Effectively using your Mid/Side EQ:
To effectively take advantage of your mid/side EQ, you are going to want to make sure you have some way to isolate the mid and side content for monitoring. Most Mid/Side (henceforth referred to as M/S since I am tired of typing that out so much) plugins will have buttons to isolate the content, but if they don’t, then there is a chance you will have a stock plugin within your DAW (like Mixtool in Studio One). Or you can use a cheap or freeware plugin like bx_solo from Plugin Alliance.
Start by isolating your mono/mid/sum material. You are now going to listen analytically and compare this to your reference tracks. If you feel you need more low end, use a low end shelf. If there is too much low end, cut with a low end shelf or peak. Your goal is to control and reshape what needs to be, and leave alone what does not.
Now pay attention to your low mid-range. Use EQ peaks to find resonances and decide if the sound is thick and muddy or sounds too thin and lacking body. As you may have guessed, if there are strong resonance and mud, some precise cuts will be necessary.
Go back to your reference tracks and start checking your work so far and listen on your different sources now (different monitors, headphones, etc.). Be aware of what your moves so far have done and make any adjustments or corrections.
As you are moving up the frequency range, now pay attention to the mid-range and top end and frequency range. This is where you (should) have the most important content in your mix, such as melodies, vocals, lead parts, snare, etc. Go back to your reference track and listen for similar content. If this content seems too low, use a wide band to bring it up. Be aware of any resonances and harshness that may have arisen with that boost. Try to minimize the amount of fine precise cuts you do make in this range since the natural sensitivity of the human ear to this frequency range will make anything you do more prominent, good or bad.
Now you have the top end. This is where you will either add air and height, or potentially create harshness and thin out the entire sound, which is not what you want. Try a strong shelf that starts above or around 10 khz and try pushing it up until you can just notice more air and height, without going to the point of harshness and making the track thin and jarring.
Reference, reference, reference. Even if you have confidence in your monitoring, you absolutely have to reference and constantly have an internal comparative analysis of your moves. Otherwise you are working in isolation which is detrimental to your master.
Now you are ready to work on the side content. What you do with the side content is generally more forgiving, but will also require you to pay attention to the composition of the source and interpret what the important elements are along with what you want to enhance or de-emphasize for creative purposes.
Start by using a high pass filter on the side content up to around 120 hz. This will ensure that no side bass content will cause any type of phase issues in the low end, and give you a solid low end by emphasizing the mono bass content. You will now basically follow the same methods as above, but what you are focusing on is going to change. A rock track for instance is most likely going to have hard panned guitars and other elements like stereo drum overheads and room captures, along with potentially having the harmonies of instruments and backing vocals within this space in the stereo spectrum. You may find yourself wanting to boost the low mid-range on the side content to bring up the strength of these panned instruments, and make more precise cuts and boosts up the mid-range to feature things like background vocals without them overbearing the mid (stereo spectrum) content and drowning your lead vocals or melodic elements. Of course you should reference the entire time and also at this point have taken a break or stepped back and listened critically.
Make sure you are bypassing the plugin occasionally and being aware of what you have done so far. You can now throw a liberal shelf on the high end and gain some more height and perceived space, but make sure you are also bring the mid content back in and comparing so your relative levels between the M and S content are not becoming too unbalanced. Avoid harshness, and if there is any harsh and glaring content in the high end of the side content you have a bit more liberty to make precise cuts, but again, do not belabor this process.
If you have made good moves with your M/S EQ, when you bypass and enable the plugin, you should hear a noticeable to massive difference and be able to distinguish if you made things better or worse. Of course gain matching is essential at this point since the bias of louder being better is going to be problematic if not addressed. You are well on your way to a workable master, but far from finished.
Listen to the difference between these two clips. The first clip is of the fully unmastered track, and the second is of the Mid/Side EQ applied. There is a greater sense of stereo image, depth, and balance.
Dynamic EQ has become a very popular tool in the recent years and has now been widely adopted by all types of audio engineers. The more prominent tool that was a precursor to modern dynamic EQ’s is multi-band compression. The advantage of dynamic EQ is more natural crossovers and also the ability to shape curves more accurately and get a better visual interpretation of your moves. I have only recently adopted it in the past year or so once realizing how powerful of a tool it is. I find the ability to dynamically shape a certain frequency range and either cut it when it passes a threshold or leave it be when it does not is invaluable. I also do the opposite, using it as an expander to make content dynamically pop more without manually having to automate the level or dynamics of that frequency range.
For example, I will of use a dynamic EQ to manage the low mid-range of a busy and potentially muddy song by only using a small initial cut, and then using the dynamic controls to cut more when the content fills that range and my dominate and unbalance the song overall. I also like to use the expansion options when it comes to a track that may lack vocal level and strength in comparison to other content in the frequency range. When the vocals come in and the frequency range is louder, the expansion will bring the vocal up but bring the level back down once the vocal is no longer there.
Again (you should be expecting this), reference, reference, and reference some more! At this point you are going to be making decisions that are reshaping the overall dynamics of your track. Every move you make is dynamically changing relative balancing of frequency ranges, and you must be totally conscious of your decisions and prevent massive shifts in dynamics across the frequency range.
Now compare the first clip to the second clip. The first clip is just of the EQ applied, and the second is of the dynamic EQ being applied. There is now more control on the low end and low-mids along with bringing up the presence of the vocals when they are featured.
Limiting and Metering:
At this point, I strongly suggest you do the car test on your master thus far, including other sources you have available, and again continue to reference. Also be aware that some sources are not going to sound good even with your reference material along with your mastering work. Try and note what you can hear with them, and how you can work with that, and go back to fix things in order to make the best of a bad situation. For example: trying to get your master to sound good playing back on your crappy mono phone speaker.
This is where managing your expectations comes in. You may think you have done some excellent work (and you may well have), but be aware that you are working in a less than ideal situation and there will be things you miss. I will restate once again, we are looking for passable work here. Unless from the time you started reading this you have constructed a full mastering chamber build out and acquired the monitors and other gear you need to take sexy mastering studio photos, acknowledge that you are learning and building a basic skill set.
At this point you should be ready for the final process of mastering in which you will be establishing your levels. I will be informing you of the basics of using your limiters and achieving the levels you desire or are required to, and I will get into the Loudness War and why achieving an optimal level is infinitely more important than getting your track anywhere near as loud as possible.
You will need to open and potentially set up your metering suite at this point. Unfortunately even some of the most popular paid metering suites do not accurately measure RMS or LUFS. I suggested above some free metering suites, but you will also need to do your research and verify the accuracy of what you are seeing since even a single dB above a LUFS or RMS rating may result in the normalization of your track or potentially a significant loss in quality when your track is re-encoded to a much lower bit rate for streaming sites.
A loudness goal in LUFS is the next decision you will be making. When I am mastering, I often do multiple masters with different limiter settings to create different versions meant for streaming, broadcast, and physical formats like CDs. Let’s consider a standard for iTunes’s Apple Music streaming being -14 LUFS. This will be your loudness goal, and also being aware that with this lossy streaming format (usually around 320 kbps AAC/MP4) you will need to leave at least some ceiling headroom. Leaving at least half a dB of headroom will prevent any degradation due to clipping, and will be virtually imperceptible to the listener. While leaving a decent level ceiling may not be essential on all content for iTunes distribution and streaming, making it standard practice will ensure that you have the ability to have your content streamed on practically any service no matter the quality with optimal results.
Since I prefer to use the Ozone Maximizer I have the ability to “cheat” and use the automatic level control where I can playback audio through the Maximizer at the loudest point and get an accurate overall integrated loudness level. While this is very convenient, it’s easy to also achieve the same results by also playing the audio through at the loudest point and check the LUFS on your metering suite, and setting the threshold of your limiter accordingly. Later in the article, I will provide you with resources that will give you the ideal loudness levels for the formats and streaming services you may be mastering for.
At this point you will also have the opportunity to change settings on your master limiter such as the attack and release controls, along with transient and character controls if your master limiter has them. Think of these controls just as you would on a standard compressor, since essential a limiter is just a very high ratio compressor. Slow the attack to let more transients through, maker it faster to get more squish and control. Modern mastering limiters may also have transient controls which give you another dimension of control. Changing these settings may give you a more “transparent” sound and interact better with transients, but the more conservative your master levels are, the less effect these controls will have. You will also at this point want to establish a ceiling, of which I would suggest about -.5 dB. At this juncture, your master is now just that, a master.
Now, you may be able to call it a true workable and decent master at this time. But, again, you need to reference across all your sources yet again and be ready to critically listen to your work. Remember, if you are using this guide, you are working in less than ideal circumstances, but you are creating quality content through good reference comparison on sources and to other materials.
Listen to the first clip, and listen to the second clip. On the second clip the limiter (Ozone 8 Maximizer) is set for moderately strong transient preservation along with a conservative level of -12 LUFS. While louder, it is still articulate and the dyanmics are still easily perceived.
What Makes a Good or Bad Master?
If you have survived this entire article then you have most likely have developed an ear to distinguish the quality of a master and also over-confidently brag to your friends that you are now an expert with a discerning ear. At this point you have hopefully finished your master with some degree of success. You should be aware of what makes a good technical master. When you hear a good master (especially if you have heard the unmastered mix), it should not betray the original balances and intent of the song, it should reveal and not obscure elements in the mix, and it should not be blaringly loud since the Loudness War is over and those still fighting it are fighting only with themselves. The master should have superb translation and even on the worst of sources still have intelligibility of the major elements in the song. While this may all be based on my own opinions, there are the facts to back it up. With all the information and technology we have easily available today, there is no reason to still be following bad rules and ideas established in a bygone era.
As previously stated, this guide is to inform you on how to do a workable master on a decent mix. My personal methods for a full client mastering job are more intensive and I use different tools for saturation and coloration in ways that make the sound more interesting while not betraying the genre or intent of the artist. For instance, I am not going to put heavy dark tape saturation on a modern pop track, and I wouldn’t try to make dirty raunchy rock track as clean and depthy as possible. Good taste and methodology will be something you develop over time. As you invest time and money into the craft you will be able to make these aesthetic decisions for the sound.
The Loudness War and Loudness Penalties:
The Loudness War is ending. The ideology of the louder a master is, the more people will hear it, and the better it sounds has been proven time and time again to be false and a futile pursuit. Yes, you will still see and hear renowned engineers mastering to extremely loud levels and also engineers constantly seeking the most “transparent” and “dynamic” (which is a fallacious way to refer to a limiter as is) since they are also working towards achieving excessive levels, and believe that they can maintain the perception of dynamics while also destroying them completely.
I like to refer to the streaming service’s website to learn what the ideal loudness levels are specifically for the services I know my clients will be using; along with being knowledgeable about physical format standards like CDs (optical media is not dead!). Fortunately there are modern free resources like Loudness Penalty: Analyzer, where you can upload your master and see what streaming services will do to your master as far as normalization is concerned. Then, you can adjust accordingly and export multiple masters with different loudness levels. If not, you will be subject to the streaming service normalizing the level of your song and potentially making it much quieter to compensate for excessive loudness.
Even when it comes to formats like CDs where traditionally master levels are extremely high and it is mostly accepted as being good practice, it is a betrayal to the capability of the format. While all CDs are 16 bit with a 44.1 khz sample rate, they are still one of the best and most popular high quality audio formats. Often you will see that a CD master is extremely loud and the dynamic range is next to non-existent. A CD has 96 dB of dynamic range, but commonly you will see popular music with a dynamic range of only a few dB. Just because you can maintain more quality with a loud master on a higher quality format like CD, doesn’t mean you should. Vinyl on the other hand has a smaller dynamic range and smaller frequency range, yet vinyl masters (proper vinyl masters) have much greater dynamic range. They make the best use of the format, and respect the format.
I hope you gained the knowledge and skill set to create proper masters, and also inspired you to expand on such. No, this was not a quick and easy guide, nor a very technical and intensive guide, but I believe it is enough to get you on the right path. Truly mastering the art and science of mastering takes time, effort, and investment in your gear and control room. You can practice all you want in a poorly on untreated room, but you will only be mastering to your environment. Use the vast amount of information online, (and this website specifically) to gain more knowledge and even if you are just a hobbyist, having the skill and integrity to create a high quality master is a noble ability and may just improve the collective quality of music, making things better for the consumer and creators.
Best of luck my friends.
Kyle Conklin is an audio engineer, electrical engineer, and PC enthusiast. His goal is to educate the audio engineering enthusiast and professional with no-nonsense information attained through real world experience. Kyle wants everyone to strive for excellence in their every endeavor and create the highest quality content possible no matter their circumstance