Transcripts
1. Introduction: Hello, welcome to the visual mixing formula. In this course, you are going to learn how to mix an eight steps using essentially your eyes. First, I will show you how to set the panorama to create separation and stereo width. Then you're going to learn how to structure the input gain for optimal plug-in performance and how to filter the low end of each track for optimal speaker performance. After these steps, I will show you how to balance your mix without even hearing using an exclusive reference chart. And we will learn how to equalize each track in an objective way with the help of spectral readings taken from professionally mixed tracks. You will also learn how to use the compressor in a visual way without overthinking it. After you will learn how to objectively said you're effects without any guessing. Finally, you will also learn how to set a central mixed bus effects as a final refinement of the process. Although you can use it that way, it is important to mention that the goal of the visual mixing formula is not to deliver a completely finished mix. In fact, the main purpose is to deliver a great starting point from where you can make artistic decisions without worrying too much about technical aspects. Let me show you a quick before and after of the song we mixed in this course using the visual mixing formula. My gosh, am I stupid?
2. Panorama: Setting the panorama is basically defining the stereo placement of the elements in our mix. We can position the instrument on the left, on the right, or on the center. Kick Snare, basically vocals and solos as the main elements of our song, should always be placed in the center of the stereo image. Everything else should be panned, either left or right to make room for the main elements in the center. In this course, we are going to use the free EMG audio track control plug-in to set the panorama of each track. The balance control defines the stereo placement. 0% is center minus 100% is hard left and 100% is hard right? Kick, snare based on lead vocals, as we already discussed, should be placed in the center. Tom's should be Pan Like a handheld fan across the stereo. The Thompson this song, however, are already printed to a stereo file. What leaves us with no option other than move on to the next track. Hi hat should be placed 50% left or right. In this case, I'll place it 50% left. Everything else should be panned 100% left or right to make room in the center for the main elements. Now, let's compare before and after processing. Before we move on to the next step, let's group our stereo pairs into folders so we can process them together. Just insert a new track, place it next to a stereo pair. Select the tracks that are going to be grouped and drag them inside the new track you have just created. Now repeat the process for every other stereo pairs.
3. Gain Structure: Now we need to structure the gain of our tracks. Drums and percussion should be leveled at minus ten decibel speak. Everything else should be leveled at 0 decibels VU. This step is important because most plugins have their sweet spots at these levels. To perform this task, we are going to use the free MV meter by TB pro audio. This little knob controls the input gain. We are going to use it to adjust the levels while looking at the meter. As we already discussed, drums and percussion should be leveled at minus ten decibel speak. Do not forget to change the reading of MB meter to peak when dealing with drums. I will speed up the video while we adjust the rest of the drum so you don't get bored. Everything else should be leveled at 0 decibels VU. Again, I will speed up the video to keep you awake. Let's hear how it sounds. As you can see, this process takes a lot of time, but you can do it way faster with the help of the horn at VU meter. This plug-in basically automates the process of Gain Staging. Set the maximum peak to minus ten decibels for drums and click the auto button. Now, copy the plugin to other tracks. Don't forget to change the maximum peak to 0 decibels for non percussive instruments. Once you are done, just hit play and let the plug and do its work. Let's do a quick comparison. Now that we have properly set up the gain structure, our tracks, let's start the processing.
4. Hi-Pass Filters: The third step is creating high-pass filters for each track. It's a really simple step and we are basically going to follow the reference chart. Just insert any q of your choice and set the high pass filter to the frequencies indicated in the PDF file for each instrument. Kick drums and bass guitar should be high, passed at 30 hertz. Tom's at 50 hertz. Acoustic guitars at 250 hertz. High hats and overheads at 500 hertz. Everything else should be high, passed at 100 hertz. Let's compare. Now, we can move on to the next step.
5. Balance: The fourth step is probably the big star of this course. We are going to use a reference value fs level chart to get a rough balance for each track. How you FS stands for loudness unit full-scale. In short, BLUF S measurement aims to get as close as possible to the way humans perceive loudness. To make this method work, I analyzed dozens of professionally mixed sessions instrument by instrument to come up with average values that can be used as a starting point for most mixes. The result of this analysis is a PDF reference chart that is available for download to everyone enrolled in the visual mixing course. To accomplish the task of balancing with LFSR, we are going to use the free lean loudness meter plug-in. Most of the time we are going to be looking at the integrated measurement. The only exception is the overheads for which we have an additional momentary measurement available to be used to balance Crash Cymbal hits if necessary. The process of balancing consists in simply finding the suggested LFS level for each instrument in the PDF chart and manually adjusting the faders while looking at the ALU fs meter. Let's get to work. The reference level for the kick drum is negative 27. Efs. Perfect. Snare drum also have a reference level of negative 27 LU fs. Good. Let's move to the next one. The wagon floor Tom's together should be hitting about negative 27. Nice. Clothes high has reference level is negative 45. Next one. Overheads when their heads are closed, should hit about negative 39 AFS or so. Rooms reference level is around negative 45 ALU fs. Perfect. Base should be hitting negative 27. Efs. Notice how the foundational elements, such as Kick, snares, toms, and base, all share the same reference l you fs levels. Acoustic guitars reference level is negative 33. Fs. Gun. Clean and crunch rhythm electric guitar should hit negative 33 ALU FS. Let's repeat the same negative 33 LU FS For the next pair of electric guitars. As she finally vocal should be around negative 24 LU fs as a starting point. As obviously, you can tweak this numbers to better fit the specific song you are working on. Don't be afraid to trust your ears. Now, let's hear how it sounds. As you can see, balancing each track manually takes a lot of time to save a few minutes and every mix session, you can use the Hornet LU meter plug-in for automatic balancing. Let me show you how fast it is. Insert the Hornet LU meter and change both the reading to integrated and the maximum peak to 0 decibels. Click the auto button and do not forget to adjust the reference level to the suggested LFS value for each instrument. Make sure there's no crash cymbal being hit at the selected starting point. Press Play and let the Hornet LU meter automatically balance your tracks. Finally, let's compare how we started this project with where we are right now.
6. Equalization: For this step, we're going to use a spectrum analyzer to visualize the frequency content of our tracks. A spectrum analyzer measures the loudness of each frequency of the audio signal. My plug-in recommendation is the free books and go span. After loading the plug-in, you need to adjust the settings to make the analysis a bit easier. Click the gear button to open the settings panel. Change the type to Max and the block size to 65,536. Set the overlap to the max value, change the smoothing to 1 third and activate the anti-aliasing. These adjustments will make the spectrum easier to read. Now the idea behind the visual equalization process is really simple. We are going to compare the spectrum of each track in the session with reference screenshots taken from professionally mixed tracks. The Quickstart folder contains pictures of overlap spectrum references for a better visualization of average levels. Let's compare the kick drum of our track with a reference spectrum. It seems we can boost some highs and lows in our kick to make it a bit closer to the reference. Perhaps we can also cut slightly the 100 hertz region to better match the reference. We are going to use the same equalizer that was used before to set the high pass filter. Let's add some highs first. Then let's boost around 60 hertz. Finally, we are going to cut a bit of the 100 hertz region. Let's compare our spectrum with the reference one more time. It looks like we need a bit more of 60 Hertz. Perhaps a deeper cut at 100 hertz. Great. After equalizing, Don't forget to readjust the LFS levels to maintain the initial balance. Now that we're done with the kick, let's compare the spectrum of our snare with a reference. Looks like it only needs a bit less of that one k to 1.5 k area. Let's use are eager to cut some energy of that region. Again, try to always remember to rebalance the LUMO levels after equalizing. It is time to deal with the toms. Our track needs a bit more highs, a little less of the 600 hertz region, and possibly a tiny bit less low-end. First, let's add some top end. Now, let's take away some mid-range. Finally, a small dip at 100 hertz. Let's compare our spectrum with the reference one more time. Looks good to me. Once you are done, don't forget to readjust the ALU fs balance. Good. Now we can move forward. We're not going to use any EQ on hi-hats, overheads, and Rome's other than the high-pass filters that we have already said. You can equalize them if you want. But I never do. Let's jump straight to the base then. The low end seems to be pretty much inside the range of the reference spectrums. The highs are lacking a bit of energy though. Let's try to fix that with a high shelf filter in the equalizer. Good enough. It is time to check the acoustic guitars. Let's compare the spectrum of our track with a reference. These acoustic guitars clearly need more highs when compared to the reference. Also goes a little built up between 12 kilohertz. So there are two things to be done with these acoustic guitars. We need to cut some meds and also make them a little brighter. Let's take care of these two things with our EQ. Let's compare with the reference one more time. We can increase the high-end boost and get a bit more aggressive with the mid-range cut. Don't forget to rebalance the ALU fs levels. Once you are done. Let's take a look at these electric guitars. Now. I will open the reference spectrum so we can compare. Looks like these guitars could use a high-frequency boost. Notice how they lack in high emit energy in comparison to the reference. Let's open the equalizer and try boosting the highs. I don't want the high boost to affect my mid-range. So I will cut a little bit of one kilohertz to avoid bringing up this region. Now, let's eat you the next pair of electric guitars. Apparently, we can cut some of the 500 hertz area and use a little high-frequency boost. Let's try that with the equalizer. And last but not least, let's check the vocals. Fast shoes. These vocals need a high-frequency boost. We can also cut some energy of the 500 hertz area. Let's pull up the equalizer and get to work. Fast shoes. But I know my stupid fast shoes. Don't forget to rebalance the ALU fs level after processing the track shoes. Now that we're finally done, let's compare how it sounds with and without any processing.
7. Compression: In this lesson, we won't go deep into the details about compression parameters, such as threshold, ratio, attack and release. Instead, I will demonstrate how I use the compressor and practice without overthinking it. The first important thing is that we are going to use analog compressor emulations. This is because hardware compressors have programmed dependent behavior, which means that they automatically adapt to the type of input signal. In other words, Program dependency makes analog compressors really easy to use. Also, these classic analog compressors have been proven effective for these specific applications for years by the most renowned mixing engineers. After this brief introduction, let's get to work the two things that usually really needs to be compressed or vocals and base. These two elements must have a controlled dynamic range in order to sound good in the mix. So the compressed vocals and base, we're going to use any emulation of the Mellotron x l2. This compressor is highly programmed dependent. Ratio pack and release parameters are predetermined and will automatically adapt to the source material you feed into it. Again, programmed dependency is the reason why this compressor is ridiculously easy to use. Let's insert the CLIA 2E, a tele tropics LA 2A emulation developed by waves and the vocal track. It has only two knobs. The one on the right is used to choose how much compression you want to apply. And the left one as an output gain control that can be used to make up for the volume loss after compression. Just by looking at the waveforms on the DAW, we can see that this vocal needs compression. Notice how there's a significant difference between the loudest and quietest parts. To figure out how much compression we need to apply, just use the peak meter of your DAW to calculate the difference between those parts. I'm going to select and loop just the specific parts I want to measure to make the reading more accurate. In this case, the quieter parts are peaking at negative 25.4 decibels. The louder parts are peaking at negative 14.6 decibels. So do even out this performance, we need to see roughly nine decibels of gain reduction during the latter passages. Remember to place the compressor before the Hornet LU meter to make it easier to rebalance the track after compression. Vast, vast. Vast. Vast shoe. Yes. Fast shoes. Scarcity. Fast shoes, shoes, shoes. Shoes. Scarcity. Piano, my stupid. Don't forget to readjust the LUMO level to maintain the balance we had at the beginning. Fashion. My stupid. Now that we're done with the vocals, let's move onto the base. The idea with the bases the same. You're simply going to spot the difference between the loud and the quiet parts and then aim for that amount of gain reduction with the LA 2m relation. The quieter parts are peaking at negative 22.8 decibels and the latter parts of negative 19.2 decibel. This means we need about four decibels of compression. So let's do it. Perfect. Now that we have Bayes and vocals dynamic range under control, it is time to take a look at the other elements. For everything else in this mix, that compressor will be used just as an effect as the remaining tracks do not have an actual needed dynamic range control, the goal is to compress each track subtly, just to add a bit of coloration to the sound. To perform this task, we're going to use an emulation of the SSL channel compressor. There are lots of SSL channel emulations out there and the vast majority of them sounds great. Here I'm going to use the CLI mix-up developed by waves, but it could have been any other. The thing about the SSL channel compressor is that it is really versatile when set to slow the attack time of the compressor works in a program dependent way on automatically varies between three milliseconds to 30 milliseconds, depending on the source material. Just set the ratio to four to one who released the fastest and use the threshold knob to control the amount of compression. I am going to move the threshold knob until the gain reduction meter indicates three decibels of compression. Remember to readjust the ALU fs level to make sure we don't mess with a balance we already had set up. Great. Now we are going to repeat this process always any for roughly three decibels of compression for the remaining tracks, with the exception of hi-hats, overheads and rooms. Snare is done. Let's move to the toms. Perfect. Now it's time to check the acoustic guitars. Let's insert another instance of CLIA mix up. We're done with the acoustics. Let's work on the electric guitars. Finally, it's time for a quick comparison.
8. Effects: Now it is time to talk about effects. The most important thing when it comes to effects is balanced. It is really easy to destroy a mix by drowning everything in reverbs. To avoid this problem and take the guessing part out of the equation. I created a PDF chart with average value fs levels that can be used as reference to balance your basic effects. Let's see how it works. Create a new track and insert an effect plugin of your choice. In this case, I'm using the Valhalla vintage verb. Let's select a vocal Hall preset, but send a copy of the vocals to the reverb track. To do so, we simply need to drag an empty send space from the vocals to the reverb. For this to work properly, you need to make sure that the mixed control in your reverb plugin is set to 100% wet. The amount of Sen defines the level of a reverb. Right now the vocal sent is at 0 decibels, which is too much. Let's solve the reverb track and measure how loud it is to find out by how many decibels we need to lower the send. Our vocal reverb is currently hitting negative 23 LU fs, while the reference chart indicates an average of negative 43 BLUF s. This means we need to lower the sound level by 20 decibels. Let's check how it sounds now. Sounds good to me. Let's repeat this process. But now we are going to select a room preset to use on the snare. First, we create a new track. Then just insert a reverb plug-in and select a preset. Let's try this fat snare room. Create the sand by dragging in empty space from the snare track to the reverb. Let's hear how it sounds. The PDF chart indicates a reference level of negative 43 LU FS. Let's measure the loudness to find out by how many decibels we need to bring down the send. It seems we need to lower the sand by 11 decibels or so. Perfect. Now, simply copy the sand from the snare to the toms. That's how I do and it works just fine. Finally, let's create another reverb track and select a plate preset that will also be used on our snare. The reference LFS level is negative 45. This means we need to lower the sand by 12 decibels. It is looking good. Now, let's hear how these reverbs sound in the mix. Now a quick before and after comparison with where we started. Oh my
9. Mix Bus Processing: To finish the visual mixing formula, we need to add a few processors into the mix bus in case you don't know, makes buses how we call the master output channel of the mixing session, or all the individual tracks go through and, and can be processed as a group. The first plugin I'm going to add to the mix Bus is an SSL Bus Compressor emulation. The mixed bus compression helps creating a sense of cohesion for the mics. Here I'm using the one developed by waves that there are many others in the market for you to choose. Set the makeup gain to one that released to auto and the attack to ten milliseconds, then lower the threshold until you see four decibel is of compression on a lattice hits. I like what it does. The next processor will be a tape emulation. I really liked the waves Kramer tape. The tape as a bit of saturation to the sound, which serves to tame the peaks of the drum heads, pulling them back into the mix. Let's compare how this project started to how it is sounding. Now, the visual mixing process is done, but you can feel free to make any further adjustments using your ears.
Alexandre Machado
