A guide to the Yamaha EX5, EX5R, and EX7 Synthesisers

table of contents

Chapter 1. Chapter 2. Chapter 3. Chapter 4. Chapter 5. Chapter 6. Chapter 7. Chapter 8. Chapter 9. Chapter 10. Epilogue About this guide Setting Up The EX Architecture Voices Performances Sampling The Arpeggiator Patterns and Keymaps The Sequencer Data Management and Utilities 72

About this book.

The purpose of this book is to provide an easy introduction to the features and potential of the Yamaha EX5, EX5R, and EX7 synthesisers. The manual tends to treat every subject with equal weight and is thus incredibly boring and difficult to read. This book assumes you know how to power a synth on and off, and that you have a basic understanding of what MIDI is or does. You may need to consult the manual if youre unfamiliar with certain terms, but generally this guide should be sufficient to get you using the EX to its fullest, programming your own sounds, and making music. The EX is driven by one of the most powerful synthesis engines in the world, and this guide aims to show you how to make the best use of it. There are several different types of synthesis available : 1. AWM 2. AN 3. VL Yamahas wave playback synthesis engine, which provides access to hundreds of superb digital waves stored on 16 MB of internal ROM. Yamahas Virtual Analogue Model algorithm, the heart of the Yamaha AN1x synthesiser, oscillators, filters, etc. Yamahas Virtual Acoustic Model, the engine behind the Yamaha VL-1, VL1-m, and VL70m synthesisers (this algorithm is not available in the EX7 model) Formulated DSP, which is a new collection of algorithms offering synthesis and processing techniques, new to the EX series. Some of these are quite mad, and deserve your abuse. You can sample your own stuff, load samples from disks, and include these waveforms as elements in your voices.

4. FDSP

5. Sampling
Included with this guide is a floppy disk with a selection of voices. These voices will be referred to in the guide, because they demonstrate specific aspects of the EX synthesis engine. If you want to follow through the guide start to finish, then you should load these voices into your EX. Dont worry about your other internal voices, they should be on a disk somewhere. You kept those other disks, right? Basically, the voices and performances of this floppy disk are the same as the EX7 UK Factory Set #1, except that the first few Performances and Voices are changed for our examples. We will also cover some of the compositional tools included in your EX, such as the fully programmable Arpeggiator, Sampling, the Keymap mode, and the built-in sequencer. But remember, mash everything up, make your own sound, and eventually try to forget any of the rules you might learn in this book. Only a wimp uses factory presets, right? Alright then.

Setting Up

If youre not using your EX with an external sequencer, then setting up is just a matter of plugging it in and turning it on. Connect the audio outputs to your mixer or amplifier and youre away.
Connecting to a Sequencer
If youre incorporating your EX into an existing MIDI setup, or connecting it to any MIDI sequencer or computer, youll have to connect both the MIDI IN and MIDI OUT to the sequencer or computers MIDI interface. (On the EX5 use MIDI A ports for the time being.) Usually in this case, youll set the EX to Local OFF mode, since the sequencer will route the MIDI from the keyboard to the currently selected instrument. (Local ON/OFF is in the Utility/MIDI page). If Local is OFF, then the EX is only responding to MIDI arriving at the MIDI IN port, so you must configure the sequencer to send the MIDI Thru, back to the EX again, on the desired channel. If you do this while Local is ON, you will hear each note flanging slightly, since it is being played twice, by the Local Keyboard AND from MIDI. Youve probably encountered this before. You can set the Global MIDI Transmit channel and the Global MIDI Receive channel in the Utility/MIDI page of your EX.

Probably by now, your Open Saw is starting to get a little less than useful. You might want to consider adding some real-time control to the voice, so that you can adjust the filter cutoffs, resonance, LFO speeds, etc. Lets exit the LFO page, and drop back down into CTRL (Control). There are two sections to the Control page, the first is called Pitch. In pitch you can set the pitch bend range, and the upper and lower ranges can be set seperately. The lower range can be set to a maximum of 4 octaves (48 semitones). This page also lets you set the portamento time, and mode. The two modes are fingered and full time. If you set it to fingered portamento only happens when notes overlap. In full time portamento is always active. Now, lets check the second of the CTRL pages, called SET. SET is basically a controller matrix. You have 16 sets, which is a little like having 16 patch cords. You can assign 16 controller+destination combinations. Okay? You can turn each of these sets on or off for each element (to the left of the screen), although at the moment were working with a single element voice. So lets turn set 1 ON for Element 1, and the turn Knob 1 on, and then down at the bottom set the Destination (DST) to 077: AWM DCF Reso. The DEPTH of this control can be set to the right, lets set it to maximum for now.
Of course, when youre in EDIT mode, the knobs are functioning as data entry for the edits you make, so if you want to check the results of this assignment without leaving the editor, hit the hardware switch called KNOB MODE to test the results of Knob 1. Hit it again to turn it off. Now lets say we want to use the Mod Wheel to control the amount LFO1 modulates the Filter Cutoff. Go up to the top of the screen and increment set1 to set2. Turn it ON for the element. Then turn on MW1. Set the Destination to 053: AWM LFO1 FMD (filter modulation depth), and set the depth again to maximum, like this:
Got it? Okay, lets have a look at the effects. As we mentioned earlier, there are two types of effects: insert effects and send effects. The insert effects can be configured in three ways; parallel, or serial (with either INS1 or INS2 first in line). You can set this on the front page of the EFCT section. This page also lets you set up the send and return levels for each of the send effects (REV and CHO) :

9. RindMod

10. Seismic
So lets check an example. Choose voice 002 in the I1 bank. Its called FDSP Saw. This voice is using the first FDSP Algorithm called EP Pickup, meaning electric piano pickup. Hit EDIT/COM/FDSP. As with other parameters in the EX, each element can be routed via the left hand of the screen. This voice uses two elements, and both of them are switched ON for the FDSP algorithm. If you turn them off, you can hear that the voice is composed of two, detuned, synth waves. Turn them back on again and start looking through the FDSP parameters.
The three types result in progressively more aggressive settings. The middle integrate type, is most useful for actual pickup simulations. Try adjusting the drive parameter to see what happens. This is like moving the pickup closer to the tine of the electric piano. Youll notice that the higher the drive, the greater the dynamic range of the pickup, meaning that the changes in timbre from low velocity to high velocity are exaggerated.
Lets try some other algorithms. Choose number 3, Water. Scroll to the last of the Water parameters and set the Dry Level to zero, so youre hearing only the effect. Now go back to the top. The pitch parameters refer to pitch of the virtual string or resonator. Key follow refers to how this pitch changes across the keyboard. (Note: With all FDSP key follow parameters, a value of 32 represents an even octave scaling. ie. a filter scaled at 32 will scale such that one octave on the keyboard moves the cutoff frequency one octave. This is useful if youre using high resonance filters, and want to keep them in tune.). Scroll down further and experiment with Resonance, Freq Mod, and Mod Speed. Resonance is that of the filter being modulated. If you scroll further, the Feedback parameter is the resonance amount of the virtual string being triggered by this filter modulation.
Now try algorithm 7, Self FM. Again, first scroll to the end and turn off the dry signal. Essentially, what we have here is a single oscillator FM system. The envelope can control the drive of the modulation. This envelope can be either in attack mode or decay mode (its very simple). In decay mode, it starts at maximum, and you have control over the time until zero. In attack mode, it starts at zero and you have control over the time until maximum. Set it to Decay with a time of about 70. Now go up to drive EG and set to maximum (63), and finally go to the top and start lowering the pitch coarse.
Getting the idea? Notice also that you can control some of the FDSP parameters in real time using the knobs or other MIDI controllers. The FDSP Saw voice has Knob 1 assigned to the pickup drive parameter. A full explanation of all the algorithms, and each parameter, begins on page 124 of the manual.

In this mode, the Sync is active and the first oscillator (VCO1) is being frequency modulated by two sources. Usually, once of these will be VCO2 (under Src). Try bringing up the depth on the FM, so you can hear what happens. The timbre gets more complex, and the perceived pitch changes. Now change the Src to VCO1, so that this oscillator is modulating itself. Now, under the FM Mod column, change the fixed to FEG (the filter envelope). Now the depth of the FM is being controlled by the filter envelope. Zero the FM depth. Lets try the Sync. Sync is the result of the one oscillator being slaved to another oscillators cycle. On analogue machines it was an nice way of creating more complex sounds. In our AN algorithm, the Sync is created within the one oscillator, so you dont have to worry about VCO2. The Slave oscillator is virtual. When you change the Sync Pitch, the harmonic structure of the main oscillator changes quite radically, but the percieved pitch stays the same. The depth in the Sync section is again to modulate the Sync oscillator over time, you can use the filter envelope or other sources to do this. Try it out. The filter section of an AN element is slightly different from that of the AWM element. There are not so many filter types, but there is always a seperate HPF (High Pass Filter) to roll off some of the bottom end if necessary (and makes an interesting controller destination, by the way). Also, the EG Depth control is more similar to an actual analogue machine, allowing you to modulate the cutoff frequency in either a positive or negative direction.
The pitch EG is slightly different also. Its not the full ADSR envelope found in the other element types, but just an attack portion. However it can be positive or negative in depth, and have a positive or negative setting (see page 117 of the manual). Try increasing the time parameter. Youll find that our voice takes more time to come UP to the pitch. Try decreasing it, take it below zero. Now the EG has reversed its setting so that we begin on the original pitch and slide away from it.
Lets have a look at the AMP page. Page EDIT/AMP/MIX.
Here we find the level for the mysterious VCO2 which weve heard about, but not heard yet. You also have access to a noise generator, a Ring Modulator (between VCO1 and VCO2), and the Feedback control. The Feedback parameter is the strange one, for a couple of reasons. Although its set at 127, youre not hearing it yet. And this is because its assigned to a controller (Knob 4). Hit KNOB if you want to try it (see Recipe box at the end of this section). Careful, its dangerous. What this does is route the output back into the mixer again. Its great for fattening or distorting basses, etc. One last point about the AN voice. If you check the control settings page (EDIT/CTRL/SET) youll see that the destinations for cutoff, or resonance, etc. are different from those in an AWM element. For example, the AWM Resonance is destination number 77, but the AN Resonance is destination 111.

You can also change the depth, offset, and curve of this Knob if you like. Hit KNOB Mode to check its working. Yep, its working. But I dont like the way the Sustain pedal is affecting the Matrix, since I want to just sustain some Strings, and then play on the Matrix above. So I go to the Controller Transmit page, EDIT/PART/CTRL, and turn OFF the Sustain pedal (Sus). This stops the Sustain pedal from being transmitted, but it will still function locally.
You can have up to 16 zones on the keyboard this way, all playing different instruments if you like. Notice that the Layer Switch is only relevant for the Local sound of the EX, you can turn ON the Tx MIDI A or Tx MIDI B for any other Part at the same time. If you decide to explore the Master Keyboard possibilites, and start building your own custom Performances for this purpose, you should also be aware of the Preset page (PRE). What this gives you is an INITIAL STATE for each Part or MIDI channel. When the Performance is selected, each Part can send a Bank Select, Program Change, Pitch Bend Value, and an intial value for all the controllers and knobs on the EX. Here, Ive set up the Performance so that it sends a default value of 20 to my Matrix 12 on controller #74 (KNOB 1). Note that you can choose whether the default is sent to the internal TG, external MIDI, or both.
This can be extremely useful for live performance. Each song in the set can be set up completely with a single Performance selection.
Edit Compare When Editing a Voice or Performance you can quickly check the existing Stored version by hitting the EDIT button. Hit the button again, and you return to your current state.

Sampling

There are several ways to get sample data into the EX. You can record samples from the analog inputs, and you can sample sounds the EX makes itself (internal resampling). You can load WAV, AIFF, or Akai files from floppy disks. And you can send sample data into the EX from a computer using SMDI.

Recording a Sample

If you want to record something directly with the EX, enter the SAMPLE page, and hit the REC f-key. You should see something resembling this:
The record mode can be set to several Left/Right combinations in the EX5 and EX5R, but in the EX7 you can only record mono samples (although you can play back stereo samples). The Source parameter can be switched to A/D (analog to digital converter, otherwise known as your audio inputs). You can choose line or mic level input. And you have a hardware input gain on a knob above the master volume fader. If you want to resample the EX itself, then set the source to Internal. You can trigger the sampling by level or manual command. When set to level the sampling will begin whenever any input occurs above the threshold you set. You can set the maximum size of a sample below that, and also use this to see how much memory you have available. Hit Standby to activate the recording stage.

The currently selected parameter in the above picture is the Loop Length. You can scroll this with Knob 4. However, you might find that the scrolling is very slow. Notice in the bottom left corner theres an KNx1000 written. If you repeatedly tap this f-key, you change the number of samples you jump when you turn the knob one tick. Set this to 1000 and you can move the points much more quickly. When you want to fine tune the placement of the points, you can set it smaller again. Notice the loop cross point zoom view to the right of the screen. If the Zoom is set to 1, this displays individual sample positions so that you can set the loop points very precisely. Notice that you must re-trigger the sample to hear the change in any points. Notice also that you cannot start a sample after the loop start point. Nor end a sample before the loop end point. The loop is always WITHIN the sample, even if it includes all of it. If you trim the front or the end of the sample, you may want to crop or extract it (ie. throw away the stuff outside your selection). You can do this with the JOB commands. In
this editor, when you hit JOB you have the choice to Extract or Normalize the current sample data. You can also delete, copy, or append two samples using these JOBs. Now do me a favour before we leave this section, set the sample to FwdNoLp (Forward, No Loop) and set the Loop In point to the end of the sample, 84966. Well see why later.
Using the Sample in a Voice
Go back to Voice Mode, and choose voice I1:007:Sample. In fact, this voice is just an Init Voice right now, but we will use it to make our Sample Voice. Hit EDIT/OSC and youll see that the current Oscillator is PRE:001:Pf:Grnd1, a standard piano sample in the Preset ROM collection. We want to access the RAM area, so lets change the PRE to RAM, the sample number is still 001, and hopefully youll see our LARGE wave written there. Like this:
But, somethings wrong, you still cant hear it, right? Right. You still cant hear it because this WAVE is currently empty. It stole the name LARGE from the sample, but in fact there still is no WAVE as yet using this sample. Jeez, you might say, Ive got to make a WAVE just to use a sample? Well, you do and you dont. You do if you want to use the sample in a Voice Element. (You dont if you only need to access the sample from a Drum Voice.) Its not difficult, though, and youll see in a second, it opens up some possibilites.
What exactly IS a WAVE? A WAVE is a group of one or more samples. The WAVE is great because it can contain 1 sample or 100 samples, and still be handled as one unit. Look at the following two diagrams: WAVE called Bill Bill Sample

R1 R2 R4 R6 R8 R10

WAVE called Fender Rhodes

R3 R5 R7 R9 R11

MIDI notes 1 > 127

Samples and Waves disappear when you power down. Back em up.
Why Not Put ALL My Samples In One Big Wave?
Well, because a WAVE is used by an Element. An Element has filters and envelopes and so on, but ONLY ONE SET. So if you wanted a different filter on the snare drum than on the kick drum, for example, youd be screwed if they were in the same WAVE. WAVES are for SAMPLES THAT BELONG TOGETHER.
How Can I Use More Than Four Samples In A Voice?
Use Drum Voice, coming up next.
Hardware Shortcuts 1 Learn to use the hardware buttons for Part and Track selection when in Performance mode, or any of the MIDI modes (patterns, arpeggiator, sequencer). 2 Also, notice the Element selection and Element On/Off function of the Bank Switches. This is useful in Voice Edit. 3 The Ten-key pad can be used for entering large values, such as sample addresses or delay times, but can also be used for quick entry of note values and velocity values in any of the MIDI modes. It can also be used for naming. 4 Whenever youre in an edit mode, the Knobs change function to become data entry. If you need to check the Knobs play function without leaving the editor (such as when assigning control destinations in a voice), just hit the hardware KNOB button.

Drum Voice

Effects Setup
Layer 1 Defined by Key AWM, points at Sample Velocity Range Voice Layer 128
Drum Voice is pretty important when you work with your own samples. If you want to have many different samples in one voice, with different envelopes and filter settings on each, then this is the type of voice to use. Essentially, Drum Voice is very similar to a normal Voice in the various sections of the Editor. The main difference is that Drum Voice is built of Layers instead of Elements. In Voice, you could have 4 Elements, and each of them had editing of filter, envelopes, etc. In Drum Voice, you can have up to 128 Layers, and each of these has its own filter, envelopes, etc. Each Layer, however, is associated with only one Key. Although, you can have more than one Layer on the same key if you wish. And of course, you can have several keys using the same sample. The Layers are created much like the Layers in Wave Edit. Lets use Drum Voice I1:008, as an example. Its completely empty at the moment. Choose it, hit EDIT/OSC/WAVE. In the upper left corner, you see written here C1. If its not on C1, choose C1 (you can type the number 36 on the ten-key pad if you want). Theres nothing on C1 yet, so lets hit ADD. This immediately creates a layer and defaults to a Preset Snare Drum wave. Scroll the wave back to BdDeep.1 (wave #1363). You screen should now look like this:
Lets imagine we want more of a click on this sound, so we need another wave on the same key. Hit ADD again. This will immediately copy the previous layer to a new layer, and the display will now say Layer= 2/002 (2 of 2, ignore the little black 3, thats the knob). Scroll the wave of this second layer to wave #1391, Stick.1. You now have two waves on the same key. If you choose the Play=default parameter, you can reverse the wave here if you like. You can also use a note-on delay, or alter the start time offset of the sample in this page. And most importantly, notice the parameter called Recv Note Off. This is set to ignore at the moment. This means that no matter how long you hold the note, the sound will play the same length. In truth, its using the amplitude envelope, but ignoring the sustain portion. If you are using samples, and want to have a sustain portion, and then a release portion, you should change this parameter to receive, in which case that layer will behave as a normal ADSR-type trigger. Note that each Layer has its own settings for all these parameters, even if theres more than one Layer on the Key. Now lets hit the MIX f-key to change to the MIX page. Here you find parameters to set the level and pan of the layers, the sends of the layers, as well as a choice of Insert Effect placement. Any drum layer can be in either insert effect (although there are only the two insert effects for the entire Drum Voice). Choose InsEF number 1 for both Layer 1 AND Layer 2 of our C1 key. Hit Exit. Then hit EFCT/INS1, and set it to type 15:Overdrive. You can now overdrive any drum in your kit if necessary. Set up the Overdrive to your liking, and lets go back out to the Oscillator page again. Look in the ZONE page. EDIT/OSC/ZONE.

Here you can set a Velocity lower and upper limit for any Layer. In this way you can build velocity splits. If you use the XFade amount, you will create a velocity crossfade. Also in this page, you can choose a Layer to be a member of a Group (Grp). If two layers are members of the same group, they will mute each other when they play. This is often used between high hat samples so that the closed hat will always mute the open hat. But you can use it for anything you find interesting. You can have 128 groups if necessary. The last parameter in this page is called Assign. In single mode, if the Layer is re-triggered it stops the previous one and begins the same sound again. If its in multi-mode it completes all of them, even if they overlap. The TUNE page lets you tune the Layers, as you might expect. Now lets choose a different key for the next experiment. Choose C#1, hit ADD, and choose the RAM sample we have. ( If youve powered down since the sampling section, you may have to load the LARGE.WAV again. ) It should be in position RAM:001. Try it.
You should hear the first couple beats of the drum loop. (This is an ideal sound to hear the difference between Assign single and Assign Multi in the ZONE page, by the way. Hit the key twice quickly, and you can hear whats happening in the multi setting.) Now, this loop is currently incomplete. You probably want to be able to control the length of the loop from the keyboard or MIDI, so we need to change the Recv Note Off parameter to Receive. Do this. Now you can hold the entire loop if you want. You might notice that the release time is rather long, and you cant stop the loop very quickly. This release time remains from the trigger mode that the drum voice defaulted to. We can fix it easily by jumping to the AMP/EG page and reducing the release time. Try this.
Lets make another Key, D1, and ADD the same sample to it. RAM:001:LARGE. Lets set the Amp release to 0, and the Recv Note Off to receive. Come back to the WAVE page:
C#1 and D1 should now sound the same. But lets change the Start Offset of D1 to somewhere around 22000 (use the ten-key pad), so that were triggering the loop at the point where the first snare drum starts. With several copies of a loop like this, you can transform breakbeats into your own new patterns very easily.
Now lets mess it up further. Go out of OSC and into FILT, enter the EG section of the filter page and set up a downward sweep, like this:
Go to the DCF section and choose the BPF (band pass filter) on both filters. Set the resonance quite high. And put the filters in parallel mode. Like this:
Now set the envelope sensitivities so that theyre opposite. This way the sweep will move the two band pass filters in opposite directions.
Its starting to get messed up pretty good now. If you want, go back to the OSC/MIX page, and turn on the insert effect for this layer. Remember that the samples are lost when you power down (unless youve used Flash RAM), but your Drum Voice can be stored in the EX, and will use the same samples if theyre loaded again. The only requirement is that the samples be in the same ORDER next time, since any voice using the RAM samples will do so by its sample position.

This is easy enough to do. If you save WAVE, it will save all the WAVES in memory. And when you Load WAVE you can load ALL of them at once, and they will be in the same positions by default. You can also load a single WAVE from a file of many WAVES if necessary, more on this in the chapter called Data Management.
Floppy Disk Format If you find that the floppy disk formatting is rather slow, use your computer. Its just a standard DOS disk.

The Arpeggiator

One of the great features of the EX synthesisers is the fully programmable, 4 track, arpeggiator. For an extensive explanation of everything the arpeggiator can do, check your manual at the section beginning on page 238. For the purposes of this guide, well show you the basics of making your own arpeggiator pattern, using an imported MIDI file as a template. Lets assume youve got a pattern in your sequencer that you think would make an excellent arpeggio pattern. Save it out as a MIDI file, copy it to a DOS formatted floppy, and import it into the EX. (The example were using is on your floppy disk, and is called Arp.MID). By default this MIDI file will be copied to the EX SONG file. If you then hit Play the onboard sequencer will play this file. Check this to make sure youve got the data. Now choose a voice to test the arpeggiator with. Turn the Arpeggio switch ON, and hit EDIT / COM / ARP. Scroll through the parameter called Type.
The first 50 are preset, you cant change them. At 51 they become User patterns, and you can choose any of these to make your own arpeggio pattern. Choose one of the User patterns, and hit F-3, Arp-Edit. Now you have to copy your MIDI file into this location. Hit JOB (beside the main Edit switch), and then section Job 2 / 3. Get Phrase. Youll now have a choice of what measure and track in the Song to copy into the Arpeggiator. Our example is in Track 1, Measure 001 002. Hit Enter and execute the Copy. Then hit exit a couple of times to get out of the Job mode and back into the Arpeggio editor. If you hold down some notes, it should make some attempt to play an arpeggio, but will probably sound completely ridiculous. Weve still got to set it up. Now it gets interesting. First, change the Arpeggio Length to 1 (since our MIDI file is only 1 measure), and set the tempo to whatever you want.

Again, with the patterns, you can offset many performance features of each track. You can introduce groove quantisations, transpositions, etc. But for now, all were interested in is the Length parameter, since this sets the length of the track.
Notice that when changing the Length parameter, there are two options. These are shown by the f-key selection below, which can be ALL or TR. If its set to ALL, then ALL the tracks in that pattern will be set to the Length you choose. If its set to TR then only the selected Track will be affected. Same goes for the other parameters in the PFX page. Im going to set the entire pattern to 1 bar for the time being. You can always go in and lengthen another track afterwards. So I set the Length to 2. Why 2? Well, because actually youre setting the Loop position, not the Length. So I want it to Loop at Bar 2, Beat 1. You can actually Loop at a beat within a bar if you like. (The Beat is displayed after the Bar in the Length parameter). This setting sets all tracks to Loop after 1 bar:
So now, I go back to the PTN page, and I see that the MaxLng display says 1. This displays the length of the longest track in the pattern. Using the Sequencer transport controls, hit Record then Play. The metronome should be ticking out its beat, and you can play something into the pattern. Its just looping the one bar around. Play a simple Kick drum or something into the pattern. Hit Stop, and then hit Play to make sure youve got it. Notice that when you hit the Record button, there are a choice of Record modes, which are equivalent to similar choices you have in your software sequencer. Here are your options.
This can be on or off. If Multi is on, then all tracks are recording at once. This can be useful if you have several MIDI channels coming into the EX from another sequencer, or from live MIDI instruments. When its OFF, you can choose which track to record on the bottom left of the screen. Replace mode. Whatever you record will replace the entire contents of that track. Overdub mode. Whatever you record will be added to the existing data within that track. Step entry. Sometimes this is the quickest way to get what you want, just choose the value of each note and hit it. In it goes. Notice two things in the step entry: 1- the note values can be quickly chosen with the ten-key pad 2 - if you set the velocity to EXT it uses the velocity at which you enter it

RPLC OVER STEP

So well record a kick drum, or something simple into Track 1. Make sure MULTI is OFF, and put yourself in RPLC mode. Hit Record, Play. Play a couple of beats. Stop, play it back, make sure youve got it. Second track. Choose Track 2 with the Program buttons on the right end of the front panel. Its a Rhodes piano sound. Hang on, we want another drum track, but this Performance has a Rhodes on channel 2. No problem, we change the channel of track 2 to 1. Hit TCH page:

If you dont, hit it again twice. The KEYMAP button is a toggle switch. Choose C1 either by scrolling or using the SEL key while hitting the MIDI note. Where it currently says off you can choose either a Pattern (ptn) or a Sample (smpl). Choose a pattern first of all. To the right, youll see a choice of all the Patterns in memory. In between youll also see a column that describes either a Track number or all. This is so you can choose just one track from a pattern. Set it to all for the time being:
Choose the Pattern that we made in the previous section, and hit C1. The Pattern starts to play. Hit it again to stop it. Cursor over to the far right where it says toggle. There are three modes of triggering for each Key in the KeyMap. Toggle means that you hit the key once to start and again to stop. Oneshot means that it will play the entire pattern once, without any looping. Gating means that it triggers much like a normal MIDI note, ie. it plays while you hold it down, and stops when you release it. This is my preferred mode, but its up to you whats appropriate for your pattern or jamming style.
Do you still have the LARGE sample in memory? If not, load it from the floppy. Choose C#1 and set it to sample, you can choose the sample and the type of triggering for the sample also. Additionally, with samples, you can tune them using the TUNE f-key to access the coarse and fine tune parameters. TUNE has no function for Patterns. Lets make one more key, on D1. Lets set this one to pattern, but just trigger Track 3 of our Pattern. This is where I put the bassline in the previous section. And our KeyMap now looks like this:
Now, theres only one problem with this KeyMap so far, and thats the fact that youre also triggering the Performance Part 1 (the Tech Kit on Channel 1). Theres no easy way to avoid this, so I suggest choosing the Tech Kit on another Part (lets use Part 4 for now) and setting that Part to receive MIDI Channel 1.
Then choose a Silent Voice on Part 1, or else turn the volume down. You may also find that if youre routing through a sequencer, all the Pattern tracks are triggering the sound on Channel 1 (or wherever your sequencer is re-routing all incoming MIDI). In this case, you should turn OFF the transmit to MIDI on the Pattern, or use a multi or Any channel track in your sequencer, and turn OFF the TG in the Pattern. For example, here is the Pattern set up so that it doesnt transmit any MIDI out MIDI port A:

Keymaps disappear when you power down. If you make a nice one, save it to disk.
About the JOBs There are many functions in the JOB section with patterns. You can copy a track or a pattern, you can get a phrase from a MIDI file in the Song section and make it a Pattern, you can quantize tracks, etc. Check the manual at page 227 for a full list of JOBs.

The Sequencer

If you have a Performance set up for a song, or just want to experiment with our Pattern Performance, choose the Performance and hit SONG. (You can also use SONG from Voice mode, but you will have only one Voice at a time in Voice Mode). Here you can see the 16 tracks of the sequencer, plus an Fx and a Pt track. The MIDI tracks 1 through 16 can be chosen quickly with the Program/Part/Track buttons.
As with the Pattern mode, each track has Play Effects (PFX) and a transmit option (TCH). Similarly, each Play Effect can be assigned to ALL or TR (individual track). You can also record Play Effect changes to happen at various times through the song. This is done on the Fx track. You can also include Patterns in the song by using the Pt track. These can be recorded real time, by entering the Pattern number while recording, or else edited in step time fashion. Choose a MIDI track first of all, hit Record.
Again, like Pattern, you have a choice of recording MULTI or not, as well as a choice of recording modes. There is one new mode here, not found in Pattern, called PNCH. Punch recording lets you define an in and an out time for recording. This is useful if you just want to record for a short passage within an otherwise perfect take. The Src parameter you see highlighted above can be switched between normal and arpeggio. If its in arpeggio then the output of the arpeggiator is captured as MIDI data, thus freeing the arpeggiator for use elsewhere.
So lets hit record and play any old stuff into a MIDI track. Done? Does it play back? Okay, hit EDIT.
This is the display of note data, hopefully the notes youve just played. There are two modes to this editor, INS (insert) and CHNG (change). In Change mode, as above, you can alter any of the fields with the data wheel or the ten-key pad. The first field is the time of the event, in bars/beats/ticks (at 480 ticks per quarter note). The next field is the pitch of the note. The third field is the length of the note, again in beats/ticks. (The note-like display to the right of this tries to display the closest musical equivalent of the duration, by the way.) Finally, at the last column, we have the velocity. If you hit INS, you can insert a note or any other MIDI event. Like this:
Lets have a look at the other types of tracks. The Tempo track lets you define tempos at any position in the song, so that you can change speed when necessary. Its usually best to do this with INS, since you cant record this real time anyway. If youve imported a MIDI file with tempo changes in it, you can edit them with CHNG. The Play Effect track (Fx) is a strange one. Here you can record real-time. Try it out. You can adjust the Groove template, change velocity offsets, etc. through the Song. Record some of this and then hit EDIT.

Once youve made a couple of these you can navigate during the saving and loading process.
SCSI drives tend to be much faster than floppies also. And if youre working with samples, youll need some space. Id recommend using a SCSI drive of some sort.

Epilogue

Where do we go from here?
If youve followed through most of this book, you should have an idea of the huge potential of this instrument. Not only is the EX providing the raw elements of many synthesis techniques, its also giving you some compositional tools to help generate ideas. Once you understand the basic architecture of the machine, its often useful to forget about the designed purpose of the various sections, and to strike out on your own. For example, the Electric Piano Pickup FDSP algorithm might sound great with a vocal sample going through it. Using the resampling feature, you can sample the results of one DSP, and then apply another to the result. You could import your drum grooves, turn them into arpeggio patterns and manipulate them with the Play Effects, transpose it realtime, etc. Put your breakbeats through the Seismic FDSP and turn them to mush. You could import each section of your track as a seperate pattern, and jam the tune live from a KeyMap. Or make a sixteen part keyboard split as a Performance, set them all to the same MIDI channel, turn the arpeggiator on and stand back. A lot of interesting sounds are creating by merely trying stuff out, experimenting without a fixed purpose. Mistakes can be very cool. Miles Davis said There are no wrong notes. Brian Eno said Make the machinery fail. In other words, go for it, break the rules, and make it squeal like a pig.
And now, my son, you must eat this book.