lot of the creative process goes into unlocking the dynamics of the mix and getting a certain balance of weight, sound, vibe, height and loudness. You constantly have to keep track on what you need to do to the stereo image and what the processing does to it. It is for instance very essential that the stereo image is optimal before you apply the sound profiles. Before you apply those you have a pre-master and that pre-master must be extremely well balanced...
Keep in mind that both a great post-mix and a great final master is to some degree the result of great monitoring, because with great monitoring you get higher precision on the volume fader moves and with that comes higher modulation quality, better stereo image, better dynamics and a more relaxed sound. The combination is beautiful. I think it is very important to have a good monitoring path both during mixing and mastering that allows you to do a lot of positive compensations. It becomes a more airy/lush/vibrant/bigger sounding master at the end. It is extremely important that the post-mix group tracks have awesome separation.
The most important stage of mastering is actually not the final stage of mastering, it is what is called the post-mix stage. The objective of the post-mix stage is to ensure the mix has been optimally ensured for a great master. I will now try to describe what such a post-mix should be and what you then do with such a mix during mastering to turn it into a killer sounding final master.
In my own opinion the ideal post-mix to receive for mastering, is one that sounds great already, that has left me with awesome options for gain staging and broad stroke fx processing and that provides me as much access as possible to things inside the mix. I want balance in terms of a wide range of dimensions, I want of course a completely clipping free signal and I want great gain staging present inside of the mix so that I have a mix that sings, that vibrates with nice harmonics and that has an already awesome stereo image. I want the chorus to peak somewhere at around -11.5 LUFS, giving me somewhere around 5 dB of headroom before the master starts to break down. I want slow attack times on the compressors as an average in the mix and I want minimal amounts of sound noise and background noise. So essentially that means I demand great balance on the most important low noise sound sources and for the more noisy sound sources I prefer them to be too quiet rather than too loud. I want key dimensions like height and weight to be near final, so a mix with way too much or too little bass is an example of a mix I send back for mixing. But a mix that is a bit quiet in the high end, that's OK. I also don't allow broad stroke pass filtering on the mix, that's for me to decide. Finally, the post-mix must have an awesome sounding side component, that's one of the first things I check before I start to master, because if the side component is noisy it means the mix is not ready. I also validate the integrated LUFS, RMS and average frequency on both chorus, verse, mid, side and individual frequency bands, just to learn whether it is where I want it or not all in all. This gives me a good indication on what kind of mastering sound profiles I will be able to successfully fit with it.
During mastering, I usually move quite quickly to the gain staging process. The gain staging process is focused mostly on the combination of the output volume faders, the compressor attack and release times on the specific tracks (most I do in mastering is in zero phase) and the pan knobs. The reason is simple, when I later apply frequency band level gain staging to compress and expand various frequency bands as well as apply broad stroke compression and post-eq, I want to be able to do so as mildly as possible - light touches everywhere.
When I apply multi-band compression, broad stroke compression and post-eq, I do so iteratively on the post-mix group level in that order.
The multi-band compression goes most deeply into the content, depending on the quality of each band I then target them so that they get the right loudness and relaxation level according to a specific sound profile I am going for on that iteration. Then the broad stroke compression is applied, which levels the resulting dynamics and then the post-eq that finalizes the frequency balance on that iteration/sound profile.
The sound profiles are designed so that when combined they smoothly shape the sound of the mix from type A to type B, 4 sound profiles per print, 2 prints. The first two sound profiles are designed to prepare for the next two that are the main ones. 4 combined are a print and I usually for simplicity aim for 2 final prints. On the master bus I have a stereo widener that relaxes the width of the master, an EQ (pass filtered), a tape saturator, a brickwall peak limiter, a dynamic processor and a clipper, each designed to gain stage the next. (except the stereo widener) The clipper is finally setting the ceiling/desired true peak. These are mostly just kissing the stereo signal towards the final perceived loudness sweet spot. The dynamic processor ensures that the transients are optimal when they are fed into the clipper to boost the performance of the clipper. This is to positively compensate for the peak numbness naturally introduced by the ceiling and is controlled using the dynamic processor's attack time. The stereo widener I typically enable when the rest of the master chain is ready, because at that point the stereo widener has the most dramatic effect, I can run it a bit more mildly and I can more instantly notice when it is at an optimal setting. After that I might do some gradual gain stage fine tuning...
Translation is typically not an issue at all, but I reference check when I start as well as when I gain stage the master bus effects. In between all of that I'm more focused on other things...
Although this might seem as a strictly technical process, in practice it is not. As a mastering engineer you are constantly focused on the vibe and emotion of the song. Sometimes you get it right on print 3 or 4. Quite a lot of the creative process goes into unlocking the dynamics of the mix and getting a certain balance of weight, sound, vibe, height and loudness. You constantly have to keep track on what you need to do to the stereo image and what the processing does to it. It is for instance very essential that the stereo image is optimal before you apply the sound profiles. Before you apply those you have a pre-master and that pre-master must be extremely well balanced...
Keep in mind that both a great post-mix and a great final master is to some degree the result of great monitoring, because with great monitoring you get higher precision on the volume fader moves and with that comes higher modulation quality, better stereo image, better dynamics and a more relaxed sound. The combination is beautiful. I think it is very important to have a good monitoring path both during mixing and mastering that allows you to do a lot of positive compensations. It becomes a more airy/lush/vibrant/bigger sounding master at the end. It is extremely important that the post-mix group tracks have awesome separation.
Last edited:
