Poly-800 Programming Guide A while ago I was asked to explain how the Poly-800 works - what each parameter does and how to use the synth. I did so via e-mail, and over the course of 5 lessons, basically explained the whole synth as well as I could. Well, seeing as I put all that effort into it, I thought I might as well make the information available to anyone who wants it. So here it is, edited for your viewing and use. Note: I do not have a manual for the Poly-800, so this is written from what I can gather. Therefore, some of it may be slightly inaccurate, but at the same time it is written by someone who had to "figure it out" themselves, which gives me (I hope) a different perspective. The Basics... OK, turn the synth on. It starts up in patch select mode, showing the patch number (as well as the letter P). It defaults to patch 11 every time. To change the patch, simply enter the new patch number you want to use. Then play to your hearts content. To change to parameter edit mode, simply press the button on the top right of the numeric keypad. There will be now 3 numbers shown. The first is the patch number, the next is the parameter number, the third is its value. Pressing on the up and down keys will change the value, while pressing the numbers will change to different paramters (eg 41 is cutoff frequency). There will be a little dot next to the third number when you change the value - this means that the parameter has now been changed. If you go back to the value that it was on originally, the dot will disappear (useful if you change the value and then change your mind about the change). To save the sounds you need batteries. The batteries can also do double duty as a power supply, though they won't last all that long (about 4-5 hours). Saving sounds: On the back near the headphones, power socket, outputs, midi etc you'll see a program write switch - its got 2 positions, enabled and disabled. Its pretty obvious which does which :). Its best to keep it in disabled most of the time - only switch it to enabled when you definitely want to write a sound. Ok, put it into the enabled position. Now press the red write button on the front - the program number will disappear, and there will be 2 flashing lines. Now all you gotta do is enter the patch position you want to save it to. Then move the switch back to disabled. And there you go, a sound has been saved. Writing sequences: Next to the program write switch is a sequencer write switch, again with enabled/disabled. You have to have it in disabled to be able to play the sequencer - it will not work otherwise. Move it to the enabled position. Now press the sequencer play button - instead of playing, it will show 01 in one of the three number spots. Pressing a key will shift it to 02 and so on. Write in the sequence, move the switch back to disabled. The display will revert back to showing the patch number, and pressing play on the sequencer will now play your programmed sequence. Viola :) Ok, thats the basic way to program the sequence. There are a few things to be aware of tho : more than one note can be played at each step ie chords :) there are 2 spots per step - but you can only "get" to the second one by playing 2 notes in quick succession. its a step sequencer - if you want to do pads and chords and stuff, it cant be done. Holding a key to get a sustained note wont do anything. rests (no note playing) can be done as well, by pressing the step button. Its also possible to go backwards by pressing the play button again I THINK. I'm not sure if thats entirely correct - I know you can go backwards... Bank hold - Underneath the keypad is the bank hold button. This simply holds the first digit in either the patch number or parameter number constant, meaning you can only change the second digit. For example, if you're in patch select mode, with patch 11 showing, pressing the bank hold button will cause a light to come on. Now pressing 2 will change to patch 12, 3 to 13 and so on. Press it again to return to normal mode. Its the same when in parameter select mode. So whats the point of it? Its most useful when you're editing a sound, and are trying to get one part of it correct eg the envelope. Instead of having to press 51, then 52, then 51 again, you can press 1 then 2 then 1 and so on. It just saves abit of time. The sequencer - It does not transmit any MIDI data ie it cannot be used to sequence another synth. A bit disappointing... Foot Switch - On the back there is a program change or foot switch input - its used for changing patches with some foot controller. Personnally I've never used it... The Synth and its Parameters... The first part to understanding the synth is to get a handle on analogue synthesis. Now before anyone wants to take issue with me, I know the Poly-800 isn't a true analogue. But it operates basically like one, and has an analogue (VCF) filter. The basic analogue setup: oscillator ---> filter ---> amplitude ---> sound out The oscillator produces the basic waveform, the filter does its work, and then the sound is amplified for final output. Thats the (very basic) way a analogue (and digitals to some respect) work. Ok, now to getting some sound. The Poly-800 has 8 digitally-controlled oscillators (DCOs), available either as a set of 8 or 2 sets of 4. This bit may sound a little confusing, so I'll explain. There are 2 "modes" for the oscillators, and which mode you choose effects the polyphony (the number of sounds you can play at once). When the mode is single, all 8 oscillators are assigned to the first set of paramters (numbers 11 -> 17), and you have 8-note polyphony. When you select double mode, 4 oscillators are assigned to parameters 11-17, while the other 4 are assigned to parameters 21-27. What this means is that when you press a key, 2 oscillators are set off - one using the values of parameters 11-17, the other 21-27. But as 2 oscillators are used for every key press, the polyphony is cut to 4 ie only 4 simultaneous notes. Thus if you want to do chords, you really need single mode, as otherwise the polyphony is too low. Oscillator parameters: 11 (21 for the second set) - this selects the octave range that the keyboard plays. It has three settings - low, mid and high. I'm not actually sure what octaves are covered in each setting, though it might be something like: -- low : C-0 to C-4 -- mid : C-1 to C-5 -- hi : C-2 to C-6 12 (22) - waveform select. This simply toggles between the 2 waveforms that the oscillators can use - square or sawtooth. Each will produce a different sound. 13-16 (23-26) Ok, I'll admit I'm not entirely sure of the musical theory behind these settings, but they seem to add different harmonic content to each oscillator. I know the measurements used to describe them (16', 8', 4', 2') are some reference to pipe lengths in an organ or something, and represent different frequency areas. Having all these set to 0 (off) kills the sound of the oscillator. 16' (13/23) gives a lower sound, 2' (16/26) the highest. Having them all on produces the most complete sound... 17 (27) - oscillator level. Simply sets the volume of each oscillator set. Runs from 0 to 31 for some reason. 18 - mode. Selects single or double mode. 31 - interval. Basically the interval between oscillator sets 1 and 2 - how late oscillator set 2 starts relative to 1. 32 - detune. How oscillator set 2 is detuned relative to oscillator set 1. 33 - noise level. Changes the volume of the white noise generator. Ranges from 0-31 like alot of parameters. I have no idea why they chose 31... Filter parameters The Poly-800 has a low-pass filter. This means it lets through all frequencies lower than the frequency set by the cutoff level. There are also high-pass filters and band-pass, but lets no worry about them. 41 - cutoff. Changes the cutoff frequency of the filter. Ranges from 0-99, but I don't know what numbers evaluate to what frequencies - thats something that never seems to get mentioned. Anyway, 99 means the filter is "wide open" - letting through as many frequencies as possible - while 0 is "shut" - letting through only very low frequencies (ie bass). This is possibly the most used control on a analogue synth. 42 - resonance. Altering the resonance means emphasizing certain frequencies near the cutoff point. This changes the harmonic makeup of the sound. Its effect can be subtle or it can be very obvious. It can produce a sort of "wet" sound if used rightly. When people talk of making a synth "squeal like a pig", they're talking about adjusting the resonance. It ranges from 0-15, though again i have no idea what the different values refer to, other than 0 means no rez, and 15 means lots :). 43 - keyboard tracking. Three values - 0 = off, 1 = half, 2 = full. What keyboard tracking does is assigns the note played to a certain cutoff frequency. A low note (eg C-1) will have the cutoff right down, while a high note (eg C-6) would have the cutoff very high. This produces and interesting effect, where as you play up the keyboard, the cutoff changes accordingly. Honestly I dont use this much, as I prefer to alter the cutoff myself. I dont know what the half setting corresponds to - it might mean only half the keyboard does it, it might mean use half of the set cutoff value as the highest. To see it work, program a low-note sequence, then start it playing. It should seem sorta subdued. While its playing, play and hold a high note. Once the high note dies away, you should notice the lower notes have had their cutoff changed to the value for the high note. Releasing the high note reverts them back to their original value. 44 - polarity. This changes the way the envelope effects the filter, by using either the standard way (the 2 setting) of the inverse way (the 1 setting). It makes a big difference to sound. I may not have explained it too clearly, cos I'm not entirely sure the way it works. 45 - EG int. Stands for envelope generator integer I THINK :). The EG part is definitely right. I'm not actually sure what this sets - all I know is what happens to the sound. It seems to effect maybe the modulation depth of the filter (modulation caused by the envelope generator). It changes how "severe" the filter is in some respects. Best to experiment with it to see what I mean. 46 - trigger. determines how the filter is triggered with different notes. In single (1) mode, I think every key press re-triggers the envelope and filter. In multi (2) mode, pressing and holding down a key will effect subsequent presses by not re-running the envelope and filter. Chorus 48 - chorus. Adds a chorus effect to the sound. A chorus is when a very short delay (echo) is added to the sound, and mixed with the original to produce a richer, thicker sound. Its not for making echos - the delay time is much too short. It works very well on pads and leads. Unfortunately, the Poly-800's chorus seems to be pretty useless (unless mines broken in this respect). It doesnt seem to do much at all to the sound, unlike the Juno-106's chorus which makes a massive difference. Envelopes parameters There are three envelopes on the Poly-800. All have exactly the same controls, though naturally they effect different things. But firstly, what are they for? The purpose of the envelopes (or envelope generators - EG) is to control something over time. The envelopes for the 2 oscillator sets control volume over time, while the third envelope controls both noise volume and filter modulation over time. The volume aspect is pretty obvious, but the modulation part may not be... The idea behind filter modulation is to change some part of the filter over time. Most commonly this will cutoff frequency, as I think it is for the Poly-800. The settings for the envelope will make the cutoff change as the envelope "runs". So if you for example you have a long attack time for the envelope, the cutoff will rise over the time the attack is running. It may effect resonance as well - I'm not entirely sure on that. The basic envelope: envelope picture Thats Attack, Decay, Sustain, Release - each refers to a part of the envelope. In truth, an envelope can be as complex as the manufacturer wants, though normally its a simple ADSR envelope. The Poly-800 has 2 further parts, the slope and the breaking point. I'm not sure where these are in relation to the picture. 51,61,71 - attack. This refers to the beginning of the envelope - how quickly the volume rises to its peak volume. 52,62,72 - decay. How quickly the volume fades from the peak volume to the sustain volume. 53,63,73 - breaking point. Actually, I'm not sure. I haven't figured out what it alters. I've heard the changes, and its something to do with the way the decay goes to the sustain part - it may be the volume that the decay stops at before going to sustain. 54,64,74 - slope. Again, I'm not sure. It may be refering to the slope of the sustain level, as it does seem to alter that in some way, but I cannot be sure. 55,65,75 - sustain. Unlike ADR, this refers to a volume level, unlike those three, which refers to a time length. It sets the volume level that the sound decays to in most synths. In the Poly-800 the decay doesn't seem to go straight to the sustain, but perhaps runs through the breaking point before reaching sustain. 56,66,76 - release. Time taken from the end of the sustain to zero volume. Ok, theres the definitions. Now what actually happens? You press and hold the key down. The envelope starts. It runs through the attack, hits its peak volume, decays down to the sustain level and stays there. Upon releasing the key, the release part is run til there is no sound. Thats the basic idea. But, like I said, the Poly-800 doesn't seem to do this precisely. Also, if you release the key while it is still in its attack or decay stages, I'm pretty sure it does not run the release part. The release part only runs if the sustain level is reached. This is different from some synths (this does not happen on the Juno-106). Thats the basics behind the envelopes. If you find there is no sound when you press a key, it might be because all the envelope settings are 0. If you don't want the sound to keep playing forever while the key is depressed, set the sustain to zero. And so on... MG parameters I actually don't know what MG stands for, but it is just the LFO - low frequency oscillator. Its used to produce rhythmic variations in the sound as it plays eg adding vibrato to a string sound. LFOs aren't made to be heard - only the effect of them is heard. The LFO on the Poly-800 is a sine wave (as far as I can tell). 81 - frequency. Changes the frequency, or speed, of the LFO. On the 0 setting, the LFO is very slow, while on 15 it is at its fastest. 82 - delay. Determines the time delay of the LFO starting after a key has been pressed. At 0, the LFO triggers instantly, while at 15 it takes a few seconds to start up. Unfortunately, once the delay time has run out, the LFO starts straight away, unlike some synths where it comes in gradually. 83 - DCO. Determines how much the LFO effects the oscillators. At 0, it has no effect on them, while at 15 it effects them fully. What it does is alters the volume of the oscillators over time. 84 - VCF. Determines how much the LFO effects the filter. At 0, it has no effect, while at 15 it effects them fully. What it does is opens and closes the filter over time. To be more precise (I hope :)) the LFO is in fact altering the envelopes of the oscillators and filter. MIDI parameters The MIDI implementation is fairly rudimentary... 86 - receive channel. Sets the MIDI channel that the Poly-800 transmits and receives on. Naturally selectable between 1 to 16. 87 - program change. Determines whether to allow program changes via MIDI. With it enabled, you can issue program change messages to change the patch that is currently playing. 88 - sequencer clock. Allows you to set whether the internal sequencer will start in time with MIDI clock data. With it set to internal, the sequencer can only be triggered by pressing the start button. But when it is set to external, when play is pressed on a sequencer or drum machine, the sequencer will start and stay in time with the incoming MIDI clock data. I think its 4 steps of the sequencer = 1 BPM. Setting up a string sound I was also asked to explain how to program a string sound in a step by step manner. Heres what I wrote. Remeber, this is primarily aimed at the Poly-800, though I suppose it could be extended to any synth with similar parameters. All "images" are preformatted text - I haven't changed them into a proper graphics format yet. The most important part of a string sound (in my opinion at least) is the volume envelope. The actual sound it makes isn't important at first, its getting it to sound (volume-envelope-wise) like an actual string. A basic idea of how the volume envelope should run: _____________________ / \ / \ / \ / \ ^ |_ key release Thats sort of how a real string volume envelope is, ignoring things like vibrato. So this first step is to model this accurately. The most important bit firstly is the sustain. Set this to 31. Never mind if it sounds loud, you can alter the oscillator volume at a later date. Now the release. This is more to personnal taste, but you want the sound to die away like a real string does. A setting of around 20 through to 31 should do this. Choose whatever you want. Next is the attack. Again, this is to personnal taste. I like longish buildups on my string sounds, sort of like a violin coming in. But you may want a more immediate sound. I don't think a setting of 0 is suitable - its too quick for my liking. Anything above 10 or so sounds alright. You want the decay to be 0 - straight into the sustain part. Now this is where it gets a little odd. On my juno-106, setting the decay to 0 makes the attack go straight (smoothly) into the sustain part. but because the poly-800 has the breaking-point and slope controls (which must come between decay and sustain), setting decay to 0 and break-point to 0 causes an audible break in the sound. The way to defeat this is to set the break-point to 31. Now, for the filter envelope, have everything set to 0 except the sustain, which you set to 31 - you want a constant filter sound. If you want some sort of dynamic change in the sound as its played, you could try something else, but it will be difficult to emulate the tonal characteristics of a real string sound. Now, with these envelope settings, you will notice something. Play (and hold) a note. Once its sustaining, release it. It will begin its release portion of the envelope. Now, before its finished releasing, play a different note. Notice how it causes the other note to sound again? There is one definite way to get around this - play the notes so they overlap. Play one note, and before you release the key, play the next note. With a short(ish) release time and a longish attack, it shouldn't sound too cluttered. You can also get around it (I think) by setting the filter's envelope release to the same value as the volume envelopes. This may work, as it seems the filter is sustaining at the end of the volume release period. more clearly : ______ \ \ \ ^^ volume slope of key 1 ____________ / / / ^^ volume slope of key 2 ______ ______________ | | | | | | ^^ filter envelope By triggering the second key before the first key has finished, it causes the filter to start sustaining straight away, thus sustaining on the _first_ key even as it releases. So perhaps setting the filter envelope release might get around this. Now, setting the sound. You're not going to get a realistic string sound out of the poly-800 - you'll get a more synth-y string sound, which still sounds nice i think. Set the octave to whatever you feel like - if you only want really high sounds, select 3 (high), but I think the middle setting is better (gives you high and low notes). On the foot settings (params 13-16), only have the 2' setting on (param 16). Set the volume of the oscillator to suit. Do not use double mode unless you really feel like it - for strings you want as much polyphony as possible, and double mode will only give you 4 notes which isn't enough for chords. Also make sure the noise generator is set to 0 as well. Real strings don't have noise generators :). For the filter, its simply a matter of setting to taste. Pushing the cut-off up will produce a more cutting sound, while a lower setting (around 40) with resonance about 7 will produce a much softer sound. Make sure the polarity (44) is set to 2. Keyboard track (43) should be 0 (off). EG int (45) is again set to personnal taste. Well, there you go. A basic guide to setting up a string type sound on a Poly-800.