The Jam Corner

Intro To Digital Recording

The purpose of this article is to go over, in broad terms, the considerations that go into recording, mixing, and mastering using modern digital recording tools. I'm going to talk about GNX onboard recorder, stand-alone digital recorders, and software recorders such as Sonar (and other Spawn of Cakewalk). While I am not a professional studio engineer by any stretch of imagination, hopefully my basic level of knowledge will be helpful to beginners in getting out there and producing decent-sounding CDs.

Introduction to digital recording, mixing, and mastering

I remember first starting out with electronic music. Initially, I was in a band with friends who had some form of a studio setup (usually Roland workstations at the time), and picked up quite a few routing basics� of the digital recording realm. In high school, I took a class on electronic music, where we learned to use an analog 4-track mixer (Tascams). Today's guitarists (and other musicians) demand great facility in recording and mixing equipment. This is because the cost of technology needed to perform high-quality digital recording has shrunk, and more consumer-level products are available for pro-level studio results. One of these products is definitely the GNX series. With the on-board 8-track recorders on GNX3 and 4, and Cakewalk software included with all of the recent GNX series, it's impossible to think back to the days where several four-track mixers were bounced to mono tracks, panned hard L/R, and printed on a record. However, these technological advances come at the cost of entry-level users being utterly confused on how to create a quality mix.

The Basics

Once again, recording is generally a three-step process. Here is our outline:

• Recording this is where we decide what musical parts stand to have a spot in the final mix. During recording, we establish a basic framework for the song. Also, one major goal of recording is to get a good representation on record of the actual sound you're producing. For many casual recording engineers, this is also the time to explore ideas and create rough mixes to help spur the imagination.
• Mixing in this step, we determine what elements from step 1 will be present in the final mix; where in the mix the these will appear; and how relatively loud various parts are going to be. Mixing is the step that determines whether the final product is going to be stellar, good, or simply mediocre (or worse).
• Mastering this is an optional step, generally performed by pro studio engineers. While I'm not going to delve into micro-mastering techniques such as click removal, I will mention a few things about compression and equalization that pertain to amateur recording practices.

Step 1: Recording

A few essential consideration that go into recording a piece (especially if yourre working with other people) pertain to deciding on common assumptions, primarily of musical and tonal nature. Here are a few things to think about:

• How fast is the song?
• What is the meter, and does it change (4/4 everywhere except one measure of 7/8 for example)?
• What key is the song in, and what tuning does everyone use? (e.g., A=440, A=432, etc)
• What acoustic instruments do you have? (for purposes of this article, any instrument that requires a microphone, including vocals and mic'ed amps, is considered acoustic)
• What overall sound are you trying to achieve? This will dictate both the performance style (picking technique, singing technique, drumming technique, and so forth) and the tones and effects that you use.
• Besides the main parts (bass, rhythm guitar, percussion), what overdubs will you need? In other words, if you want a fill in-between sections, backing vocals, or some synthesizer effects, you'll want to leave room (both physical and acoustic) for these to take place. You'll also need to decide on how the tone of the overdubs will differ from main parts.

The first step is deciding the tempo and the meter. If you're using a stand-alone digital recorder (e.g., Korg D-series) or software (such as Cakewalk), you'll be able to pre-arrange the click track or drum machine before you ever start recording. Doing so in the beginning will make mixing and editing much simpler in the end. Also, if you're using software synthesizers, software recorders generally let you dictate the key changes that occur throughout the song. TAKE YOUR TIME TO PROGRAM ALL THIS before you start recording, or you'll [u]regret it later.[/u] Traditionally, the first thing that's recorded is the drum track. This is the most difficult thing to fix after you've begun recording if you're using acoustic drums, so make sure you get a good take of the drum part before going on to other parts. If you're using a drum machine or a click track, I still recommend either pre-programming it or recording it first, so that you can base the rest of the song on the feel of the drum track. Much literature is available for recording acoustic drums, so I'm not going to go too deeply into it. My band uses a Korg D1600 digital recorder, which allows us to record up to 8 mono tracks simultaneously. Various setups we've tried for recording drums include:

• One mic on snare and one on kick; one large-diaphragm condenser overhead;
• One mic on snare and one on kick; two small-diaphragm condensers overhead;
• One mic on each snare, kick, hi-hat, rack-1, rack-2, floor tom; two small-diaphragm condensers overhead.

The main things to consider when placing microphones on drums are:

• The microphones are not isolated; if your snare mic is pointed too much towards hi-hat, you will get significant bleed.
• Having two overheads, one covering primarily left side of the kit and the other covering the right, allow for a good deal of stereo separation between the sides. When you mix later, things like tom fills will whoosh by from left to right, creating really nice, professional-sounding mixes.
• Putting separate microphones on the toms allows to pan them L-C-R, creating an even greater degree of stereo.
• Mic'ing the hi-hat separately aids during mixing, because you can separate snare from hi-hat much easier. More on this in the mixing section.

The recording levels, for drum mics and any input in general, use this important guideline: Nothing recorded digitally should ever go above -0.1dB at the input. Make sure you audition the LOUDEST parts (fast drum parts, heavy rhythm guitar, slap bass are all good candidates for loudest parts), and set the levels accordingly. This presents the issue of softer parts not being loud enough; we'll discuss how to address this issue in the Mixing part of this article. Once you have the drums, usually bass comes next. With bass, it's important to capture a good, clear tone. It's very easy to muddy up a nice bass part with too much gain, too much low-mid, or too many effects. Remember, you can add both mud and effects later, but it's very difficult to get rid of them. If you must record a processed track (meaning, not just amp/amp modeling, but also effects), it's a good idea to record a dry track alongside of it again, you'll be thankful you did later, when you go to mix your masterpiece. Once again I'll use my band as an example. Using the D1600 recorder, the following setup is used: Notice how in this setup you get a variety of sound sources to mix later: mic'ed bass amp (good to have if you have a nice amp), bass amp model + effects (from BNX3), and clean unprocessed track (to reamp later, or to mix with the other two if you have a good-sound bass). You may not end up using all the tracks, but at least you'll have that option down the line. Just make sure that you're not wasting time/space by recording an identical signal for example, if your DI box pass-through output is recorded alongside the BNX/GNX dry digi-out, you've wasted time and space. We recorded guitar in a very similar fashion the one exception is that there was no mic'ed amp. The guitar was plugged directly into GNX3 and S/PDIF was used to get a dry track. Later I ended up using DX plug-ins to reamp the dry track, which was cool for stereo effects on solos and certain backing rhythm parts. For most people (especially using GNX4 or GNX3000, Line 6 PODXT, or Boss GT-Pro) this will be the most likely setup because most stand-alone recorders don't allow real-time USB audio. Of course, if you're using something like Atomic Amps (those ones designed specifically as tube warmers for modelers), you can use the extensive output selection on GNX4 and 3k (as well as some of the Pro series) to both record a direct signal and a mic'ed amp… you decide on which possibilities suit you best. When recording any acoustic instrument (especially vocals, but also drums) it's imperative to achieve the most sound isolation you can get. Record in a closet if you must. Carpet your floors and put up dampeners on walls and ceilings. Just make sure that the only thing coming through the mic is what you want to be in the final mix. Only pro mastering has a chance to remove miscellaneous noises that accidentally get onto the recording this can cost $7-10 per track minute to fix, so take care to get it right from the get-go. You might be thinking, How the heck can I fit all these multiple tracks into my 16-track recorder? The Korg D-1600 allows up to 8 virtual tracks per each of the 16 track channels. My band recorded all drums on Vtrack D, all bass on Vtrack B, all guitar on Vtrack G… many times, we also had room to take down some alternate takes. Then, we imported all those tracks directly into Sonar which has unlimited tracks available. We then labeled them and picked out which ones stay, which ones go, and which ones get turned down this was done during mixing. So if you have capabilities of this nature (Sonar is a great one), you can let loose and record everything and more. If you're limited in number of tracks, you'll have to decide what to keep and what to cut during the recording process planning and common assumptions mentioned above come into play a great deal if you have severe limitations. Also remember that acoustic instruments will all have a bit of room ambience to them. 99% of the time, you'll want to record them in the same room. This will let you add reverb/ambience effects later without the result coming out sounding… weird. So the [u]key points[/u] for recording are:

• PLAN EVERYTHING AHEAD OF TIME!
• Get good levels.
• Get multiple takes and versions (layers).
• Capture the stereo (but keep it in separate L/R tracks).
• Go easy on effects and dynamic processing (compression, etc).

Step 2: Mixing

When it comes to mixing, things become a little more tedious. A few nights ago, after exporting all the tracks for one of our songs from the stand-alone recorder, I was faced with 108 tracks. ONE HUNDRED AND EIGHT of them! To add to the troubles, the recorder exports all tracks in the 8x3 DOS file name convention. This meant I had to go back and audition each track, label it, etc etc. Once you're done mixing, the result may sound vastly different from your scrap tracks or preliminary mixes. How do you cope? Well, if you're recording directly into PC, you're in luck: the ability to label everything is second to none when using computer software recorders. If you're dealing with the scenario similar to mine, you might have to spend two-three days just going through and identifying which track is what. Taking notes during recording alleviates these problems, and coupled with a standard labeling/numbering procedure on outboard recorders eliminates the guesswork and headaches. Once you know what all your tracks are (Sonar lets you arrange them into track folders), it's important to decide what kinds of subgroupings you're working with. A nice feature in Cakewalk software (and others, including many stand-alone recorders) is the ability to create BUS tracks, to which you send the output of certain tracks to be mixed a certain way. My band has used many different configuration of using buses to pre-mix in real-time. These will be applicable depending on your song's need, the complexity/number of recorded tracks, and the amount of system resources your mixing software has available. It may help to look at a diagram. Here are some examples of buses to be used:

• Instrument groups for example, all guitar tracks go to Guitar bus and run through the same compression setting. This helps to adjust panning and equalization in the final mix.
• Acoustic groups for example, all drums, alternate percussion, and vocals have a Send to AcousticReverb bus, where they're patched through Lexicon Pantheon. The same acoustic instruments each have a main out to the instrument group (so drums will go to drum group for example); the Acoustic Reverb is then mixed in with the instrument groups at the mains. You may also consider using a master reverb instead, where you send things like keyboards and amp modeler guitar/bass (not just the acoustic instruments). Adjusting the amount of send to a master reverb bus will play an important role in the overall mix more about this in a minute.
• Drums compression mixing uncompressed drums with compressed drums helps avoid the issue of having some parts too soft. You want to make sure you don't send too much cymbal to this bus (level of each send is adjustable). This technique really brings out and solidifies acoustic drum parts.
• Lead buses routing lead tracks (guitar solo, bass solo, etc) around master dynamic effects will allow you to make them stand out without losing the rest of the mix.

Notice in the diagram how certain buses are mixed at the master, while others are mixed at the mains. This is because, often enough, you'll want something like lead guitar to really jump out. If that's the case, you'll probably add a lead guitar bus with the relevant tracks sent there. On this bus, you'll adjust the EQ and compression for the guitar, perhaps add some reverb that differs from the acoustic reverb to make it stand out. Master bus generally has some overall compression and pre-mastering effects on it (Izotope Ozone and BBE Sonic Maximizer are both good choices), so bypassing it for lead guitar brings it up front and center. Speaking of front and center, our next mixing step is determining how much of each instrument we will here, where, and how. Respectively, these correspond to level, pan, and EQ. These are probably the three most important elements to a good mix. First, we'll deal with placement. This is fairly easy in theory: so long as all (or most) of your recorded tracks are mono (you can import stereo WAV files as split mono tracks in Cakewalk), adjust the pan setting FOR THE BUS SEND based on the desired placement of your track (i.e, after you've performed an "insert send" operation, adjust not the main panning for the track, but the one for that particular send). Here are some guidelines:

• Always leave bass and lead vocals in the center. If you have more than one bass track, try to put them slightly off to the sides, but definitely no more than 25% either left or right.
• If you have multiple guitar tracks, put the off to the sides about 60% each is generally good.
• Set backing vocals off-center; we'll also use some EQ tricks to make the fit better into the mix.
• In the general case, you want to position all tracks at different pan settings, so that they stay out of each other's way when you adjust levels.

A note about split stereo tracks: by adjusting the panning for each channel of a stereo track separately, you can define both the positioning and the perceived width of the sound source. One of the pan values is negative (usually Left), so keep that in mind. The formula is as follows: Center = Left+Right Your perceived center of the sound source (for example, vocal track) will be the Center. Your perceived width will be from Left to Right. This is nifty trick if you really want to narrow down the placement of your stereo tracks. Next, let's take care of levels. Once you're done deciding on tracks to use for the mix, and the bus configuration, follow these steps:

• Put the fader for Master channel at 0dB (unity gain). Turn off effect buses (such as reverbs).
• Put all other faders at 0 assuming you've auditioned all tracks and assigned them all to their respective buses, you can use bus faders to save some time in this step just be sure that the bus sends are panned properly coming from the tracks.
• Starting with drums, followed by bass, vocals, and guitar, bring up each track (or bus) fader until you hear a decent amount of the instrument. Note: when doing this with drums, having made appropriate compression settings on the drum compression bus is necessary, if you're using this technique.
• Keep making adjustments until you get a good overall mix level, and each instrument is as loud as you'd like it to be with respect to the others. Keep in mind, however, that this process alone will not make a perfect mix just settle for good levels for now.

Now that we have our panning (placement in the stereo field) and levels (relative volumes), our mix is starting to come together. In software like Cakewalk, as well as on certain stand-alone recorders/mixers, you'll be able to automate level and pan settings. This can create motion and add some nifty effects to your song. Finally, we need to adjust EQ settings for each bus. For this purpose, many stand-alone recorders offer per-track equalizers. Cakewalk's Sonar PE also has per-track EQ, that's very easy to adjust in the Console view. Equalizer settings are very important during mixing. This stuff can really make or break your final mix. The process consists of two primary steps: (1) identifying which tracks have similar frequency ranges, and (2) equalizing them to fit together. This is key when you have tracks with similar range (such as guitar and vocals) close together in terms of their panning position. The Internet is a wealth of information when it comes to EQ settings; many sites offer characteristic frequency response charts, overtone frequencies and levels for different instruments, etc. There are also plug-ins available (some as freeware) that emulate real-time frequency analyzers. These are all useful tools to figure out which frequency range(s) a particular track's tone is coming from. If none of these are available, just solo each track/bus, and play around with a large boost setting at various frequencies with patience, you'll eventually find exactly what makes your tone. Keep in mind that, after adding equalization, many of the tracks that once sounded full and awesome will sound puny and weak. That's perfectly acceptable by themselves they may not be much, but in the mix they will be clear and complement other instruments. Here are some broad guidelines for EQ:

• For all instruments: lower EQ gain on anything below 50Hz. This range will be restored during mastering, if necessary.
• For bass guitar, give a light boost between 60Hz and 200Hz. Cut out the kick drum in this range.
• For kick drum, give a light boost between 100Hz and 300Hz. Cut out the bass in this range
• For toms, you may want to boost around 250-500Hz. This is probably not necessary, but will depend on your style and overall sound you're trying to achieve. Make sure you cut out some of the kick drum in this range. A boost around 1.2kHz will add more kick to the kick drum.
• For snare drum, give a light boost between 500Hz and 700Hz. Cut kick drum in this range. Also a boost around 2.5kHz can help get more of a snap.
• Rhythm guitar is a difficult one due to the many variations and a fairly rich overtonal harmonic spectrum, you'll have to try various settings to see what causes the best sound.
• For cymbals, you'll want to shelve off anything about 8kHz (and especially above 10kHz).
• Lead vocals will largely depend on the vocalist. Generally, just above 1kHz is a good frequency center to address. You'll want to cut backing vocals in this range.
• Lead guitar most likely somewhere between 2kHz and 3kHz will require additional boost to make it stand out. Make sure there are no lead guitar parts clashing with vocals if there are, you'll have choices to make.
• Backing vocals a boost around 3kHz to 5kHz (again, depends largely on the vocalist) helps. By accenting backing vocals at higher frequencies than lead vocals, you'll make them more distant and they'll fit better in the overall mix.

Obviously, of the three key elements, this one is the most involved. It requires practice and expertise much of which I myself lack. The above suggestions are based on my experience with my band your experience may guide you along a different path. Some finishing touches, such as adjusting the amount of send to mastering reverb (or acoustic reverb), will either bring the track forward in the mix or back. More reverb generally equals greater perceived distance of the sound source; the inverse is true. Play around with these settings, and you can get a quick studio-sounding mix by simply getting the right reverb settings. Here's our summary for mixing:

• Always begin by labeling all your tracks and deciding which you want to keep, which you want to delete, and which you want to alter;
• Set up your bus/output configuration before you start it'll be all that much more effort to redo the whole mix;
• Make adjustments to the following (in this order): panning, levels, EQ as necessary;
• Use light reverb effects to achieve back-to-front placement of instruments, if you feel so compelled. You might add a master-reverb-send to each track, and then use automation tools (in Cakewalk, they're called envelopes) to increase or decrease the amount of send for each track or bus to the master reverb.
• A good idea for getting a decent-sounding mix is not blasting it when mixing - playback should be at conservative levels.

Step 3: Mastering

When your final mix is ready, you'll probably want to do some mastering. The best way to do this, in my experience, is to export a WAV file of the entire mix, but setting the Master Bus output at a low level (maybe try for -10dB). This will leave you enough room to bring up the good parts of the mix, and turning down the bad ones. Next, load the WAV file into a new Cakewalk project (or if you're using a stand-alone recorder, into a new song). Now, you are free to make some useful tweaks to the final version:

• Adjust the amount of overall reverb, if you haven't done so during mixing (experiment with both, sometimes one or the other works better)
• Add overall compression to the mix. Boosting around -70db through -40db ranges can make the mix appear louder and more full, but be careful not to compress/boost noise (also beginning to appear between -70 and -40dB).
• Use a multi-band compressor (Sonar PE includes a plug-in for that) to clean up muddiness or bring out frequency ranges that are too soft.
• Use EQ and mastering effects (again, Sonic Maximizer comes to mind) to shape the end result of your mix.

The basic idea behind mastering is (1) make sure everything is audible that needs to be, (2) even out/flatten the frequency response, and (3) bring everything up as much as you can. However, it takes a lot of trial-and-error for a newbie to get a final mix that's both as loud as commercial recordings and sounds good. You may want to try some factory presets on your mastering plug-ins and go from there.

Conclusions

I know that I've mentioned a lot of various techniques and tools to use, and many people may not have them available. However, my intention here is simply to outline what tools are available and how they can be used. Not all of them are necessary in fact, none of them are really, truly necessary. Nonetheless, it behooves us to use all the latest technology that we can get our hands on to get the most out of our recording sessions. Check out this article for some insight into the kind of equipment you might use to record your songs. One final note: the single best source of information on all these effects will be you guessed it! THE MANUAL. For plug-ins and other software, help files are available more often than not. For hardware, you can usually go to the manufacturer's web site and download a PDF version. Make sure you read this stuff otherwise, you'll be pretty lost.

2157 Views | 0 Comments | Rating: (0 rates)