STUDTRIX.ASC (c) 1990, Living Skill Media First posted Feb. 1990 Publishers' permission to upload this file to electronic bulletin board systems and public or private electronic databases is NOT required. Spread it around. ----------------------------------------------------------------- Brought to you by Steve Winter, Living Skill Media and Texfiles'R'Us. "If you can almost stand to read it, we probably wrote it." ---------------------------------------------------------------- This is the second in a collection of tip sheets for home recordists, MIDI enthusiasts, songwriters and working musicians. You get 'em free because no magazine on earth was prepared to pay for them, and because with Steve's sales record as an author, not even we would risk publishing them in book or manual form. And it's a continuing collection because we were so pleased with the response to REALDRUM.TXT that were able to guilt Steve into transcribing more of his Little Black Book for the modeming MIDI world. As with all Living Skill/Textfiles'R'Us textfiles, this file is not public domain. It may be freely exchanged and posted, complete and with no changes in text content, in magnetic or optical form only and may not be copied in printed form except in the case of single-copy printing for personal use only or in the case of brief quotations embodied in reviews or educational texts where the source and author are credited. Feel free to upload this to any BBS or net that doesn't currently have it in its library. Others might benefit, and we certainly wouldn't mind the exposure! Details on the complete collection of Living Skill's MIDI/music-related textfiles will be released with the next posting. CORRECTION TO REALDRUM.TXT: It was erroneously reported in the REALDRUM.TXT posting that an upgrade to the Ensoniq VFX was available which permitted the addition of the drum and percussion PCM's in the VFX-SD to the original VFX. This resulted from misinformation provided by Ensoniq's Canadian distributor, and Dave McMahon of Ensoniq has stated officially that there is no current or planned method for upgrading the PCM ROMs on the VFX. The new VFX-SD drum/percussion PCMs are available ONLY in the VFX-SD, and neither version of the VFX is planned to be released as a tone module. (Considering the quality of Ensoniq's percussion PCMs, that's a real pity. Enough of a pity that I bit the bullet on my VFX and traded it for a Proteus. -SW). We apologize to any users or dealers who may have been inconvenienced by this error. It was also reported that the VFX will allow keyboard zoning of individual percussion instruments in addition to level scaling of more than two voices per instrument. In fact, the VFX' architecture will allow this, but at the expense of layering of sounds on the keyboard. A maximum of two DISCREET percussion voices may be provided in a single-patch split. The addition of more voices will result in stacking of tones, an unwanted result when using the level-scaling techniques described in REALDRUM.TXT. ----------------------------------------------------------------- ************************************************* * STUDIO TRICKS #1: * * SPICES AND SEASONINGS FOR THE HOME STUDIO: * * GETTING THE MOST FROM DIGITAL EFFECTS * * by Steve Winter * ************************************************* Having been called in on a number of pre-pro and fix-mixes, it continually surprises me how little people tend to explore the depths and breadths of their equipment. Most people with home 4- and 8-track rigs tend to have not one, but SEVERAL digital effects devices and/or synthesizers that can do a lot more than the manuals let on. Here are a few of my special tricks, tips and techniques for tweaking fresh new sounds and unique treatments from your existing gear. Some of them might seem painfully obvious to you, but keep in mind that the ideas included here seemed intuitive to me too, until I discovered how many people DIDN'T know about them! Here's hoping others in the RT will add their favorite ideas and expand all our horizons! A number of these tricks require the use of more than one multi-effect unit, but don't view that as the problem it seems to be. Most of us own one, perhaps two at the most, but come time to mix our precious project, we can usually count on friends and neighbors to come through with five or six others for a day or two. These devices have become so cheap and plentiful there's no reason why you can't reasonably expect to lay your hands on two DDL's a pair of reverbs, an SPX-90, a DSP-128 and an SGE if you really need them. And believe me, if you get serious about mixing, you might find that's still not enough to mix an 8-track demo on many occasions. (If it is now, it won't be by the time you finish reading this!) Honestly, in this day and age your biggest mixdown hassle should be a shortage of cables and auxiliary inputs, not processors. DIGITAL DELAYS Whether you use an 8-bit stomp-box delay for your Fostex X-15 or a Lexicon Prime Time for your Otari, you should find a few of these tips helpful. Since The Edge made the heavy-echo sound almost de rigueur for the new "mutant blues", everybody and their brother has been using perfectly-timed repeats which fall directly on the beat to add thickness to sparse arrangements. But there's more to this than meets the ear. Calculating Delay Times First of all, how do you calculate how long a delay to use? If you know the formula for calculating beat length in milliseconds, you should never have to fight your delay line to find the proper echo length. (Believe it or not, I've seen guys waste $10 worth of studio time trying to find a single delay time that fell on the beat!) For a standard quarter-note delay (the benchmark), divide 60 by the tempo in beats-per-minute (e.g. for a 120 bpm quarter note, you divide 60 by 120, giving a total of .5; the beat length is therefore .5 seconds or 500 milliseconds.) For three-quarter time, divide 60 by the tempo and add one-third of the result. (e.g. for a 120 bpm 3/4 quarter note, you take the .5 figure and divide it by three, giving .16666. Add this to .5 and you get .66666, or 666 milliseconds.) If you use it for nothing else, this single application will make a RAM-resident or D/A calculator pay its byte count on your sequencer's boot disk. Dragging and Pushing with a DDL Now you've got an exact millisecond count of the delay length for a single beat. But that's not enough for some applications. If you're doing heavily rhythm-dependent music like funk or hip- hop, you'll want to use that exact millisecond count, but if you're playing metal or kick-ass, you'll want to pack every bit of urgency into the music that you can, and dead-perfect delays are bo-ring! Echos aren't used that often in commercial soundtracks, but when they are, adding a drag or a push to it can accentuate the emotional tone of the video and voice-over. To add a "rushed" or "pushed" feel, SUBTRACT a few milliseconds from the delay length and each repeat will sound like it's rushing the beat slightly. Unfortunately, if you're using a live drummer who doesn't have great discipline, it might rush your drummer, so it's tricky in any application where you're recording live off the floor with a WYHIWYG (What You Hear Is What You Get) headphone mix. Best to add the "push" at mixdown and keep the live cue-mix-only delay dead on the beat, even tied to the tempo map if necessary. You can do the same thing on blues or ballads if you want to drag the beat a bit, except that you ADD a few milliseconds to the delay time. The more you alter the delay time, the harsher the effect. One caution, though. If you're using severe delay feedback, don't vary the delay time more than a few milliseconds off the beat or the trailing echoes are going to sound so far off the beat that it may seem to the listener like a mistake was made. Non-Standard Delay Times A number of players have added a whole new dimension to their style by performing an additional calculation on their delay lengths. I've used it myself to great effect, but it requires some experimentation. The trick is to calculate the delay time based on a time signature different from that of the music. For example, I recently needed a "misleading" 1/8-note echo on a 6/8 boogie at 120 bpm, so I used a delay length of 250ms, when the actual delay length of a single beat was 333ms. In other words, I interpolated a 4/4 feel on a 6/8 rhythm. Conversely, on a 4/4 rhythm, you'd program the delay to triplets instead...using a 333ms or 167ms delay instead of a 250ms on-the-beat delay. For lead lines, effects and spacey rhythms, it can give superb results, but you must practice with it first because it can play hell with your sense of timing. One of The Edge's BEST tricks involves not actually changing the signature, but dividing the timing on a delay. On arpeggiated lines, especially those with a bouncing feel, try a 3/16-length delay (at 120bpm, this would be 250ms [an eighth-note] plus 125ms [a 16th-note], or 375ms in total). This or a 3/8-length delay can create some spectacular results on lead breaks. One of the earliest uses of this technique was by Brian May on Queen's "Brighton Rock". Jean-Luc Ponty and Eddie Van Halen have also made excellent use of this technique. A 3/16-delay on a flute line literally saved one of my favorite compositions from being chopped up for my "riffs" file. MIDI-fied Delays and Custom Delay Tracking Many of the new multi-effectors now provide real-time MIDI control over several parameters, including the depth of any of several parameters which can be set from your synth or sequencer, as well as MIDI program change and MIDI effect on/off. But there are few reasonably-priced units that have MIDI-controlled on-off or mix for input. One of them is my trusty Alesis Quadraverb, one of the new- generation processors that offers mappable real-time control over virtually every effect parameter imaginable. This sort of mapping control has been a godsend for me, because I frequently work alone on a setup that doesn't usually leave me any empty tape tracks. But it doesn't have input on/off, and that's a problem for me. One of my favorite tricks, especially on "new rock", rap and hard funk, is using delay on the downbeat only. This means if a one-bar, quarter-note pattern is played, the note played on the 1 is repeated once, twice or three times through the bar and the other three notes are uneffected. Here's a graphic representation of what it sounds like. Direct: ONE!\ Two! Three! Four! Delay: \.One! ...one ...(one) It's a killer effect, but if you try to use an Effect Depth=0 message from your sequencer, the only option you get is the single repeat, because if you want to have the one note echo on all quarter-notes, the notes played on the 2 and 3 will start to intrude on the delay signal by the time you reach the end of the bar. From MIDI control, the same effect would look like this: Direct: ONE!\ Two!\ Three!\ Four! Delay: \ One!.\..one....\..(one) \ two.....\.(two) \three Not good. You wind up with echo signals you don't want. The only way around this is to either bounce the direct signal over to another track, have someone on hand to kill the delay send while you record, or to dedicate an extra track to the delayed signal. If your sends can be switched on and off with a pushbutton, here's how you'd track this downbeat-only delay effect as shown above (it cannot be done via MIDI control with the QuadraVerb, just to settle that question): Start the track with the delay set to a 1/4-note length and the delay send ON. As soon as the note finishes decaying, or JUST before the 2 (as close as you can get; you might have to do a LOT of punching to perfect this), kick the delay send OFF so that the next note doesn't get into the DDL. As soon as the 4 decays, or just before the 1, kick the switch ON again, and OFF again just before the 2. This effectively shuts all notes except the 1 out of your DDL, and that's the ONLY way to filter out the other notes regardless of whether you do it manually or with a MIDI mixer. If you have an extra pair of hands with you, you can rehearse your engineer to do these "punches" while the track is going down, and if you don't have the option to switch the delay input out with a switch, you can dial it out quickly instead, or "slap" it out if you're lucky enough to have a linear fader to do the job. Filter-Echoing MIDI Notes This trick is also useful if you just want to delay certain parts of a lead line, such as a tremolo dump or an arpeggio, and leave the rest of the music unaffected. Try it...you'll LOVE it! But fader-slapping really is the hard way to go if you're trying to achieve this effect from a tone module or drum machine. Better to insert notes into the sequence, and if you want a volume decay on the echoes, reduce the velocity of the "delayed" notes in software. It might not be quite the effect you need, especially if the patch in question has its overall timbre tied to velocity, but it's a quick, permanent, tapeless way to do it. The other option you have if you can spare the MIDI channels is to copy the patch to one or several other patches and simply reduce the overall volume of each successive patch to simulate the volume reductions you get on an echo with feedback. Then you can enter the notes you want to hear as echoes on that patch's channel and preserve the true timbre of the patch in the echoes. It's time-consuming, especially if you've got a lot of echo-with- feedback happening on one or more patches, but it's the best solution to a tough problem. MODULATED DELAY TRICKS Wide-field Mono-proof Chorusing with a DDL Another delay trick involves chorusing. I learned this several years ago while working with what was then a very sophisticated multitap delay unit, but anyone with a chorus pedal and a DDL can use it. It's especially effective for pads, rhythm guitars and anything else for which you'd like an ultra-rich, wide stereo field, and if you're using an analog stereo chorus, it will prevent that ugly who-stole-the-chorus phase-cancellation you get when you hear the track back on a mono system. I'll use rhythm guitar as an example. First, send your guitar or other instrument into the mixer and pan it dead center. Then take an effect or auxiliary output (NOT the direct output, unless you can keep your input signal as- is while using it) into the chorus box or pedal. On cheaper units, the output will probably be set up so that the only difference between left and right channels will be that the oscillator waves are 180 degrees out of phase from each other. From here you can route your signal one of two ways. I recommend sending one chorus output directly back to the auxiliary return and sending the other into a DDL and THEN into the other aux. return. I recommend it this way since most "stereo" digital effects actually bung your stereo input back down to mono again before processing it, but if you don't mind this (and you bloody well SHOULD mind it), you can send both chorus outs into the DDL. All you do now is add a single 10-millisecond delay on of the two chorused signal and kill the direct signal before sending it back to your mixer, so that one channel is still out of phase but a few milliseconds behind the other. You can add 20 to 30% feedback to the delay if you like, but this will add extra "weight" to the chorusing on one of the stereo channels and has to be counterbalanced by some other element of the mix. Best to set feedback to zero, and voila, mono-proof wide stereo chorusing! Mind you, if there's no risk of your tune ever being played back in mono (i.e. it will never be played on FM radio or TV or used on a mono ghetto blaster or house sound system), you don't have to worry about this anyway. Emulating a Tape Echo with a Digital Delay A lot of people still like the sound of tape echoes, but owning that huge 20-pound behemoth that needs more maintenance than most 10-year-old Toyotas can be a hassle. Fortunately, there are ways to emulate a tape echo with digital gear you probably already have. The first thing you need is a DDL that will add modulation to long delay times, not just to short chorus/flange delays. Many of the newer multi-effect units don't have this option. All you really need to do is add a bit of low-depth, long-wave oscillation to the delay and dirty up the sound a bit as the echo decays. Run the signal into the DDL and bring your direct signal back into the board. Take the effected signal and send it into the noise gate (an old Hush unit is perfect, but any frequency-dependent gate is fine; even a Rockman Sustainor will do the job) and set the gate up so that it trims just a touch more treble off the signal with each echo. Then send the signal into the cheap preamp and back into the board. You should get everything the tape echo gives you...the treble loss, the wobble, the hiss...everything except for the timbral distortion you get on an infinitely-decaying echo. In order to emulate this timbral smearing, you'll need somewhat more specialized gear. You need a device such as the ProVerb, MidiFex or MidiVerb II to perform the delay and another processor such as a REX-50 or a SPX-90 to execute initial reflections, and the effect takes place entirely in an auxiliary loop, meaning you'll have to have some way to preserve your dry, uneffected signal. Notebooks ready...this is a tricky patch! Take your aux send into the right input of your delay device and set the delay device to 0% direct (100% effect). Take the delay output into the right reverb input and set that device for between 30% and 50% initial reflection, no direct signal, no predelay, no reverb. Take the reverb output and bung it into the LEFT input of your delay, take the LEFT output of the delay into, you guessed it, the left input of the reverb, and back to your auxiliary receive. Here's a graphic representation, and I hope you can view this with your CPU's ASCII. ___________________ _______________ /->Rt Rt>\ Mixer Aux Send>-----/ INPUT **DDL** OUT \ --------------- ____>Left Left> \ / ------------------- \ \ / ______________________\__\ \ / \ ------------------------\ \ \ __________________ / \ \>Rt Rt> \ ______________ INPUT**REVERB**OUT \ Mixer Aux Rcv.<-\ />Left Left> / -------------- \ / ------------------ \ / \ \_____________________\___/ \________________________/ Trust me...it'll work, and if your gear is less than 4 years old, it won't hurt a thing if you wind up feeding everything back like crazy. Now you're going to have to fiddle quite a bit with the reflection density, high (treble) rolloff on the reverb or delay (an ambient bandpass delay on your ProVerb or MidiFex would be ideal; it will add even more authenticity), the ratio of dry-delay to reflected-delay, and the type of early reflection algorithm. You'll also have to adjust the overall output level of the reverb until the echo sounds realistic. But you should find once you're done that the early reflections emulate that characteristic "smearing" effect you get on long tape echoes, and the EQing done in the reverb amplifies itself each time it passes through the delay so that each repeat sounds thinner and thinner. Flange and Chorus on Bass Lines So what's the big deal about these new "bass choruses" and "high band flangers"? Nothing, really. They're plain old ordinary bucket-brigade modulation delays, except that they add a circuit that filters out all the low frequencies from the signal before it gets sent through the delay processor. You don't need a separate device to do the job properly once you understand how it's done. The basic concept stems from the observation that the bass' fundamental function (pardon the pun...or don't; see what I care) is to carry the bottom. If it's run through a flanger or chorus and is allowed to wobble at the lower, fundamental frequencies, flanging tends to steal bottom-end harmonics through phase cancellation, especially at delay times between 1.5 and 35 milliseconds; and chorusing makes the fundamental sound just plain rubbery. All you have to do to get a clean, solid fundamental WITHOUT losing the characteristic swoosh of the flange or thickening of the chorus is to EQ the effect signal before mixing it with the dry instrument signal. Just dump the bass or other instrument into the mixer as usual, send it into the delay for processing, and bring back the chorused or flanged signal ONLY, NOT the direct signal, into another channel of the mixer. Set up the processed channel so its EQ is identical to the direct bass channel, with one exception: turn ALL the low EQ on the effect channel OFF! That's right, as hard to the left as it will go. This brings the low-frequency effect level down to a volume so low that it won't bother the bass' fundamental tones, and still preserves the effect sound in the midrange and treble bands where you want to hear it. And high-band flanging? Not much different, really. As near as I can tell, Boss' HF-2 flanger is the same as its brother except that it has delay times from 0.5 to 7 ms (regular pedal flangers go from 1 to about 16 ms) and a high-pass (bass-cutting) filter. I shouldn't have to tell you how to duplicate this one. But why stop at high-pass filtering? If you've got a midrange control, or parametric or semi-parametric "sweep" EQ on your mixer, you can try selective EQing on all kinds of effects for an infinite variety of "filtered flanges", top-end choruses, ambient delays, you name it. Then just mix the effected signal in with the direct signal at an appropriate level. No fuss, no muss, and most important, no difficulty on vocals trying to stay in tune with a wobbly fundamental. Adding Depth to Background Vocals More and more home recordists are relying on multiple-repeat digital delays to thicken up background vocals and save on tape tracks. Personally, if I really need a choir of a dozen voices on a backing track, I think it sounds better to record and bounce, record and bounce, twelve bounces deep if necessary, then spread one or two tracks in stereo with outboard gear and damn the distortion. But if you insist creating massed choirs from just one or two tracks of vocals and some multi-multi outboard processing, there are a few more things you can do besides adding short delays and soaking the tracks in reverb. The trick lies in making the "digital" vocals sound real, and in this case, "real" is each voice slightly off-pitch, out-of- phase and out-of-time with the others. Piece of cake. (Ha!) For optimum results, run the backing vocals through a 20 to 40 millisecond delay, out-of-phase if your DDL can do it; add a slow-wave, low-depth chorus to the delayed signal (this actually adds an additional delay to the original, in addition to a bit of detuning on most newer multi-effect units), and then the piece de resistance: early reflections! But since we're not using feedback, we don't need to worry about that "smeary" tape echo effect. Using an SPX-90, R-100 (not R-1000) or some other device that offers you an early-reflections algorithm, set the reflections to "Random" or "Gate", the pre-delay to zero and the delay length, density and room size to the setting that sounds best. Mix the effected signal back into the direct signal and bring down the level and time of the early reflection output level until you get that sweet choral sound you're after. "Bohemian Rhapsody" it's not, but it's about as good as you're gonna get without a Lexicon. (You HAVE a Lexicon? And you want to use it for THIS instead of as your main reverb?? For shame!) And if you can't do all these things, use as many as your gear will let you. Just keep in mind that the most important effects are chorusing and early reflections. Chorus with Class Bear with me for a moment. This next point might seem like a trivial matter to you, but it could well save you the cost of one or several recording sessions so I'm going through it in detail. One of the biggest problems I've had with fix-mixes is an out-of-tune lead vocal. Sometimes it's better to refuse the job than try to mix "around" a good singer who's been misled into singing off-pitch by a wobbly fundamental. No problem if you've got a vocalist who sounds like Lou Reed, but if the singer has good technique, it ruins the effectiveness of that singer's ability, and the only real fix is to run the vocal full-on through a quality harmonizer and write a controller sequence that corrects his or her pitch whenever it's needed. This is time-consuming, needless work, and bloody expensive for the client, especially if you have to rent a $2000 harmonizer for the day. And it's almost entirely preventable...providing the vocalist is relatively drug- and alcohol-free when the track gets laid down. The most common cause of this problem is an overdose of chorus. It's not much of a problem with analog stomp boxes, but I've heard good masters ruined by bad chorusing printed to tape. What many amateur recordists do is mix the detuned, delayed and modulated signal (all three operations are performed within multi- effect units' chorus algorithms) at a level equal to the direct signal. And here's what happens. You've got a blistering headphone mix and passion to spare. You're ready to squawk. But you don't have perfect pitch. The bed tracks are great...rich, thick bass; rich, thick guitar; rich, thick keyboards...but they're all tracked essentially at at least two different frequencies (the dry and the modulated frequency), and damned if one of them won't stop wobbling up and down! Yeah, it sounds dynamite...but WHERE THE HELL IS THE REFERENCE PITCH? You find yourself trying to stay in tune with tracks that your head is telling you are OUT of tune! Solution 1: Fire the vocalist and find someone with perfect pitch. I'm available and cheap, and I don't mind writing harmonizer sequences if you can afford it. Solution 2: Never, EVER print chorused signals on all taped instrumental tracks! If nothing else, preserve the bass track dry on tape so the singer can pick out a reference pitch. Even better, make it a habit to keep the chorus-effect level at least 20 percent below the level of the direct signal. You usually don't need a 50-50 mix to get the richness you're after, and when the true note is clearly distinguishable in the mix, it seems to add snap to your mix without sacrificing "liquidity". The other level to watch carefully is the depth parameter. Many people use far too much oscillator depth on their choruses. My opinion, one shared by many top producers and engineers, is that chorus works best when it can be described as "transparent", meaning audible, but almost invisible. Once a dedicated rubberhead, I've finally disciplined myself to use only enough oscillation depth to set the sound in motion, and my mixes are the better for it. Flange Tricks Interesting story about flanging...this may be partly inaccurate so please don't quote me...seems John Lennon was doing a track for a Beatles album and wanted some sort of effect on his voice that would emulate the sound of two vocal tracks. The engineer, either through earlier experimentation or quick thinking, I'm not sure which, decided to use two high-speed tape decks and vary the tape speed on one deck so that the pitch on the second deck would wobble slightly and be far enough behind the original track to sound like a twinned vocal. The effect not only doubled the voice, but when the signal was fed back through the first deck again, it could result in anything from a weird thickening to vicious jet-takeoff effects to unusual "metallic" filtering. Lennon asked the engineer what the effect was. Since it didn't have a name at the time, he called it "flanging", a name allegedly pulled out of nowhere. Since then, it's been thought that the name stemmed from riding the "flange" on the tape deck, the device that holds the reel in place, as a means of varying the speed. But the engineer who started it all swears that the name was a lark. Seems to me that riding the flange wouldn't be the as effective a way to pitch-modulate a tapedeck as riding the reel rim in any case. But enough of this crap. Flanging has generally come to be thought of as a short delay effect with feedback and oscillation, but flanging is really just any short delay effect that produces filtering. Understood in these terms, a whole new pallette of sounds opens up to the user of the well-tempered DDL. Frozen Flanging Most pedal flangers use a 50-50 delay-direct mix, which need not be the case with a DDL; and analog or digital flange effects are almost never applied by amateurs without modulation, which is a shame. One of my favorite effects is "frozen flange", or flanging with the modulator turned off (depth and speed are the modulator controls; "Feedback" controls the actual depth of the effect on flange pedals). There are numerous applications for this effect. One of the most useful applications is on a Fender Precision bass track played by a finger-style bassist. Fenders are not especially well-suited for modern recording, and I find adding a no-modulator flange, actually a simple delay with not a whole lot of feedback, 1.5 milliseconds or less, adds a nice upper-midrange presence to a Fender that it lacks when played without a pick. Careful of the feedback level, though, or the bass will sound "windy" in the mix. And remember that you don't have to have a 50-50 flange-direct mix to get this effect. "Frozen" (unmodulated) flange can be added to any bass track, but the delay time is CRITICAL! When you start using delay times between 1.5 and 35 milliseconds, you start cancelling out the necessary lower harmonics of the bass and murdering its presence. Since flanging is generally thought of as "transparent" echo (a delay so short that the human ear can't distinguish the echo from the original), delay times of less than 1.5 milliseconds must be used when flanging bass, unless you've got the necessary bass-cut EQing mentioned in the chorus section. Frozen flanging can also be used as a metallic "resonance" effect on virtually any track. It's often used on cymbals, though usually with some modulation, to add to the cymbal's brassiness. Fiddle with the delay time and you'll discover that it seems to act as a frequency control, while increasing feedback seems to increase resonance at that frequency. It's actually adding AND subtracting resonance in SEVERAL narrow frequency bands (this makes the wave look "spiky" on an oscilloscope, which is why flanging is sometimes called "comb" filtering), but your ear seems to tell you that it's behaving like a stoned parametric EQ or a synth's resonance control. It's a nice effect for metal, especially where you've got two guitar tracks that sound a bit too much alike. Two excellent examples of frozen flange used in metal are "One Way Street" from Aerosmith's first LP and the semi-crunch rhythm guitar in Judas Priest's "You Got Another Thing Comin'". (Ahhh...so THAT's how they did it!) But don't get the wrong idea...it's by no means a dinosaur effect! And hey...want some REAL monster flanging? Try combining an Early Reflections algorithm with flange-plus-feedback! Also a nice seasoning for dull hip-hop samples and sound-effect stings that don't. A NOTE ABOUT HARMONIZER SEQUENCING I probably shouldn't let that little anecdote about harmonizers go without a little more explanation. If you do much mixing for profit, this could be a nice way to boost your hours and help provide the client with a better product too. And if you ever get stuck with a magic track with some glaring pitch waver, here's proof that it can at least be repaired, regardless of what your engineer might say. "Sequencing" a harmonizer is a new concept. Until recently, harmonizers haven't had real-time MIDI control and haven't been clean enough to swallow whole something as fragile as a vocal track and vomit back a result that wasn't noticeably deteriorated. Without oversampling filters, harmonizers are still not as accurate as good studio monitors might demand of them, but when you're faced with a lead vocal or primary instrument track that's markedly out of tune in spots and can't be re-recorded, you could do a lot worse than to run the whole track through a Digitech IPS or a Lexicon LXP-5 (and if your client can afford better, rent it!) and let the sequencer pitch-map the track to perfection. Don't worry about the characteristic brittleness you might be used to hearing from harmonizers. You'll be shifting the pitch in cents, not semitones, and only in certain places. Any good 16-bit 16kHz harmonizer should do a fine job in these narrow shift ranges. Setting up the harmonizer to correct the pitch is easy. Simply run the track out from the direct output and back through the return, set your harmonizer so that it's spitting out 100% effect signal (no direct signal), and you're getting pure harmonized signal. Setting up the actual pitch correction is a bit more exacting. First you have to map the harmonizer's pitch shift to the pitch wheel of your controller keyboard, make sure it's usable as a real-time controller of the AMOUNT of pitch-shift in your harmonizer. I use the pitch wheel as a controller because it's most intuitive and fastest to program. It also allows smooth pitch shifts. Unless you're punching the harmonizer in and out on gaps in the signal, which is NOT recommended, any sudden shifts in pitch will sound jerky to the ear. Next, take the MIDI OUT from the keyboard to MIDI IN on the harmonizer, and the MIDI THRU on the harmonizer to the MIDI IN on your computer or MIDI interface. Assign the harmonizer its own track, and be sure your record filter is set to allow recording of controller data. Then set the master keyboard so that the pitch bend amount is +/- 1 semitone. (Anyone who goes out of tune more than a semitone shouldn't be in a studio!) This narrow range will allow you to make minute pitch adjustments quickly and accurately from the pitch wheel. All you have to do now is rock the pitch wheel smoothly back and forth from its center position so that it shifts up where the singer went flat and shifts down where he or she is sharp. Your sequencer will record your rocking motions on a separate track (NOTE: Controller tracks eat a TON of MIDI data!), and duplicate your actions every time you replay the track. If you can lock the sequencer to tape with Song Position Pointer or SMPTE, you can fine-tune the controller track as much as you like. But if all you've got is an FSK stripe that syncs but won't lock, you can still try rehearsing the track a few times before laying it down, or else try editing the sequence within the computer itself once it's recorded. Just be careful about making too many adjustments. The best course of action is probably to pitch-shift only the most severely warped notes and leave the other, less serious violations alone. Once the serious pitch errors are corrected, minor pitch errors are less noticeable. And if you're working on a track by a player or singer who's never in perfect tune anyways, sugar-coating all of the player's mistakes might very well disturb some of the emotional thrust of the track. Otherwise capable players with serious technical deficiencies often make up for this weakness with a greater emotional strength in their playing...witness Neil Young and Jimmy Page! REVERB TRICKS On units such as the MidiVerb/Fex/MidiVerb II/ProVerb, where the reverbs are burned into ROM at the factory and can't be tweaked, there's not a lot you can do to get more than you've already got. But you can try to make better use of what the factory gave you. One of my favorite tricks, and it only works if you've got good ears, good judgement and a lot of tracks, is to layer reverbs. On cheaper gear, when you record every track with its own reverb, you'll get a funny "summing" effect on mixdown. The various reverbs will interact with each other creating unusual phase cancellation effects, and if you've done it right, you'll notice "phantom" tracks arising from the background. "Phantom" tracks are instruments and sound effects that were never actually put on tape, but seem to appear out of nowhere in the mixdown. (You might be able to produce a musically-valid "phantom" track one time in ten as a beginner; once you can produce these effects on demand, it's time to hire out your services to the highest-paying studio you can find! My ratio is still only about one time in two.) Control Your Enthusiasm! One of the biggest complaints A&R people have about home- produced demos is the density of the sound. This "density" generally comes from using too much reverb. Anything that makes such a radical improvement in your sound will tend to put you over the deep end, and you can get into a nasty habit of over-using these effects on every track. Other devices that fall into the same category are parametric EQ's, which require enormous skill to use properly; aural exciters; BBE's Maxie units, which give a "forced" sound when used to excess; and dynamic range expanders, which seem to amplify small technical glitches into audible errors. With all such devices, my advice to beginner and intermediate-level home recordists is ALWAYS to set the device to the level where it sounds awesome to your ears, then BACK OFF!! The reduced effect level you set at this point will almost always be a happier medium. Reverb room size and decay time are two other areas where you might be well-served by peeling back a bit. Drum tracks tend to sound much crisper and cleaner when run through a medium or small room algorithm, even if the main reverb is a large hall. Large hall reverbs rarely need to have decay times longer than 2 seconds; a confusing statistic when you consider most programmable reverbs can decay for at least 20 seconds! Don't be too worried about "shrinking" your ambient environment to the point where it might seem unnatural. In actual fact, only the first second or so of reverb decay is truly important to the overall sound of the mix. Save your long "bloom" reverbs and huge halls for special effects or for ultra-sparse arrangements. Reverbs this long have no fit place in today's pop music...at least, not until the Gothic sound comes back. High-Calorie Reverb This trick makes a hell of a lot of sense, and it's simple, too. Still, it never even occurred to me until I noticed it in one of the preset programs on my Alesis Quadraverb: chorused reverb! For adding richness to a thin or tacky reverb sound, it's a terrific idea. You see, the richness of a reverb is in large part dependent upon how much RAM memory it has and how fast it can access what's in that memory. Pre-1987 digital reverbs and multi- effectors tend to be light on memory and pretty limp in comparison to even some pedal-type reverbs on the market today, and they can probably use all the help they can get. Adding a touch of slow- wave chorusing either before or after the reverb makes any reverb seem twice as lush. Just be careful about leaving "holes" in the music. The modulation will be painfully obvious during hard dynamic downshifts when the reverb is allowed to stand on its own. Pure Heaven Another favorite effect of mine is the "pure reverb" instrument track. Essentially it's a guitar or keyboard track that is...you guessed it...nothing but reverb. I simply run the instrument into the reverb, set the predelay to zero, drop the early reflections level (usually to zero so there's no hard smearing), and print the track with nothing but the thick, trailing reverb sound. The sensation is similar to that of being in the arena washroom during a concert soundcheck, only with more definition. It seems to sound best with hall- or chamber-type reverbs, and I've noticed that ART and Yamaha processors can't create this effect to my satisfaction. They both seem to have been programmed with a slide rule rather than the human ear and sound brittle in this application, but to be fair, digital reverbs were never designed with this particular application in mind. If hall-type reverbs are too big for this effect, try shortening the reverb to a second or so to simulate a room reverb decay. The diffusion of the large hall reverb will preserve a thick, smooth instrument tone even if the decay time is more suited to a medium room. And if shorter decay doesn't do it, try using a room reverb algorithm. It might sound brassier, but that brassiness might be just what the track requires. In any case, try this trick on trebly leads or languid rhythm tracks and see if it doesn't do a lot to improve "track separation" on your mixes. Especially nice on pianos! There are plenty of other low-calorie sweeteners available for digital reverb, but you can find most of these ideas in any good introductory article on reverb or in popular home-studio publications. The ones included here were those I found few people either knew about or knew how to use properly. ----------------------------------------------------------------- COMING IN MARCH: STUDTRX2.TXT Our next textfile will cover a subject that should be of interest to everyone with any kind of investment in analog signal processing. Among the topics covered will be noise gating and compression, spring reverbs, distortion and overdrive, graphic and parametric equalization, and - don't sell your pedalboard yet! - unusual and idiot-proof modifications you can make on older stomp- box effects. Once again, Steve will concentrate on tricks and ideas we haven't seen in common use. ----------------------------------------------------------------- A BAD CASE OF PLUG-ITIS We're capitalists. Naturally we're not taking this much time and trouble for nothing. We realize the concept of shareware textfiles is pretty pointless, but if you feel so inclined, you're more than welcome to order any of the following from us (All prices postpaid in US funds; Canadian funds accepted at par): LIES AND DAMN LIES: The latest cassette release from Steve Winter and MODEM. Recorded on a Fostex A-8/350 system, this should give you an excellent idea of what can be achieved with bare-bones studio tech combining analog and MIDI signals. This tape currently has four songs; each cassette is individually dubbed from the master upon receipt of your order, so more tracks will be added as songs are completed. A special sheet outlining the techniques used to achieve certain effects will be included on request. The music is basic, guitar-driven "mutant blues" that's been described as suitable for fans of U2, Tears for Fears, Robbie Robertson, Gene Loves Jezebel and 54-40. NOTE: It is NOT "musicians' music"; the accent is on strong songs, solid production and emotional integrity. Chrome cassette: $6 THE ULTIMATE SHOPPER: The funniest, most informative moneysaving guide ever published, and one of the most highly- publicized consumer handbooks in Canadian history. Written by a musician for the North American marketplace (not just Canada), it provides hundreds of the most effective and unusual moneysaving ideas you'll ever see on everything from food and household items to insurance and garage sales. This is the consumer handbook for people who love to buy but hate to "shop"! 192 pp. quality softcover: $9 THE MECHANICS OF HIT SONGWRITING: Based on Steve Winter's acclaimed songwriting seminars, this program outlines step-by-step how to transform your songs into hit material, usually without the creative compromise you've been led to believe was required by The Industry. This is perhaps the most complete, artist-friendly step-by-step program available for today's pop and rock songwriter and covers everything from the most fundamental concepts to advanced tricks and "loopholes" you can use to revive your old demos, energize your current "projects in development" and how to turn what might seem like industry roadblocks into real creative opportunities. Designed for everyone from the weekend hacker with a four-track to the career songwriter to the performing original artist. Complete step-by-step tutorial, including our exclusive A to Z project checklist (an absolute must for ALL songwriters...we don't start tape without it!) and chrome cassette with recorded examples, $25. Forward orders, inquiries, hate mail, etc. to: Robert R. Shrier or Steve Winter Living Skill Media Unit 2 6508-33 Ave. NW Calgary, AB ('The City that Pop Forgot') T3B-1L1 CANADA ----------------------------------------------------------------- NOTE TO ALL READERS: In order to save you download costs, we are seriously considering using the ARC compression utility on all subseqent uploads. Please inform R.SHRIER via Email before March 15 if you will be adversely by having these textfiles ARCed. -----------------------------------------------------------------