FIG. 1: To find the tempo, the original track is cut into a 2-bar section and placed at the beginning of a measure (a), then the sequence tempo is changed until the end of the section lines up with the bar line (b).
A remix is so much more than its name implies. In many cases, it is an entire reworking of a song; the only parts that remain from the original are the vocals — and even those are re-arranged and edited to be different. The word remix is an insufficient term to describe a process that incorporates all aspects of music production from start to finish.
Being creative and producing music is my passion, but it is also my business, so I always need to work efficiently and stay on task. Even though every project is different, I try to approach my workflow in a standardized way to keep me organized and moving forward. I have three main phases in my workflow: First is the basic song prep and discovery phase, second is the creative writing phase and third is the final mixing and mastering phase. In my new DVD course, The Art of the Remix, I demonstrate step by step how I do this. In this article, I'll cover some of the most important concepts involved in remixing.
The first thing I do when starting any remix is to analyze the original song and see what parts I have available. The main goals of this phase are to identify the tempo, line up the vocals and find the key of the song. In some cases, I am given all the original project files so I can immediately get the tempo and also have control over which vocal and musical parts, if any, I want to use from the original mix.
Most of the time, though, I only get the original stereo mix and various a cappella files. I then bring those into a new project and find the tempo. This is easy in Apple Logic Pro, the sequencer I use, as well as in most other DAWs. What I do is cut out a one- to two-bar region at the beginning of the song and edit that region to start right at the downbeat of a measure and end right before the downbeat of the measure that would follow the section. This is easiest during a part of the song where there's a beat with easily recognized transients. Then I make the timeline conform to the region by adjusting the project's global tempo control until the start and end of the audio file lines up with the corresponding bar markers (see Figs. 1a and 1b).
Once I've matched the tempo, which is a nondestructive edit, I simply drag the start and end of the region to reveal the rest of the song file. This works great for tracks that were originally sequenced or recorded to a click, but even in more live recordings, it gets me close enough that I can make fine adjustments later if needed.
FIG. 2: The bass sound on this remix comprises four different MIDI bass patches and a live electric bass.
Next, I line up the vocal parts, and this again varies from project to project. Sometimes, the stems line up on the downbeat, allowing me to place them on adjacent tracks to the original mix and slide them into place. If the vocal stems don't seem to line up easily on a downbeat, then I will trim the region before the start of the vocal waveform, line it up by hand and nudge it until it is right. One trick I like to use when lining up the vocals is to pan the original mix reference hard-left and the vocal files hard-right. This allows me to raise and lower the original, in and out, to make sure the a cappella is precisely lined up.
The next thing I do is spend a little time critically listening to the original song. My main goals here are to explore the key of the song and to dive into the overall feeling of the lyrics and music as I begin to get creative. Finding the key of the song is, of course, critical. There are some simple methods you can use to quickly do this. First, I will bring up a simple piano patch and find the root note of the downbeat. From there, I will identify whether it is major or minor by playing scales of both types against the original mix.
Once I have found the tempo and key and have all the vocals lined up, the last technical step I do before I start getting creative is to change the tempo. This is one of the easiest ways to immediately give the remix a new feel. Even if the song is already in the bpm range in which I plan to work, I will usually still change it just to make it different. Changing the tempo with Logic's new Flex Time feature is simple. Once I have the tempo map of the original mix and my vocals are all lined up, all I have to do is turn on Flex View and analyze each vocal track. After that, I can simply change the tempo of my project and everything automatically follows along (see Web Clip 1). Other DAWs with time-stretching capabilities, such as Digidesign Pro Tools with its Elastic Time feature, make the process equally simple.
FIG. 3: Sometimes adding distortion to the bass helps it punch through the mix better.
After I have spent a bit of time organizing the parts and getting familiar with the original song, it is time for the fun part: getting creative. The first thing I will do is come up with a basic beat with a kick, snare and maybe a hi-hat or some other simple accents. The genre I am going for will dictate the kinds of sounds and patterns that I'll start to experiment with. For projects aimed at the clubs, it is critical to create a DJ-friendly drum pattern, typically built around a foundation of a “four-on-the-floor” kick drum.
After I construct the basic foundation, I will spend some time creating or editing loops and other accents to make the groove have a signature feel. I will add such things as percussion elements, reverse claps and other sounds to create something unique. Because the foundation usually revolves around a simple kick/snare/hat pattern, these loops and percussion parts allow me to get creative with the beat.
On the Bottom
Once I am happy with the initial groove, I quickly move on to the bass line. While writing the instrument parts, I will usually loop an eight- or 16-bar section as I build up the music. (I usually choose a section from tracks underneath the verse or chorus parts of the original vocal so I can simply mute the vocal tracks and constantly reference my instrumental ideas against the vocals.)
FIG. 4: Organization is the key. Here, all the tracks are arranged and color-coded with mix subgroups (in orange).
In modern electronic club remixes, the bass line is one of the most important elements. It carries the weight of the record, and in many cases it can be a feature of the mix. There are two things that make up the bass line: the musical pattern and the actual sound or sounds.
In some cases, I will begin my work on the bass line by focusing on its sonic characteristics. Other times, I begin by working out the pattern and deal with the sounds later. It depends on the flow of the session, and the most important thing is to stay in a creative frame of mind.
Probably the most important thing I do when constructing bass lines is to layer the sounds. One of the easiest and best ways to make signature sounds — and this goes for all of the elements, not just the bass — is to blend multiple layers from a number of synths or instruments (see Fig. 2). By using differing layers, I can create complex textures and better control the frequency spectrum.
For example, to sound massive in the club, I like to use synths with a solid low end that can really drive the subs. In the studio and in the club, this works great. However, when the track is played back on smaller systems, that really low bass line can disappear altogether. So when choosing my bass patches, I will do one of two things to compensate. I will either pick a synth patch as a second layer that has a higher frequency range or I will duplicate the main bass patch and insert a highpass filter and distortion, or a fuzz plug-in, to add some gritty high end that complements the low end of the bass and helps it punch through in the mix (see Fig. 3 and Web Clip 2). My goal when layering is to make what sounds like one cohesive complex sound. I also like recording live electric bass as a layer because it gives the part a nice, live swing feel.
The subject of frequency masking is another important one relating to layered bass lines and the use of layered sounds in general. You have to be careful that the layers are not fighting each other for the same frequency range. If I first use a lower sub-synth patch, I will complement it with another sound with a higher frequency. As I add layers, I will do so in a way that helps to expand the soundscape and avoid too many sounds concentrated in one part of the spectrum.
You also have to factor in the kick drum and how it relates to the bass line and other musical parts. For instance, if I have a sub-heavy kick with a lot of extreme low end, I may choose a bass that can sit slightly above it or vice versa. If I love the kick and bass patches but they're fighting each other frequency-wise, I will put a compressor with a sidechain input on the bass and trigger it with the kick drum. This will cause the bass part to duck down when the kick hits — to what extent is dependent on your compressor settings — thus helping the bass drum punch through the mix better.
FIG. 5: A series of plug-ins on the master bus, including several meters and a limiter, help di Pasquale keep his levels optimized.
After the bass line is solidified, I then begin to flesh out the other musical elements of the track. At this point, I am still working in a loop of eight to 16 bars and usually building around a section of the vocals. I try to experiment with as many musical ideas as I am hearing in my head and just let the vibe of the track take over.
I try not to get locked into one thing and usually construct at least two or three contrasting musical sections that I can use as different song parts when I do the arrangement. I listen constantly to make sure each musical part works well with the vocals.
Sound effects are another critical aspect of a modern remix, especially in the case of electronic club music. Effects are the finishing touches that create certain emotions and add contrast to various sections of the arrangement. This is where you can create that tension and release that works so well on the dancefloor. Sound effects will also work to create that powerful energy flow during transitions from one song section to another.
Some of the main types of sound effects I use are sweeps and fills. I have collected my favorites over time and have many EXS sampler instruments and sound libraries that I use. Sometimes I will just pull up a white-noise tone generator with a flanger and automate in and out with it throughout the mix (see Web Clip 3). One technique that is popular now is to use a sidechain to duck down such effects. For this, I will create a global sidechain track by duplicating the kick drum part and having it play constantly throughout the whole song. I will then set its output to Logic's No Output setting, which ensures that it only acts as a trigger for the elements I want to sidechain.
One of the most important parts of my workflow, as I turn the corner creatively and begin to transition into the arrangement and finish the mix, is the organization of my project. Up to this point, I work as quickly as possible and focus primarily on the creative musical aspects of the project. After getting the main musical sections added, I will spend a little time getting organized.
I always organize my projects in the same way so that I can find things and work quickly. I reorder and color-code the tracks and set up my subgroups. I like to arrange the tracks in the following order: I put the vocals at the top, followed by the drum tracks, the bass layers, the musical elements and the sound effects. I will create subgroups for each of those track categories by sending the outputs of the tracks within them to a corresponding aux track, which serves as subgroup fader. That's a key part of my workflow and speeds things up as I start to mix. I will put the subgroup aux channel strip for each group underneath the corresponding groups of tracks in Logic's Arrange page. This keep things organized and allows instant access to each subgroup (see Fig. 4 and Web Clip 4).
It Will Be Arranged
Up to this point, I have usually been working in sections and have arranged at least two contrasting sections, both around eight or 16 bars long. If I am doing a more commercial remix that will use a verse/chorus type of song form, these sections will allow me to follow that format. If I am doing a more underground mix with fewer vocals, then I may keep a more consistent groove and use breakdowns and sound effects for contrast.
As mentioned, for club mixes it is imperative to have DJ-friendly intro and outro sections. These typically comprise about a minute of drum elements. Once I get into the main section of the track, my arranging approach depends on what the music and vocals are doing and what type of energy I want to create in the early part of the song. Maybe I will want a groove-based vibe with the vocals coming in with more of a verse-type feel. Maybe I will create a breakdown after the intro and build some emotion and energy so when the beat drops back in, the impact really grabs some attention.
When it comes to the arrangement, there are limitless possibilities. For me, the breakdowns are probably the most important sections. These are the areas of the remix where you can really change things up, give a break to the drums and build things up to reintroduce the record with some impact and energy. When it comes to the dance market, this is the payoff, and it can really light things up on the dancefloor. I tend to stay away from the cliché drum rolls and prefer to get creative with tape delays, sound effects and vocal parts to do the job.
Mix and Master
After the arrangement is complete and everything is organized, it is time to move to the final phase: mixing and mastering. In the mix, one of the single most important things to do is pay attention to the master bus level. So many people do not do this, and at the same time, have the volume knob in the control room turned up really loud. Listening at high volume doesn't tell you the whole story, and in some cases it can be misleading as to how loud your mix really is. You may discover that after you bounce down your final mix and reference it somewhere else, it is way quieter than other commercially released songs. To keep that from happening, you need to constantly pay attention to the master bus level throughout the project. When finished, my mixes typically live somewhere between -10dB and -5dB RMS (see Fig. 5 and Web Clip 5).
My standardized processing chain on the master bus is as follows: multimeter, EQ, compressor (bypassed initially and not always used), multiband compressor (bypassed initially and not always used), Logic's Adaptive Limiter (set to -0.1 on the output to start), another multimeter and a simple level meter (set to show peak and RMS levels; see Fig. 5). The most important meters in the chain are the multimeter plug-ins because they show the pre- and post-processing levels. When I start to mix each subgroup, I make sure that I always have enough headroom and that I am not getting things too hot, too early.
After I set up the processing chain on the master bus, I go through each subgroup and begin mixing. I will always start with the drums and get them blended and mixed just right as a group. After that, I blend the bass layers until I have the sound I want and adjust the bass level relative to the drums. After that, I move to the other instrument parts and then the sound effects tracks.
Being organized and having my subgroups set up is the key here. Doing so allows me to get the individual tracks sitting right as a group, then I can use the main subgroup faders to blend the groups together. This workflow gives me total control over the individual tracks and the groups together. Lastly, I will blend the vocal group and work on getting its basic level right before adding any automation. I wait to add automation as long as possible because I don't want to commit to levels until I have a rough balance setup. The only time I will automate tracks before the final mix is when the automation is an integral part of the sound. For example, if I am using long delays or reverse-reverb effects, I will automate them early to get the timing just right. However, for the final vocal rides and other automation, I will hold off as long as possible and until I have the basic blends done with each subgroup.
I've found that the areas most likely to need automation are the drums during the intro and outro sections, the sound effects and the vocal rides. Because the intro and outro are simpler and more stripped down, the levels of certain elements may need to be automated to help them blend better during the quieter sections. Once the whole song is going and much more is happening musically, certain elements may need to be turned up in the mix. After the basic mix is done, a vocal ride throughout the whole track will help the vocals sit on top of the entire arrangement. They may need to be lowered during breakdowns and raised during peak sections of the track to remain on top.
The sound effects usually get the most automation. With so many sweeps, effects hits and sometimes a pink-noise track running throughout, careful automation is key to maximizing their impacts. I use a lot of delays as well on vocal parts and effects tracks, and I will automate the delay tails and feedback to be exactly how I want them.
Throughout the mixing process, I constantly refer to the master bus meters to ensure I have enough headroom. Making up gain at the end is easy, but if you mix too loud and too fast and run out of room, by the time you have automation and a lot happening, it can be tedious and time consuming to go back and lower things.
When metering, it is important to remember that you perceive the loudness of a mix based on the RMS (or average) loudness. The Adaptive Limiter in Logic allows me to ultimately make up the gain I need and achieve a loud RMS level without overloading the master bus. I will rarely touch the master fader, preferring to leave it at unity gain (0.0 dB). Because the inserts are pre-fader, any make-up gain I need can be added on the output of the Adaptive Limiter.
At the end of the day, it's important to remember that the main purpose of all this studio technology is to help you transform your musical ideas into finished recordings. Yes, the production process is free-flowing and ever changing, but having a consistent workflow and a personal step-by-step process allows you to capture and harness that creative energy and get the results you want.
Vincent di Pasquale is a Miami-based producer/remixer who works out of his project studio (vcdstudios.com). To learn more about remixing, be sure to check out The Art of the Remix, a comprehensive 7.5-hour interactive remixing course taught by di Pasquale available now atfaderpro.com.