OWNERS MANUAL

Owners Manual
Copyright 2004, by Way Out Ware, Inc. All rights reserved. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, from or to any form of media, without the prior written permission of Way Out Ware, Inc. Requests for permission to reproduce any part of this work should be addressed to : Way Out Ware, Inc., attn: Copyright Adminstration, info@wayoutware.com.

Table of Contents

1 The ARP 2600, and onward.. System Requirements, Installation, Configuration, Setup and Usage. 5
In this chapter you will find all of the platform-dependent information you need in order to install and operate your TimewARP 2600 software synthesizer.
3 The Craft of audio Synthesis. 11
This chapter is about the facts physical, mathematical, and auditory that make the TimewARP 2600, and the hardware that it emulates, possible. We have to spend a few minutes here distinguishing between physical signals, and the sounds that people hear in the presence of certain kinds of signals.
4 Modular Components of the TimewARP 2600. 27
In this chapter youll find detailed explanations of the TimewARP 2600s features and functions.
5 Patching the TimewARP 2600. Appendices. 47
Table of Alternate keyboard tunings compiled by Robert Rich

7 Index

The ARP 2600, and onward.
The ARP 2600 was the second product of ARP Instruments. It was released in 1970, and continued until the manufacturer ceased operations in 1981. Its design combined modularity (for studio flexibility, and for instructional use) and integration (for realtime performance). Functionally, the ARP 2600 was completely modular: any signal output could be routed to any signal input, with a patch cord. Operationally, the ARP 2600 was integrated, using internally-wired default signal paths that made it possible to create a wide range of keyboard patches by simply opening up slide attenuators, as though sitting in front of a mixing console. The ARP 2600 earned an early reputation for stability, and for flexibility exceeding that of its competitor the Minimoog. Used 2600s in good condition command premium prices on eBay today and businesses around the country can make a living reconditioning, rebuilding, and customizing 30year-old units. Among rock musicians, the ARP 2600 was used by Stevie Wonder, Pete Townsend, Joe Zawinul, Edgar Winter, Paul Bley, Roger Powell, Jean-Michel Jarre, Mike Oldfield, Herbie Hancock, and many, many others. Its modular design, and the popularity of its Owners Manual, made the ARP 2600 a widely used teaching instrument in many schools and music conservatories around the country and internationally. We are proud to bring you this software emulation of the 2600. Have fun with it, learn from it, but above all, make noise with it.

CHAPTER 1 - The ARP 2600

Foreword
Unfasten the seat belts of your mind. The TimewARP 2600 will be an astonishing, exhilarating, and enlightening experience. Creating this manual has been an astonishing, exhilarating, and enlightening experience for me. How many are ever given the chance to revisit an earlier life, an earlier project, a project like the ARP 2600 Manual, decades later, and get it right? Its time travel. Im grateful to Way Out Ware for providing me that opportunity. When, at Alan R. Pearlmans invitation, I began work on the original 2600 manual in September of 1970, the 2600 itself barely existed. For the first two months, I was writing blind - without a machine in front of me. My first hands-on experience with a synthesizer had been only six months earlier (it was a Putney VCS3). I finished the text in March of 1971, Margaret Friend created the graphics, and the Owners Manual for the ARP 2600 began what turned out to be a surprisingly long career. In spite of the many defects that my inexperience contributed the gaps in coverage, and outright errors - it became quite popular. To this day, it still gets an occasional respectable mention in the analog-synthesis community. When Way Out Wares Jim Heintz called, early in 2004, to tell me about the TimewARP 2600, a lot of time had passed. Regarding software synthesizers, I had grown weary and cynical. Analog-modeling software had been a decade-long disappointment; some products did interesting things but not the things that real analog modules do. Jim, however, had already encountered, and thought about, and solved, these problems. He owned a real 2600. He really aimed at getting it right and would not be satisfied with anything less. It was a pleasure, finally, to accept his invitation to do an Owners Manual. Its clear, now, that Way Out Ware has set a new standard for software-based audio synthesis. The behavior of the TimewARP 2600 software both module-by-module and integrated into patches - is effectively indistinguishable from that of the analog hardware that it emulates. Soaring and swooping through the free air of analog synthesis a world of nothing but sliders and cords and continuously evolving patch configuration - was a capstone course at the Boston School of Electronic Music in the 1970s. That is the world that the TimewARP 2600, for a new generation of musicians in a new millennium (that means you), provides access to: it is the first and I believe only - software synthesizer to support real-time performance by sliders and patchcords alone. So here it is: your new Owners Manual, for the new TimewARP 2600. Unfasten the seat belts of your mind. How else can you hope to experience time travel? How else can you enjoy free flight?

white noise has equal energy in any two equal frequency bands

18 20K

so its NOT equal in equal pitch intervals

1.25 2.5

5 10K 20K
octave intervals even when you use dB for the vertical axis

-9dB -6 -3

0dB -30dB
THIS is why white noise sounds so shrill
Steady-State Attributes: some sounds, once they get started, remain pretty constant. They dont change much while they continue, and when they stop, they just stop. We say such sounds reach a steady state, in which we can pick out a definite Pitch, Loudness, and Tone Color. Notes from organ pipes or from electronic drawbar organs are steady-state sounds. Time-varying Attributes (i.e. Envelopes): some sounds have a pretty clear start and end, but while theyre happening they change. Think of a bird song, or ordinary human speech. Such sounds can sometimes be analyzed as more or less rapidly changing in one or more of those fundamental auditory attributes: pitch, loudness, or tone-color. Think, for example, of how the sound of a plucked guitar string evolves from the moment you pick it to the moment you cant hear it anymore. It starts out loud and fades away; and it starts out bright - with lots of harmonics and is slowly muffled as it fades out. One way to conveniently diagram what happens in such events is to chart each changing attribute separately, in its own time-domain graph. Such a graph is often referred to as an envelope. Over the years, some standard vocabulary has developed for talking about envelopes: Attack (for the first part of an event), Decay (for some later parts), and Release (for the last part, when the event is coming to an end).
CHAPTER 3 - The Craft of audio Synthesis 3.4 How Signals and Sounds Go Together. Sort Of Signal activities, being entirely physical kinds of things, are easily nameable, measurable, catalogable, countable. Sounds, as we pointed out above, are not quite so easily domesticated. The music and other sounds that we listen to correlate in several wellestablished ways with the signals around us; but the correspondence is not simple. There are always surprises. 3.4.1 Signal Frequency and Audible Pitch This is probably the best-established and most reliable correspondence. It dates all the way back to Pythagoras, the Greek philosopher who 2500 years ago worked out that halving the length of a vibrating string made the pitch rise by one octave. 3.4.1.1 How Adding Pitches Means Multiplying Frequencies The range of audio frequencies of human hearing, in other words is conventionally stated to be 20Hz to 20KHz. And when you think linearly, it sounds tragic to learn of someone, say, who can only hear up to 10KHz. But in the realm of human pitch perception, such a person has kept 90% of his hearing range: from 10KHz to 20KHz is just one octave out of the 10 we hear.

Frequency in cycles / seconds
10KHz Each division is one octave. To add an octave, you have to multiply F x 2. 5KHz
This is a really important fact in audio synthesis; you will encounter it over and over again, in every 20Hz patch you create. It governs much 0 of the arithmetic of generating and controlling spectral distributions and patches:

625Hz 1250Hz

2500Hz

middle C

Pitch by octaves
4 Equal pitch intervals require exponentially increasing frequency increments; 4 Equal frequency increments make a harmonic series. In a harmonic series, as we go up the series we get smaller and smaller pitch intervals.
CHAPTER 3 - The Craft of audio Synthesis 3.4.2 Signal Amplitude and Audible Volume This is a very dicey relationship. It is true that, for any given signal, increasing its amplitude will increase the volume of the associated sound. But a lot of other signal characteristics have a greater impact on our perception of volume than amplitude does. For example, the difference between talking and shouting at someone is far more a matter of pitch and spectrum tone-color than of mere amplitude. When I yell, I raise my voice; that is, I raise the pitch of my voice, and I put more energy into it, which generates more harmonics, which is a matter of spectral content, not amplitude. Stage actors have to learn to overcome this tendency; in order to be heard onstage without shouting, they have to learn to increase their speaking amplitude without yelling. If you examine visually - the recorded signal from a pipe organ, or even an orchestra, you may be surprised to find the apparent amplitude of the softer signals almost equal to that of the loudest. This has to do with the spectral distribution of the sound. Of two different signals of approximately equal amplitude, the one with the broadest spectral distribution the most harmonic content - will sound louder. 3.4.2.1 How Adding Volumes Means Multiplying Amplitudes

Reference Power level

A 3dB rise or fall in sound level is noticeable, but it's not much; it represents a factor of 2 change in signal power.
Nonetheless, for a given signal and its spectrum, it is always true that a bare increase in amplitude will be heard as an increase in loudness. But how much of an increase? It turns out that, just as with frequency and pitch, the relationship here is exponential. A century of research has established that equal steps in loudness are represented by multiplicative ratios in signal amplitude. In other words, if you double the amplitude of a signal, you will hear an increase in loudness. But then to get another increase equal to the first one, you have to double the amplitude again.
decreasing Power level Power 2

Power 4 Power 8

-33 -30 -27 -24 -21 -18

-15dB -12dB

decreasing sound level
-6dB Ref-3dB Reference Sound level (any)
Signal Spectrum and Audible Tone-Color This is a quite solid relationship. In fact, the word spectrum has achieved a meaning in both worlds: depending on the context, it can refer to a measurable attribute of physical signals, or a character of perceived sounds. Any change you hear in the character of a sound in its tone-color or spectrum - must have an associated variation in the character of the physical signal that is arriving at your eardrums. Tone change = waveform change. The reverse is not quite so certain. Its fairly easy to find waveform changes that listeners cant hear. For example, we humans simply arent sensitive to phase relationships within a complex spectrum. But, in the time domain, two identical spectra with shifted phase relationships among their components can look unrecognizably different.
CHAPTER 3 - The Craft of audio Synthesis 3.4.4 Signal Envelopes and Audible Event-Contours One of the most fascinating areas of audio synthesis is listening for the envelopes of time-varying events. Here there are all sorts of mysteries, in which signal attributes get regularly misperceived: frequency variations get heard as volume, spectral evolutions are heard as pitch, and amplitude envelopes generate an elusive spectral twitter. 3.5 Modules and Methods for Generating Signals How can you get access to the signals and processes weve just described, so that you can play with them on your own? Throughout the past century people have been noodling around with electronic ways of generating audio signals. In particular, beginning in the 1960s, people such as Bob Moog, Don Buchla, and Alan R. Pearlman began to settle on some ideas that have become almost standard for audio synthesis: independent, modular functions for signal generation and signal processing, capable of being controlled not only by hand but also by signals of the same kind as they generate. This was the idea of voltage-controlled operation, and it was completely revolutionary. Using independent, modular functions made it possible to change one attribute of a signal without necessarily affecting any other attribute; so the craft of synthesis became the craft of constructing, and tuning, integrated configurations of modules. The cables that connected modules were called patchcords, and so connecting modules together came to referred to as patching them, and so, finally, any working configuration came to be called a patch. Lets take a look at some modules that generate signals. 3.5.1 Oscillators A device that repeats the same motion over and over is an oscillator. In audio synthesis, oscillators typically produce very simple geometrical-pattern signals such as sine, triangle, pulse, and sawtooth waves, named simply for what their time-domain graphs look like.

Sawtooth

f 2f 3f 4f 7 8f 16f
Oscillator Output TIME - domain
SPECTRUM FREQUENCY - domain -3dB / Octave

OdB -10 -20 OdB

f 2f 4f 8f 12f 16f OdB

Triangle

f 3f 5f

-10 -20 OdB

In an analog synthesizer, the 1/t underlying medium in motion t is usually electrical pressure, or 9 voltage. In a digital synthesizer, the signal is actually generated as a sequence of calculated numbers. (Conventionally, these are generated at the standard sampling rate for music CDs, 44.1KHz.) This does not become motion until, at the output of the synthesizer, the number stream is converted to variations in voltage, and then amplified, and then used to drive a loudspeaker. (A loudspeaker is a motor that moves back and forth instead of around and around.)

-10 -20

Because oscillator-generated signals are periodic, their spectral components always form a harmonic series. 19
CHAPTER 3 - The Craft of audio Synthesis 3.5.2

White noise spectrum

Noise Generators
A device that jiggles at random without ever repeating itself is a noise generator. Waterfalls, steam, wind, fans, and such things are all noise generators. The spectrum of a noise signal is a statistical distribution of frequency components. (This is the opposite of a sine wave, which is exactly one frequency.) A noise spectrum that is perfectly balanced throughout the musical range is called pink noise. Pink noise is very useful in listening tests of loudspeakers, because a trained human listener can hear even tiny differences between two different noise spectra. Filtering and equalization can shape a noise spectrum into almost any sound.
+15 Pink noise spectrum 0dB -15 0dB Red (LF) noise spectrum -15dB -30dB
Envelope Generators A device whose output is intended to control some time-varying attribute of an event is called an envelope generator. These are sometimes referred to as transient generators, to call attention to the fact that their output is not constant but transient. An envelope generator produces an output signal only on demand. The demand is made by means of timing signals called gates, and triggers.
envelope generators are controlled by gate and trigger signals.
trigger ADSR envelope AR envelope
Sample & Hold Processors The idea of sampling a signal does not directly relate to any particular characteristic of audio events; instead, it is an idea from electronics that has turned out to be useful for creating patterned control signals.

3.6 3.6.1

Modules and Methods for Processing/Modifying Signals Inverters A signal inverter works exactly like a seesaw. When the input goes high, the output goes low, and vice versa. An analog inverter would output negative voltages on positive input; a digital inverter simply multiplies its input number-stream by (-1).
CHAPTER 3 - The Craft of audio Synthesis 3.6.2 Signal Mixing A signal mixer adds two or more signals together and outputs the result of the addition. This is a more complex signal, usually, than any of the inputs. But not necessarily; if signal B, for example, is the exact inversion of signal A, then mixing the two will produce a signal of exactly zero.

Inverters

input output
these four signals added (mixed) together make.

ou tp ut

two sines added together
Attenuators/Amplifiers An attenuator cuts the strength of a signal passing through. Digitally, this is accomplished by multiplying the input by some value ranging from 0.0 (which passes no signal) to 1.0 (which passes the signal at its full input amplitude). An amplifier may increase the amplitude of a signal. But not necessarily; it is usual for a voltage-controlled amplifier to have a maximum gain factor of 1.0. In fact, the purpose of VCAs is actually to chop the signals passing through them. It would make more sense to think of them as voltage controlled attenuators.

3.6.3.1

Gain Factor To describe the behavior of an amplifier or attenuator, we may use the expression gain factor to mean the ratio of output signal amplitude to input amplitude.
Filters A filter is a device that works better at some frequencies than at others. (The inverters, mixers, and attenuators we have been describing work the same at all frequencies, so they are not filters.) Because of this frequency-dependent characteristic of filters, they change the shape of any complex waveform passing through. And so it will be important for you to get to know what filters do to signals in both the time domain and in the frequency domain.
CHAPTER 3 - The Craft of audio Synthesis 3.6.4.1 Low-Pass Filters

f | 4f | 8f | 16f 32f |

A lowpass filter in the time domain.
Any device or mechanism that passes along slower motions better than faster ones can act as a lowpass filter. Picture yourself stirring a cup of tea with one of those little wooden paddles they hand out in the coffee shops. Stir it back and forth, fast. Now slow down. Now imagine the tea has turned to syrup. You can still stir it slowly, but if you try to go fast the stick will simply not move. Thats a lowpass filter. You can see the effect of this on a signal quite easily.
. makes it difficult for the medium to change value.
For audio signals, you will usually be more interested in the frequency-domain effects of filtering. For subaudio signals, it is usually the time-domain effects changes in waveshape that we care about. In the time domain, a low-pass filter rounds off any sharp transitions in the signal. A good example of this is the lag processor described in section 3.6.4.5 below.

. and in the frequency domain, this cuts off higher-frequency components.
In the frequency domain, it weakens spectral components that are higher than the filter cutoff frequency the frequency at which the filter begins to have an effect on the signal. 3.6.4.2

High-Pass Filters

Any device or mechanism that passes along faster motions better than slower ones is a highpass filter. Take the drinking straw from your water glass. Seal the end of it with your thumb and dip it back into the water. Notice that you can pump it up and down in the glass, fast or slow, but the water never leaks into the straw as long as you hold your thumb over the end. Think of the up and down motion of the water at the bottom end of the straw as the signal. Now start letting a little air leak into the straw as you move it up and down. The water level at the bottom end of the straw no longer stays down when the straw signal goes down it starts to come up again. And then when you draw the straw back up, the water leaks back down more or less rapidly depending on the position of your thumb at the top of the straw.
A highpass filter makes it difficult for the medium to keep a new level - it "leaks" away.
. and so lowerfrequency components are weakened.
This is a high pass filter. In the time domain, it constantly leaks its output signal level back to zero, at a rate related to the cutoff frequency and slope. In the frequency domain, it passes all spectral components higher than the cutoff frequency, and attenuates those below the cutoff frequency by an amount proportional to the cutoff slope.
CHAPTER 3 - The Craft of audio Synthesis 3.6.4.3 Cutoff Slope This is the rate at which a filter attenuates spectral components, as a function of their frequency. It is usually a multiple of 6dB/octave. 3.6.4.4 Feedback and Resonance It is usually possible, with any signal-processing device or system of devices, to mix some of its output signal back into the input signal. This may be intentional, or it can happen by accident; everybody who has ever worked with a PA system has experienced the terrible screech of a system in feedback. In filter modules for audio synthesis, it is common to provide controllable feedback. With just a little, the filter response begins to peak around its cutoff frequency; as the feedback level increases, the peak gets stronger. Eventually just as happens with a PA system the filter falls into oscillation. Regardless of the input signal, it screams a sine-wave at its cutoff frequency. In this state it is no longer behaving as a filter at all; it has become an oscillator. Systems in feedback have been very well studied in physics. Their behavior can be described mathematically. Whereas feedback in a PA system can be unpredictable and uncontrollable, feedback in a filter module for audio synthesis can be and is a controllable and useful feature. 3.6.4.5 Lag Processors A lag processor is a low-pass filter intended specifically for processing subaudio signals. It introduces a lag in the output signal wherever the input shows a sharp change in value. How much of a lag depends on the cutoff frequency of the processor. 3.6.5 Modulation Methods The simplest possible signal, we said above, has exactly three attributes and no more: frequency, amplitude, and phase. If, starting from a steady-state signal, we systematically modify any of these characteristics, we are said to be modulating the signal. And so, based on these three signal attributes, there are three possible forms of modulation: Amplitude Modulation (AM), Frequency Modulation (FM), and Phase Modulation (PM). The first two are more commonly used in audio synthesis than the third; we wont say anything here about phase modulation. Using AM and FM methods, it is possible to generate waveforms and spectra that are far more complicated and interesting to the ear than anything that can be produced by merely mixing and filtering signals. Mainly that is because of sidebands.

CHAPTER 3 - The Craft of audio Synthesis 3.6.5.1 Sidebands and Sideband Spectra What happens to the spectrum of a sine wave when we modulate its amplitude? What happens when we modulate its frequency? Clearly, since the signal that results from AM or FM methods is no longer a plain vanilla sine wave, then, in the frequency domain, it must have some additional components. These additional components are called sidebands. They have been studied for at least a century, and are pretty well understood physically and mathematically. Audio synthesis is probably the only application of AM or FM modulation where we are interested in sidebands for their own sake, as something to listen to directly; in the past, this stuff was only interesting to radio engineers, radar, sonar, television broadcast engineering. In those disciplines, modulation sidebands are products of broadcasting methods, in the electromagnetic spectrum. Because of that historical background, some of the conventional language for talking about modulation processes is a little weird: the signal being modulated is often referred to as the carrier signal, and the signal that signal provides the modulation pattern is called the program. (Guess why.)
Given this for a carrier: carrier spectrum
. and this for a program program spectrum
here's the AM result: lower sideband
carrier weakend upper sideband
Any form of modulation generates, for each component of the original signal, at least one lower sideband at a frequency equal to the component minus the modulating frequency and at least one upper sideband, at a frequency equal to the component plus the modulating frequency. If the carrier the original signal - is itself complex, with multiple spectral components, then each of its components will produce its own sidebands independently of all the others. Likewise, if the modulating signal the program - is complex, the arithmetic applies separately to each of its components. So, just for example, if you modulate a 10-component carrier signal with a 10-component program signal, the signal resulting from the modulation will have not less than 100 spectral components. This can get very messy; the most useful thing you can do with such a signal, before you do anything else, is filter it to get rid of some of the fuzz.

Modular Components of the TimewARP 2600
Top Row Control Panel Buttons and Indicators Just above the panel graphics, outside the case of the TimewARP 2600, is a horizontal row of buttons and indicators for patch storage, import/export, voice-cloning, and other operations. These powerful features of the TimewARP 2600 have no equivalent in the world of analog synthesis; they are unique digital extensions of the original ARP 2600 synthesizer.
Patch Lock Button (Padlock Icon) This padlock button helps you to avoid accidentally overwriting your favorite patch. Basically it disables the Save button, while leaving the Save As button enabled. Under these conditions, you can save your current patch only by assigning it a new name.
Group, Category, and Patch Drop-Down Lists The TimewARP 2600 gives you a three-level hierarchy for storing and organizing your patches. All Patches are sorted into various Categories, which are in turn sorted into major Groups. Each of the three patch selection buttons generates a drop-down list associated with one layer in this hierarchy. Groups, Categories, and Patches can also be selected by keyboard shortcuts. The up/down arrow keys on the computer keyboard select Patches, the left/right arrow keys move between Categories, and using the control key with the left/right arrow keys moves between Groups.
Save Button. The Save button saves the current patch configuration and settings under the name of the most recently loaded patch. This button is disabled if the patch is locked (see 4.1.1 above).
CHAPTER 4 - Modular Components of the TimewARP 2600 4.1.4 Save As Button. The Save As button saves the current patch configuration and settings under a group, category, and patch name of your choice. Within the Save As dialog, you may create new groups and categories at will. There is no limit to the number of groups and categories you may create. 4.1.5 Patch Manager Button
Use the Patch Manager to organize and reorganize your patches; to move patches from one category to another, and to move whole categories from one group to another. Use it to export and import patch collections dozens or hundreds of patches at a time. The Patch Manager window displays all three levels of the patch hierarchy, and supplies a number of tools for managing the entire collection. These tools are listed in a column on the left of the window. To use them, select one or more items from the hierarchy, and then click on the operation you want to perform. Any operation that cannot be applied to the current selection of items will be disabled in the list.

CHAPTER 4 - Modular Components of the TimewARP 2600 4.1.7 Reset Button The Reset button removes all patch cords and returns all sliders to a standard position. 4.1.8 MIDI Indicator This virtual LED glows when there is any MIDI input to the TimewARP 2600 not just keystrokes, but also controller input and sysex dumps. 4.1.9 Output-Level Meter This shows the output signal level. If it reaches into the red segment, your signal will distort. 4.1.10 CPU Load Meter This meter shows, roughly, how much of the time between samples (the sample period) is being devoted to the TimewARP 2600 emulation process. In a complex patch, or a manyvoice polyphonic performance, the meter may indicate overload; when this happens, it is likely that the TimewARP 2600 output signal will be interrupted, so your audio feed will develop a glitch. To avoid this, you will have to simplify your patch, or decrease the number of voices, or acquire a faster, more capable computer. 4.1.11 The Magic Logo At the lower right of the main panel is the TimewARP 2600 Logo. Clicking on this brings up a menu:

4.1.11.1 4.1.11.2

About TimewARP 2600 identifies the team; the people who worked together to bring you this software.
Load/Save MIDI Maps Use the Load/Save MIDI Maps commands to save and reload the MIDI-controller to slider assignments that you set up. In the TimewARP 2600, these are global assignments, independent of any particular patch settings; saving a patch does not save these assignments, and loading a patch does not change the current assignments. You can, if you want, set your mappings once, and they will be there throughout all of your personal patch changes.
CHAPTER 4 - Modular Components of the TimewARP 2600 4.1.11.3 Load Microtuning You may also load alternate tunings for the keyboard. These are described in Appendix 6.1. The TimewARP 2600 does not allow you to modify these tunings or to save or create new ones. Microtunings are a global attribute of the keyboard; once loaded, the tuning will govern anything you play until you load a different one, regardless of your patch changes
4.1.11.4 MIDI Beat Synchronization You may synchronize the Internal Clock (IC) (see section 4.13), to the MIDI Beat Clock (MBC) by specifying the number of MBC pulses per IC transition. As a reference, there are 24 MBC pulses per quarter note. The keyboard LFO (see section 4.14.1) may also be synchronized to incoming the MBC, independently of the Internal Clock. Setting different sync counts for these is a fun way to program complex rhythms that are locked to the tempo of your MIDI tracks. In order for the MBC messages to be sent to the TimewARP 2600, you must enable MIDI Beat Clock in the Pro Tools MIDI Menu, and select the TimewARP 2600 as a recipient of these messages. Also, MBC messages are only sent when the Pro Tools transport control is running. MBC synchronization is a patch attribute, not a global one; the sync counts you set here will be stored with the current patch when you save it.

VCO 2 VCO 2 generates sine, triangle, sawtooth, and pulse outputs. A pulse-width slide control can adjust the duty cycle from 10% to 90%; at the middle of its travel, the pulse width is 50%, that is, a square wave. The default signal to the first (unattenuated) FM Control input is from the keyboard. The Audio/LF switch above this input switches the mode of the VCO from Audio (10Hz - 20,000Hz) to LFO Mode (0.03Hz 30Hz). When the VCO is in LFO Mode, the default connection to the keyboard is removed. This can be overridden in this mode by patching a cable to the Keyboard CV output on the left side of the front panel. The default signals to the next three FM Control inputs are from a) the Sample & Hold, b) the ADSR Envelope Generator, and c) VCO1 square. There is a fourth attenuator-governed input, for digital control of the pulse width. The default signal at this PWM input is from the Noise Generator.
CHAPTER 4 - Modular Components of the TimewARP 2600 4.3.3 VCO 3 VCO 3 generates sawtooth, pulse, and sine outputs; the pulse width is manually variable. The sine output is a TimewARP 2600 extension; the original ARP 2600 VCO3 provided just sawtooth and pulse outputs. The default signal to the first (unattenuated) FM Control input is from the keyboard. The Audio/LF switch above this input switches the mode of the VCO from Audio (10Hz - 20,000Hz) to LFO Mode (0.03Hz 30Hz). When the VCO is in LFO Mode, the default connection to the keyboard is removed. This can be overridden in this mode by patching a cable to the Keyboard CV output on the left side of the front panel. The default signals to the next three FM Control inputs are from a) the Noise Generator, b) the ADSR Envelope Generator, and c) VCO2 sine.
Voltage Controlled Filter (VCF) The Voltage Controlled Filter has variable cutoff frequency (Fc) and resonance (Q). The response below Fc is flat down to DC; above Fc the response falls off at 24Db per octave. Fc range is from 10Hz to 10KHz without control voltages; under voltage control, Fc can be driven as far down as 1 Hz and as high as 20KHz. Fc is controlled manually by a coarse tuning slider (labeled initial filter frequency) and a fine tune slider. Fc may also be controlled by external voltages; the sensitivity under voltage control is 1.0vV/oct. The Q, or resonance, of the filter circuit is controlled by a single manual slider. As the Q is increased by moving this slider from left to right, the response below Fc is gradually attenuated until a sharp peak remains at the cutoff frequency. (Gain at Fc is always unity.) At this Q setting, just below the point at which oscillation begins, the filter will ring distinctly in response to any sharply defined pulse presented to its signal input. In this state it is effectively analogous to a highly resonant physical system, and may be used for various percussion effects depending on its resonant frequency (identical to Fc) and on the impulse spectrum exciting it. As the Q is raised still higher, beyond about the halfway point in the slider travel, the filter will oscillate. Operating in this state, it generates a pure sine wave. even in the absence of any signal input. The VCF has five Audio signal inputs. They are fed through logarithmic attenuators to a summing point, and then to the VCF itself. The default input signals are from the Ring Modulator, VCO-1 Square, VCO-2 Pulse, VCO-3 Sawtooth, and Noise Generator. The VCF has three frequency Control inputs. The first is normally from the Keyboard pitchcontrol. The slider that governs this input is a TimewARP 2600 extension; on the original

CHAPTER 4 - Modular Components of the TimewARP 2600 4.9 Ring Modulator The Ring Modulator is essentially a multiplier; from its two inputs A and B it produces the output function A x B / 5. The kind of transformation this effects on input signals depends to a large extent on what they are and on whether the modulator is AC or DC coupled to them. This is selected by the Audio/DC switch at the bottom of the modulator. When the inputs are AC coupled (Audio position of the switch), any DC component ( present in them is canceled before they are fed to the modulator. Thus a sawtooth that starts from zero and goes to+l0vV will instead start at -5vV and move to +5vV so that its overall positive and negative deviation cancels to zero. Under these conditions the modulator will generate from any two periodic signals an output signal consisting of the sum and difference frequencies that can be generated from the frequencies of the two inputs. The input frequencies themselves will be suppressed. If both signals are audio-frequency, a large variety of harmonic and inharmonic timbres can be produced from the modulator, depending on the ratio of the input frequencies and on their own harmonic content. If A is a sine wave and we represent its frequency by Fa, and B is a complex waveform of frequency Fb with overtones 2Fb, 3Fb, 4Fb, etc., then the output of the modulator will be a complex waveform with frequency components Fb + Fa, Fb - Fa, 2FbFa, 3Fb Fa, 4Fb Fa, etc. A moments experimentation with the prewired sawtooth and sine inputs to the modulator will demonstrate the complexity of the timbres that can be generated by this simple means. If, still with AC coupling, one input is subsonic and the other at some audio frequency, there will be an output from the modulator only when the value of the subsonic input is changing, and the output will be roughly proportional to the rate of change. If, for example, the subsonic input is a square wave, the modulator output will be a series of short, decaying tonebursts one at each rise or fall in the input signal. When the inputs are DC coupled, any DC component in either one of the inputs will pass into the modulator and affect the modulating process. The effect when both inputs are at audio frequency is to allow into the output waveform some of the input frequencies in addition to the sum and difference frequencies. The effect when one of the inputs is subsonic is that the modulator operates as a voltage-controlled amplifier: the output amplitude will be in direct proportion to the instantaneous amplitude of the low-frequency input and will vary as its absolute value varies. Also, the output phase will reverse when the low-frequency input signal changes from positive to negative or vice versa. The AC-coupling time constants are 235 msec, for the left input and 90 msec, for the right input.