Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

Understanding the DX7 (Part 2)

The second part of Jay Chapman's guided tour through the Yamaha DX labyrinth, with an in-depth analysis of one of the factory-programmed voices.

Jay Chapman begins his in-depth look at how the DX keyboards generate their sounds by examining one of the factory-programmed voices — 'E. Organ 1'.

This article is the first of two in which I'll be analysing the makeup of two of the Yamaha-supplied pre-programmed voices. This will not be done in exhaustive detail - I'm not going to look at every single parameter setting - for several reasons, not the least of which being that we would probably all fall asleep after a couple of paragraphs!

The object of the exercise is really to pick out and discuss the important settings and the workings of the algorithm in use. In other words, the parts that give each voice its essential character.

I hope you find this approach a fairly logical one - if it's successful you should soon be able to explore the machine under your own steam, of which more in a later issue.

Analysis - 'E. ORGAN 1'

Voice ROM1 A17

If you have never downloaded any of the ROM voices into the DX7 internal memory (thereby overwriting the latter) you can simply press the green INTERNAL keypad (under the MEMORY SELECT legend) followed by the green '17' keypad, to call up this particular voice. Check the display:

If you have overwritten this internal memory, press the green CARTRIDGE keypad instead of the INTERNAL one (having first set and inserted the correct ROM) and the green '17' keypad and you should end up with:

Having got to the correct voice we now want to examine its internals, so press the purple EDIT/COMPARE keypad to get into edit mode. Your display will change but exactly what it shows depends on what keypads you pressed last time you were in this mode. The information regarding the number of the algorithm used and which Operators are switched on will be present, though, so that's where we start the analysis - part of the top line of the display shows:

The algorithm used then is number 32 - the diagram for this algorithm, taken from the DX7 front panel, is shown in Figure 1.

Figure 1. Pedalboard circuit diagram.

Straight away we can see that this voice does not use any frequency modulation, as all six operators connect directly to the horizontal line (below the operators) which represents the output. The string of '1's in the display tells us which operators are actually in use and as we have a complete string of ones, ie. '111111', we know that all six are turned on. In fact we have six 'carrier' operators - they all contribute directly to the sound we hear.

OK, lets start some experimental keypad pushing.


The first thing to do is examine the sound that each carrier is producing. To do this, turn all the operators off by pressing the green keypads '1' through '6'. The display should change to '000000'. Now turn on one carrier at a time by pressing the green keypad bearing the relevant Operator's number, and play a few notes. When you want to move onto the next Operator, press the same keypad again to 'toggle' the current Operator off. Your display of ones and zeros should cycle through the sequence '100000', '010000', '001000', '000100', '000010' and '000001' as you listen to each Operator in turn.

With the exception of Operator 6, you should hear a simple sinewave, of a different pitch for each Operator, which is turned on and off by the keyboard acting simply as a switch, with no hint of any envelope shaping. Thinking back to the 'drawbar organ' discussion in last month's introduction this state of affairs is not at all surprising, but don't be fooled, there is a little subtlety to be uncovered yet.

We know already that to create the timbre of a sound we must use frequency modulation and/or combine frequencies that are (usually) harmonically related. Since we do not have any modulating operators in use in this instance, we should look for what frequencies are being combined. To do this, simply press the green FREQUENCY COARSE keypad and turn all the Operators back on ('111111'). The display will not show the actual frequency of the Operator unless the operator is in Fixed Frequency mode (see last month's article). For most 'musical' sounds, Frequency Ratio mode is chosen so that the pitch output by the Operator is related to the keyboard note played and thereby indirectly to the frequency of other Operators also in Frequency Ratio mode - this is the case for all the Operators used in 'E. ORGAN 1'.


By pressing the purple OPERATOR SELECT keypad you can cycle through all the Operators displaying their frequency ratios. This is what you should find:

It's always nice to start from a known base during investigations, so you'll be pleased to see the '1.00' for Operator 5. This means that the pitch produced by Operator 5 is exactly that selected by the keyboard note played. The ratio of 1.00 is usually, but not always, going to represent the fundamental pitch of the overall sound being produced.

Of the other Operators, number 3 is probably the easiest to consider next. The frequency ratio of 1.50 means that the pitch produced by this Operator is one-fifth above that selected by the keyboard note being played. To hear this, turn off all Operators except number 5 ('000010'), play middle C, turn on Operator 3 ('001010') and play middle C again. You should hear middle C and the G above middle C (this being the musical fifth above) at equal volume.

With Operators 3 and 5 on and all the others off, the sound produced has a slightly oriental flavour to it because the interval of a fifth is a characteristic of such music. If you feel brave enough, have a look at the voice 'PIANO 5THS' (ROM 1 B6) - you should now be able to work out how this voice produces the fifths.

What next? Well, let's consider Operator 1. Its frequency ratio is 0.50, so it is producing a pitch one half that played on the keyboard: in other words, we get a note one octave below that played. If you want to hear the difference, switch operator 5 on ('000010') and 'toggle' Operator 1 on and off whilst playing the same note ('100010' then '000010'). It's just like using the 'sub-octave' facility on a cheap synth that 'fattens up' the sound a little without the extra cost of another oscillator.

By this time you should be getting a bit suspicious about Operator 4, which also has a frequency ratio of 0.50. Isn't it just being wasted duplicating Operator 1? The answer relates to the subtleties referred to earlier. To explain what is going on we must look more carefully at the tuning of Operators 1 and 4.

As far as the FREQUENCY COARSE and FREQUENCY FINE parameters of the two Operators are concerned, both of these produce the same pitch because their frequency ratios are identical. There is, however, a last 'fine tuning' parameter we must consider - this can be displayed by pressing the green DETUNE keypad. Turn all the Operators except 1 and 4 off and use the purple OPERATOR SELECT keypad to select each of the two Operators in turn when the display will be either:


The difference in pitch produced over the -7 to +7 range of the DETUNE parameter is small and is, in fact, quite difficult to hear. To convince yourself there is a difference, try just Operator 1 on its own and play the C above middle C while (quickly) varying the DETUNE value between -7 and +7 using the Data Entry slider. This feature is also to be found on two- (or more) oscillator synths where slightly detuning the oscillators causes beating due to phase effects. Beating can be used both to 'fatten up' the sound and, in the case of 'E. ORGAN 1', manages to simulate something of the feel of the Doppler effect given by a Leslie cabinet. To hear the beating, get back to having just Operators 1 and 4 on and play a note. At the bass end of the keyboard the beat frequency is about ⅓Hz, rising to about 1 Hz at the treble end.

If you look back to the frequency ratio of Operator 2, you will see that at 1.01 it is very close to Operator 5 at 1.00. Since the difference in tuning is greater than that between Operators 1 and 4 the beat frequency is slightly higher, but the idea is the same. Note that the difference is not as great as it looks at first, as Operator 2 is detuned down by -6 starting from 1.01 and Operator 5 is detuning up by +2 from 1.00.

Hammond Sound

As we have two pairs of Operators beating in this manner, we have two low frequency waves weaving in and out of each other which makes the overall effect less clinical and more ethereal (very subjective comment that last one!). As a last point on this detuning business, note that all the Operators are detuned slightly relative to each other, thus heading toward that good old Hammond tone-wheel sound, albeit from a completely different angle of approach!

The last, but by no means least, item on the agenda for 'E. ORGAN 1' is Operator 6 which has a very important job to do. Turn all the Operators except 6 off and play a note on the keyboard. You will hear a short burst of sound which is fairly percussive. The pitch of the sound is one octave and a fifth above the keyboard pitch played (frequency ratio 3.00) and it is the combination of percussiveness and a more piercing pitch that adds some 'bite' to 'E. ORGAN 1', emulating the famous Hammond 'key-click'.

The percussive effect is governed by the Operator's envelope generator which we will look at next. Make sure Operator 6 is turned on and selected, then press the RATE keypad under the EG legend (the keypad is green and has '21' on it). If you have just Operator 6 on, your display should look like:

Keep pressing the RATE keypad until you've seen the four rate values. Do the same with the LEVEL keypad just to the right (the word RATE in the display will be replaced by LEVEL) to see the four level values. The values are:

Before you dive off into drawing a fancy eight-parameter envelope diagram a la the DX7 front panel, note that level values 3 and 4 are the same as level value 2 at zero, and their rate values are therefore meaningless (this is a simplification but it really is true enough in this case). The envelope is very simple and can be seen in Figure 2.

Figure 2.

It's not too difficult to see why Operator 6's output comes and goes rather sharpish!

Well, that's quite enough time spent on just one voice... next month we'll move on to something a little more complex that shows off more of the DX7 features.

Readers may be interested to know that an introductory booklet on the DX synthesisers entitled 'Getting Started' and compiled by Dave Bristow is available free of charge from Yamaha. Just write to Martin Tenant, Yamaha Musical Instruments, (Contact Details), enclosing a stamped addressed envelope.

Series - "Understanding the DX7"

Read the next part in this series:

All parts in this series:

Part 1 | Part 2 (Viewing) | Part 3 | Part 4 | Part 5 | Part 6 | Part 7

More with this topic

Browse by Topic:

Synthesis & Sound Design

Also featuring gear in this article

BeeBMIDI (Part 3)
(EMM Aug 84)

BeeBMIDI (Part 7)
(EMM Mar 85)

Load Baring
(12T May 85)

Load Baring
(12T Aug 85)

Yamaha DX7
(12T Nov 83)

Yamaha DX7
(ES Nov 83)

(EMM Feb 84)

(EMM Mar 84)

(EMM Apr 84)

(EMM May 84)

(EMM Jul 84)

(EMM Aug 84)

(EMM Jan 85)

(EMM Feb 85)

(EMM Apr 85)

(EMM Jun 85)

(EMM Jul 85)

(EMM Feb 86)

(EMM Mar 86)

(EMM May 86)

(EMM Jun 86)

(EMM Aug 86)

(EMM Sep 86)

(MT Nov 86)

(MT Dec 86)

(MT Jan 87)

...and 19 more Patchwork articles... (Show these)

Browse category: Synthesizer > Yamaha

Featuring related gear

Previous Article in this issue


Next article in this issue

Modular Synthesis

Electronics & Music Maker - Copyright: Music Maker Publications (UK), Future Publishing.


Electronics & Music Maker - May 1984


Synthesis & Sound Design


Understanding the DX7

Part 1 | Part 2 (Viewing) | Part 3 | Part 4 | Part 5 | Part 6 | Part 7

Gear in this article:

Synthesizer > Yamaha > DX7

Gear Tags:

Digital Synth
FM Synth
FM 6-Operator

Feature by Jay Chapman

Previous article in this issue:

> Patchwork

Next article in this issue:

> Modular Synthesis

Help Support The Things You Love

mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.

If you value this resource, you can support this project - it really helps!

Donations for June 2022
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £49.00

All donations and support are gratefully appreciated - thank you.

Magazines Needed - Can You Help?

Do you have any of these magazine issues?

> See all issues we need

If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!

Please Contribute to mu:zines by supplying magazines, scanning or donating funds. Thanks!

Monetary donations go towards site running costs, and the occasional coffee for me if there's anything left over!

Small Print

Terms of usePrivacy