Review Zoo Catch-Up with Mastering Engineer Barry Gardner: The Art, Science & Truth Behind Professional Music Mastering

“AI doesn’t listen to music — it processes numbers.”

From First Vinyl Cuts to AI Debates: Barry Gardner on Craft, Creativity & Why Human Mastering Still Reigns Supreme

What first drew you into mastering, and what kept you there?

My first ever mastering job was about 20 years ago. I used to be a full time employed sound engineer for a big production company in London (Somethin' Else - now Sony). A job came in that was to record 6-8 buskers in a day and prepare the recordings for a vinyl cut for an art project.

The vinyl cut need was a baptism of fire and yet the job went very well. I later went freelance as a sound engineer and was asked about mastering and said I could oblige. From that point the mastering jobs grew. I then created my website and that is the story of how I became involved in mastering.

Mastering is the passion in my life, I sincerely love the work, it is not always easy, it is challenging work. I find it enjoyable and seemingly I am pretty good at it ! I simply want to help people make their musical creations sound better. Musical passion tends to find mastering passion.

Can you explain what mastering is, in your own words, for artists who might not fully understand it?

Mastering is the improvement and quality control of music before it is released. Using layers of tools for their primary function (i.e. an equalizer) and secondary (box tone) sonic characteristics. Along with sonic technical optimization for intended use purposes.

Depending on the mix you can make a mix translate better across all reproduction systems, sound more full, classy, deep, smoother, more rounded and so on, nudge tone, nudge instruments/voices and enhance the musicality of a mix.

The key is what you choose to do to enhance the clients music from the 1,000s of tools available. This relies on engineer knowledge, experience and internal musical and sonic references. You need a sonic vision to master music, a vision coupled with the knowledge of what the best tools are to use to get there, with the least side effects.

What do you listen for first when you receive a mix?

The entire mix, wholistically. I simply listen to the entire piece of music. It's mastering so that is what needs to happen. Once I get an impression of the whole, my monitoring system allows me to hone in onto distractions initially. Distractions that may not be enhancing the listening experience. This is usually done in a wordless manner and quite quickly. First objective impressions are usually right. I simply make an auditory note or sometimes immediately select a well known effective tool to quickly and almost effortlessly adjust an aspect of the sonic presentation.

After hearing literally thousands of tracks across all genres, sonic issues jump out within seconds. It has got to a point now where any issues cannot go without being noticed. And that is a good state to be in when listening. It is taking care of the issues that have been missed in production before fans and listenership hears them. I am also considering my internal genre references and how the music in front of my ears fits in.

What separates a “good” mix from a great one in your eyes?

Great mixes incorporate great sound sources, be it recorded sound sources or synthetic. Great sound sources have great sonic qualities. These produce great mixes more easily.

Great sound sources are clear, defined, smooth (or at least not overly harsh) and punchy, they are attractive to the ear in some way. Those descriptors work for great mixes as well, they go hand in hand. A great mix also amplifies and articulates the emotions in the music as well.

Great mixes have good tonality and an instrumental balances which does not draw the ear to any specific element without an auditory enhancing need. Great mixes never have excessive vocal sibilance where a vocal is present which is usually distracting and rather annoying.

Are there times when a track barely needs touching—just that final polish? What tells you that?

In my own personal experience this is rare, but it must always be kept in mind. Fortunately, exceptional monitoring and good internal sonic references are telling you what not to do as much as they tell you what may need doing.

When I do the wholistic listen and my ear has not been overly drawn to any given instrument, song section, specific instrument, technical issue, extraneous noises or element then I know the mix is already very good.

What’s one piece of gear or software you couldn’t live without?

Monitoring, which is actually a system. As such I will go with PMC IB1S, detailed, high definition, very deep bass (and defined), accurate, superb stereo imaging, very well balanced sonic presentation. (The only thing is they are so closely coupled as a system to my DAC - Crane Song Solaris, Nord - Hypex amplifier and room that it is a system rather than a specific individual item of kit.)

That's mastering summed up well, the entire music mastering interaction is an alive, interdependent system between human and equipment, a responsive reproduction/listening system. Where the results for a clients music are greater than just the sum of parts.

This is probably why mastering has been somehow shrouded in mystery. When it all comes together it becomes easier to understand how meaningful improvements can be made to most audio.

What are the biggest misconceptions artists have about mastering?

That's a good and tricky question. In these times it could arguably be that AI is listening to the music that you have created. Let's be clear, in no way is AI listening to the music. It's running streams of numbers for data comparison.

AI aside probably that it is just adding a limiter and that you can comprehensively master music using small monitors. Mastering is really a unique, bespoke listening service for the client here. I employ a considered, skilled response to every piece of music that passes through the studio. It's my passion and I sincerely care about the end results.

Mastering can be different things to different clients and this you get to understand after working with someone a few times. And most times I try and understand what the objectives are in early communications.

What are the real benefits of professional mastering compared to DIY or automated tools?

Objective, experienced human listening using accurate monitors in an accurate acoustic, communication with a listening professional, ultra high quality tools (not server based number crunching), and a respectful, considered response to one of the most powerful human art forms.

I try and provide a sense to the client that they are not on their own with their music. A good ME has got your back. It's a relationship of mutual respect and learning.

How do you feel AI mastering services are affecting the industry right now?

It has been around for about 10 years and I consider it mastering in name only, new and younger generation musicians and producers may incorrectly believe it is some kind of ultimate standard. It really falls short in so many ways compared to what actual mastering is despite a curiosity.

Sometimes a new client will mention these services when they contact me or send a track completed with automated services. (I rarely listen to them, I listen to their mix instead. I may check the LUFS integrated numbers on a meter of the "mastered" file.) I know what good sound is, so I work objectively and from my own knowledge and skillset. Objectivity is very important in mastering and I do not want that all important objectivity to be skewed by something poorly mastered by an automated service (One example being, too loud, too bright which easily fools the human ear into being better.)

Obviously, automated is very quick and we live in a rather impatient world, however the value of true listening to music is not a rushed experience. (and not cold maths either) Automated means you don't get musical listening, and musical listening is a powerful factor related to mastering. It has a parity with an artist or label's listenership. Which means such musical enhancements will be non existent with an AI service. It fundamentally cannot listen, so musicality enhancement cannot be expected.

AI is ultra low cost and quick, but at what actual cost to the first listen of your music by listeners ? Which people have typically put a lot of time and effort into. It does not offer mix feedback (double checking something never harms), does not listen in the true sense of the definition of that word. You really need an experienced human professional to engage, consider and feel your music, music is human, not just numbers.

Neither does it hear clicks and glitches, missed bad audio edits, edit out count ins or not, automate EQ or dynamic processes, edit tracks and process different parts of a track differently (that happens more often than one might think) It cannot use analogue equipment, cannot work at an upsampled sample rate, cannot sympathetically space music for an EP or album project, balance multiple tracks sensible by ear for a human ear. The list goes on and on.

Sentience is the unique awareness of a biological being, an evolved organism. Human music reflects life, feelings, and many times has a back story. Human music is an organic product even if it is synthetically produced.

The substrate of a listener's intelligence is organically evolved neurons in a brain that took billions of year to evolve. That is the most complex matter thus far discovered in the universe. It is intrinsic to a unfathomably complex human psyche and body, not predictive processing in copper and silicon. Humans should never denigrate themselves hastily, just because we are at times in awe of new technology.

Do you think we’ve reached a more balanced place after the loudness wars? What changed?

Yes, streaming is volume normalized and of course this has helped greatly in some respects. I have never been against loud and yet very good sounding mastering. For CD and non volume normalized platforms there is no logical need to produce specifically low, LUFS integrated masters. As long as the sonic integrity is maintained within the boundaries of good taste.

There is usually a fine line between loud and overcooked and after many years this is easy enough to judge. Volume normalization is not perfect, it is not an exact mirror of human volume perception.

There are music genres that still sound great when loud, but how loud is too loud ? This is part of the art of mastering, to find the perceived volume middle path for a musical EP or Album product. Mastering is optimization of a multitude of sonic factors.

I have various adaptive approaches to perceived volumes, depending on what a project needs. Some of this is genre expectation based as well. Lots factors in, but I am confident I get it right the vast majority of time as indicated by a low revision rate. Good mastering is bespoke work, track by track, project by project.

Do you think alternative music benefits from a rougher, less polished master compared to mainstream genres?

The mix dictates a lot before I get to it and yes it can do, however that must always be balanced against listenability. Too harsh sounding results need to be considered, out of kilter instrumental balances, sound amateurish. (demo tape style)

Raw and brash can veer into this territory and I don't think many alternative/underground/aggressive styles of music want to sound intentionally bad. You can usually retain edgy, raw, punchy and powerful in mastering without repelling people from wanting to turn the volume up. Alternative or not, you want people to want to turn the music up, not down.

Again the style of mixing determines a lot of how I approach the tracks. I think it would be good for the mix engineers of heavier styles of music to consider that harshness builds quite easy when you have multiple aggressive sounds occupying the mid range and highs. Take a little time out to reflect on the mixes a day later with fresh ears and assess whether this has gone beyond good taste.

A little tip if your monitors are not revealing this, is to look at a spectrum analyzer and see if there are a vast number of partials/harmonics persistently occupying 3.5kHz through 14kHz. Though quality studio monitors should be alerting you to potential ear fatigue.

What’s the strangest or most unexpected track you’ve ever been asked to master?

An apt question, that also relates to the prior question. That is probably a track that came in so unbelievably distorted it was hard to imagine it was not a serious technical error. And I don't mean heavy tube amp distortion on some guitars. Rather it was totally and completely obliterated and seemingly with intent after checking with the client.

I had to step back and work out what I was going to do. In the end you always come back to listenability and how will this sound if cranked up. And communicate this to the client.

For more info on Barry, please see below

Barry Gardner's website : Safe&Sound Mastering - Barry specializes in high end mastering that remains very much affordable to all.

https://www.masteringmastering.co.uk/