Stem Mastering for Enhanced Audio Quality

Stem mastering is a relatively recent development in the world of audio engineering. It’s a process that can have a pretty dramatic effect on the final result of a musical production, but what is it and how does it differ from traditional stereo mastering? That's what we are going to dive into.

Traditional Stereo Mastering

In a traditional mastering scenario, you are given stereo files. No matter how many tracks make up the mix of a song, it is mixed down to a single stereo track. Within the confines of these stereo files, the mastering engineer works to correct any issues and increase the fidelity of the audio being processed. Typically, the goal isn’t to rebalance the mix, just make corrections and enhance an already balanced mix.

There are quite a few limitations with stereo mastering since you are dealing with a mix that has multiple elements spread across the entire frequency space. These limitations aren’t usually problematic assuming a well-balanced mix, but let’s take a look at a scenario.

You have a mix where the vocals are too bright but the instrumental portion that makes up the music is dull. You want to brighten up the instrumental but doing that would make the vocals which were too bright to begin with, insanely bright. Sure, you can try and resort to some creative processing to compensate for this, but it is less than ideal. Neither the instrumental nor the vocals are going to be where they need to be and that's the tradeoff. This is just one of many potential scenarios that happen with increasing regularity.

In the previous case, you go back to the mixer/artist/producer and ask for a revision explaining the situation. Hopefully, on the next run, you get what you need to finish the master.

Stem Mastering

Stem mastering aims to address some of the issues that can’t be addressed by traditional stereo mastering. Instead of just having one stereo mix of the entire song, you have multiple stereo files that make up the core elements of the mix. You then take these individual stereo files and create a single stereo master from them.

These audio files provided could be as simple as an instrumental and a vocal stem to the individual subgroups of instruments that make up the mix. There are no rules here on which stems to provide and every engineer is going to have their preference, but let’s take an example of what you might provide for a modern a rock song:

• Drums
• Bass
• Guitars
• Leads
• Synths / Pianos
• Vocals
• Sound Effects (Car crashes, sub drops, reverse cymbals, etc) 
• Effects (Delays, Reverbs, etc)

This is just 8 stereo tracks. So the entire mix, which may have been 100's of tracks is now broken down into just 8 tracks. Or maybe even fewer tracks, here is an example with 5:

• Drums
• Bass
• Guitars
• Vocals
• (Everything Else)

Obviously, you could be more or less granular here, but you get the point. Now you can see where the mastering engineer would have more control over individual elements that make up the mix. So that overly bright vocal can be fixed without affecting the other elements that comprise the song. Same goes for countless other issues including the level of each of the stems.

Issues don’t have to be as drastic as the ones pointed out. Sometimes even well-mixed songs can benefit from a bit of additional clarity and processing on individual elements. Mix engineers typically spend more time with the material so ear fatigue and just getting accustomed to how things sound can happen.

With the stem mastering approach, you end up with material that is more punchy, clear, and balanced with fewer tradeoffs. If this mix needed a bit of help from the get-go, this can be exactly what is needed.

Why Stem Mastering

Stem mastering is a product of the modern music production environment. Mix engineers are often asked to do more with less in a shorter period. In many cases, there are also fewer people involved in the music production process. One person may have done the production, recording, editing, and mixing of the material. With fewer ears working on a project and less time available, it’s much easier for issues to creep in and persist throughout a production.

Another thing to think about is that more people are dabbling in audio engineering. Often they are working in less than optimal listening and monitoring conditions. These factors either in part or whole can allow a lot more issues to creep into the production before the mastering stage.

In all of these cases, the perspective provided by an experienced mastering engineer providing stem mastering can be invaluable.

Recently I received a mix that had some issues in the high end. It was difficult given the stereo WAV file of the mix to determine just what was causing the problems due to the wall of sound coming out of this rock track. What were the biggest offenders? Was it the cymbals? The electric guitar? A combination of both? Come to find out it was a combination of both. I wouldn’t have known exactly or have been able to correct these issues without the individual stems of the song.

But what about cases where the mix is well balanced?

Even in cases where the mix is well balanced, stem mastering can still be beneficial. After all, it is still another set of ears in another environment. The benefit here is that where there aren’t significant issues, small things can be corrected without affecting the overall balance leaving more of the mix balance intact. Stem mastering can increase that amount of clarity and loudness that is capable out of the mix.

How is Stem Mastering Different Than Mixing?

This is a good question with a cloudy and non-specific answer. The truth of the matter is that it can certainly be similar to mixing and given that the stems of a song are given to the mastering engineer, it’s very easy to rebalance the mix during mastering. The reality of the situation though is this may very well be what needs to happen.

How I look at it is the mix engineer is the one who sets up the initial balance and levels. They process the individual elements and buses of the song. Mixers get the potentially hundreds of tracks down into just a few subgroups. They also add the creative effects and set the direction of the mix along with the artist and producer. They should be free to explore their creativity.

The mastering engineer is in a different room with a different monitoring, listening environment, and perspective. The mastering engineer typically is in a more critical listening environment and taking an approach that is more clinical. Both the mixer and mastering engineer working together to create something better because it’s the sum of both their work. Teamwork, that’s what music is all about.

The mix engineer is the pitcher throwing a great pitch, and the mastering engineer is the batter hitting that great pitch out of the park. Yes, I know the pitcher and the batter are on different teams, but you get my point.

Stem Mastering Viewpoints

Some mastering engineers don't like stem mastering. They feel that the onus of getting all of the elements balanced should be placed solely in the hands of the mix engineer. In a perfect world, this would absolutely be the case. But we all know in the world of modern audio production this just isn't the reality we are living in. 

Personally, I'd rather help someone that needs it for their production and do the best I can for them. If they can work on something and be proud of it in the end, then I've done my job. 

Stem mastering can also be a good learning tool for mixers who are looking to gain experience. Hearing the result can help when applying mix decisions to future songs.

Downsides

So now that I've made Stem Mastering sound like some sort of silver bullet, I think it would be an excellent time to point out a couple of potential downsides.

The first thing to keep in mind is that it's certainly possible to make things worse. Things like an inexperienced engineer, sub-par room, poor monitoring, and a variety of other things can mean that the mix is even more unbalanced than it was previously. Ouch!

Stem mastering can cost slightly more than traditional mastering services. This is because there are more tracks and it takes more time for setup and completion of the master.

It should also go without saying that stem mastering cannot fix a bad recording. A poor performance or severely compromised files are something that can't be fixed by this process.

Moral of the story is work with someone you trust and create a relationship with them.

Our Approach To Stem Mastering

We will work with you to get the necessary stems and set them up in a session making sure there are no problems with the files. We can provide insight on which stems to provide given your particular songs and genre.

We will ask that you take any dynamics processing on the master bus off (this is because the balance can change without all of the elements flowing through the master bus).

Do not adjust any faders or perform processing tasks like normalization. We want to be able to pull in your files, hit play and hear the mix pretty much as you hear it.

Just like most projects we handle each project differently based on what it called for but being a hybrid environment, there is almost always an analog component.

What you receive in the end is stereo masters for your project which you can then use for distribution just like you would in a traditional mastering scenario.

Conclusion

Stem mastering can be a great tool in the music production arsenal allowing the mastering engineer to make corrections and enhancements that they couldn’t do in a traditional mastering scenario. This flexibility can be beneficial for both balanced and unbalanced mixes alike.

Stem Mastering for us is a collaborative effort, we work with you to exceed expectations and provide you the best result possible.

If you have any questions about The Freq Zone’s Stem Mastering services or any services we provide, please don’t hesitate to reach out for more info {at} freqzone {dot} com

http://freqzone.com/mastering

Mix Bus Processing Isn't Mastering

This week I thought I’d take some time to cover something I end up talking about quite a bit. It’s not uncommon to see people referring to their “mastering” chain and it’s a few plugins on a mix bus. Let’s clear something up, that’s still mixing. If you decide to work with just plugins on your mix bus, then you have decided not to get your music mastered.

The mastering stage of audio production and the processes it contains can’t be distilled down to a couple of plugins on a mix bus. I spend a bit of time covering subjects and issues like this in the book I’m writing, but for this post,  I'd like to really quickly hit a couple of highlights.

• Mastering isn’t just aesthetic processing
• Loss of Perspective
• Different Workflows
• Different Mindsets
• Lack of Quality Assurance

Mastering isn’t just aesthetic processing

The activities that make up the mastering stage of audio production go beyond just the aesthetic processing that some often associate with the process (louder, brighter, etc.). The engineer is identifying issues that have cropped up or persisted through the recording and mixing process as well as preparing the audio for distribution on various platforms and encoding formats. Error checking, metadata, deliverable creation and many other components make up the mastering phase.

Loss of perspective

It’s no secret that mixing a song takes longer than mastering a song. It’s easy to lose perspective in these situations and become accustomed to hearing elements of a mix a particular way. So if there is a problem, the problem starts to sound normal.

Different Workflows

Most modern mix engineers don’t mix a song in passes. What I mean by this is they don’t start from the beginning of the song and play in through in passes working on the song as it plays. This was much more of a thing when tape was used in the mixing process. Today many mix engineers will tackle various parts of the song in pieces.

Mastering, on the other hand, should be done in passes. This way you are looking at the song as a whole and not in its various pieces and how they fit together.

Mastering and mixing are two different mindsets

This is a subject I don’t often hear people talking about. Mastering is more surgical and precise than mixing. It’s part creative and part science. Rather than working with a distribution of elements in a frequency spectrum, we are working with a distribution of frequencies that have elements in them. We also have to deal with physical limitations of formats and the expectation of the public as it relates to the various genres we are working on. The analytical side kicks in with correcting of errors as well. Mixing leans much more to the emotional side, simply doing what feels right even if that is doing something extreme.

In mastering, it's a balancing act between frequencies, emotion, and elements trying to come up with a best of both worlds compromise that is better than where the mix left off.

Lack of Quality Assurance

It’s not possible to be your own quality check. This is especially true in the moment while you are mixing. Different than the previous point I made about the loss of perspective, this one has to do with someone else performing the master. A different set of ears, in a different room, with a different monitoring system goes a long way to catch issues before releasing music out into the world and ensuring it sounds as good as possible.

What You Should Do?

Remember when music creation process used to be collaborative? Make it that way again. Pass the mastering on to a mastering engineer, someone who makes the process their specialty. The right engineer will provide perspective and make your work that much better.

If you decide to master the music yourself, print the mix and master the music in another session. Take a break and come back to the song later, preferably a day or two after you have put some time between the mix and mastering process. Why put it in another session and not keep it in the mix session? Because there is too much temptation to just start mixing again. Maybe that's what needs to happen, but there is less temptation when just dealing with a stereo mix. You may even surprise yourself.

Audio Reference Gap

Comparing Audio

Let's talk about something that every audio engineer does and is of particular importance in a mixing and mastering context, that's comparing audio. If you are mixing, you may be comparing your mix to a rough mix of the song that was provided or maybe you are comparing to a commercial release. 

In mastering it's critical to compare the mastered version of the song to the raw mix to ensure that the processing steps taken are making an improvement to the mix and essential elements from the mix are retained. The small subtle details are crucial at this stage. 

I've started to use a term recently to describe accuracy issues with comparisons of audio material. This is something I call the Audio Reference Gap.

Audio Reference Gap

The audio reference gap is a gap in process or quality causing inaccuracies with A/B comparisons. The larger the gap, the more inaccurate the comparison.

The reference gap consists of the following items. Timespan, DAC Difference, Level Difference (optional)

Compromises in these areas make it difficult or far too inaccurate to make any reasonable comparison. Why is this important? Well, because decisions made during this process are made off of a skewed perception of the audio.

This has to do with things like the length of time between the comparison or the physical difference of the audio between the comparison (made by different DACs).

Let's take a look at a few of these areas in more depth.

Timespan

The length of time between comparisons is critical to the accuracy of that comparison. The human brain doesn't have the capacity to remember details of an audio comparison beyond about a second. The more subtle the difference the closer together the comparison should be. This means that if someone were to listen to audio, get up and patch a piece of equipment in and listen again, they wouldn't perceive any of the details of that comparison. Sure, if something is so stark in contrast that it's overwhelming that may be perceived, but we are talking about being able to recognize details, even minute details while performing audio processing tasks.  

The timespan of the comparison should be kept as short as possible and well under the 1-second mark for critical comparisons.

Monitoring Path Difference

Differences in the monitoring path can have a pretty large difference in the way that two pieces of audio are perceived. The monitoring path consists of everything from your Digital to Analog Converter (DAC) to your ears. 

Comparing audio through two different DACs adds problems to the comparison. Not all DACs are created equal and different specifications, converter chips, components in the signal path can affect the audio in various ways.

One DAC may be clear and punchy, and the other may be less clear and have problems with low-end focus. So if you are comparing a mix or master, you may be making adjustments to the audio that are unnecessary or have a negative impact. 

Mastering is all about subtleties so even if two different DACs are both excellent and nearly identical they are still different and compromising your comparison.

Level Difference (Optional)

We all know that level difference can fool us into believing one piece of compared audio is better just because it is louder. The reason I added this as optional is that there may be times when you are comparing audio when you want to know about the level difference. 

Better Comparisons

The point of all of this is to reduce the reference gap and remove as many obstacles as possible. This will put you on the path to working better and making better audio decisions in all of your processing tasks.

The timespan should be kept as short as possible to ensure your perception of the material is as accurate as possible. This span should be well under the 1-second mark preferably to where it feels like it's instantaneous. A good monitoring controller is essential for this and will vastly improve your workflow.

The same DAC should be used for all A/B comparisons. This way inconsistencies won't cause you to make inaccurate processing decisions. I can't stress enough the importance of having good digital to analog converters. Since every decision in recording, mixing, and mastering (assuming digital audio is involved) is based on your DAC.

Lastly, if level difference isn't something you are comparing in your A/B, then you are going to want to make sure you level match the material you are comparing. This will allow for a more reasonable comparison and not get fooled by the louder version.

Hopefully, this gives some food for thought while you are performing your critical audio tasks or considering purchases for your studio. Your workflow and audio quality will thank you.