The Prophet '08 Among Prophets

[tumble2k]

I often use VCA velocity amount on the Prophet08. If I do that, I have to dial back the VCA env amt, because its initially at 127. There's no higher value possible, so the VCA vel amt can't add anything to it. This is why I always use the inverse value (e.g. vel amt 64 -> env amt 63 | vel amt 80 -> env amt 47), if I want to retain volume but not cutting off high velocity values.
I also, like you said, dial back VCA env amt, if my patch seems to distort, but it seems to me that lowering the overall patch volume (misc parameters -> voice volume) is achieving the same result.

I hadn't experimented with this the VCA velocity amount, so this information is very useful. What you say makes perfect sense now. Also I didn't know about the voice volume parameter. I suppose the VCA Env Amount is nice because I can twiddle it using one knob instead of having to select the parameter with another knob.

Jason, the audio mod parameter is a type of FM synthesis where the actual output of oscillator 1 modulates the frequency of the filter. If you have the resonance way up you can get real FM sounds, but with the resonance lower audio modulation just adds some more higher harmonics to the filtered sound. Mephistofeles suggested using it to brighten up the filter when the cutoff is set really low.

[Jason]

I believe you can create a modulation with the key number as the source and an individual LFO frequency or all LFO frequencies as the destination.

Yes, that's it! Thank you. That's exactly what I was looking for, and playing with it gave me a better understanding of how the modulation can affect the LFO's. My string sounds are definitely sounding better.

I also, like you said, dial back VCA env amt, if my patch seems to distort, but it seems to me that lowering the overall patch volume (misc parameters -> voice volume) is achieving the same result.

I have to say that that is also what I found when experimenting with it. It's nice to have the ability to take the volume down with one knob, but both adjustments seemed to give the same effect for me. So I feel like I have to either live with the slight distortion or turn down the volume of the patch.

I haven't been able to get good results from using the Random LFO. Mephistofeles says that the rate is best between 70-100 IIRC, but I still hear the oscillators burbling. Higher LFO rates did not help this problem. I think this is what you're saying doesn't give good results.

.

Yes, that is what I was doing, and "burbling" is a good word for it. But I did have better results today with the suggestion of keeping the rate between 70-100. Although it didn't improve any of the patches that I tried it with today, depending on what I am going for, I think that could definitely be useful. I honestly can't say that I liked the other method with a Mod source of noise any better. It seems to give a low level of distortion, which I can't say I liked. In the end, I think Slop is a bit more useful for me, or at least for the patches that I was working with today. But I want to keep experimenting with both of these options.

Jason, the audio mod parameter is a type of FM synthesis where the actual output of oscillator 1 modulates the frequency of the filter. If you have the resonance way up you can get real FM sounds, but with the resonance lower audio modulation just adds some more higher harmonics to the filtered sound. Mephistofeles suggested using it to brighten up the filter when the cutoff is set really low.

Thank you. That's probably the best description of Audio Mod that I've seen, and it gives me a better idea of how to use it. I did notice today how it added some highs to the filter (I was using pretty low resonance.) Did Mephistofeles or anyone else offer a range that is most helpful?

[tumble2k]

I did notice today how it added some highs to the filter (I was using pretty low resonance.) Did Mephistofeles or anyone else offer a range that is most helpful?

Actually I had Mephistofele's instructions wrong. Here they are from another forum, but I don't know which one. He is responding to some bellyaching about the Prophet '08's sound.

-------------------------------------------
See my comment above.

I thought all of these at one point but it can actually sound quite different with the right techniques.

You need to do 2 things:

- "Destabilise" the oscillators.
- Find the lower rich filter sound.

Use the LFOs to "wobble" the oscillators, experiment with different waves, speeds etc. I tend to keep the speed fairly high (60-80) and level very low (low single figures). 

If you are using a PWM based sound do the same at the same time for an even bigger sound. 

Another trick is to route noise into the filter or oscillator, gives it a strange grainy quality. 

The filter thickens up nicely at the low end. But to get it low enough you need to know that the other controls (velocity, envelope, keyboard) also effect the cutoff. You need to turn these all down to get it low. Also make sure you're using the 4 pole filter. 

Once you have it low turn up the resonance so it's not quite self-oscillating. You will probably have a very rich but somewhat dull sound. 

Add in the filter mod, this adds in higher harmonics allowing you to some brightness even with a low cutoff. 

After this it's just tweaking to get a much more vintage type sound.
Try it. You'll probably be very surprised at the sorts of sounds you can get - I certainly was.


BTW These techniques should apply to all of DSI's synths including the Evolver and Tempest.

[Sacred Synthesis]

The view that destabilizing the oscillators and filter produces a vintage sound still mystifies me.  I'm never striving to create a vintage sound, and the last thing I want is a sound that is destabilized anyways.  Nor would I ever attribute that characteristic to the essence of a fine synthesizer sound.  But if some one can post an audio sample of the above sound, I'm all ears.

[MartinM]

Moinmoin,

IMHO the most interesting - if not the best - sounds from an electric guitar are achieved, just before "something really bad" happens: feedback, screamig, collapsing, You name it.
I found out that this is also true for synthesizer sounds. Parameters may be different, the principle remains: For a simple example there is no acoustical feedback, but filter resonance, electrical feedback, or other methods to achieve the same kind and degree of "danger" 

In these situations, the musician walks the small line between "wow" and "aaaargh", which requires direct control and fast reaction. This BTW is the reason why I prefer analog synthesizers with a lot of knobs, as they give instant and direct access (analog) without fingering through menus (knobs), while sound, ears and minds go berserk...

If this is meant with destabilisation, I do agree. Otherwise I am with Sacred Synthesis. "I'm all ears" shall be the whole of the law for any kind of music(ian) anyway.

Martin

[tumble2k]

I don't have much sound design experience but when I've listened to pure sawtooth waves with the filter wide open on different synthesizers I hear differences. For example, on a Moog Thin Fatty, the sawtooth is very pleasing on its own. Contrast that with a single sawtooth on the Prophet '08. It sounds flat and static by comparison--like a signal generator in an electronics lab. An acoustic instrument never sounds static like this, one reason why sampling is such a fine art. Perhaps that is why Marc Doty recommended modulating the frequency of the oscillators when he created his video about making the Pro 2 sound "vintage"? I'm still trying to figure this out.

Jason affirmed my observation that modifying the frequency using a random LFO was plainly audible, and in a detrimental way. I suspect this is because the modulation amount of 1 is still too high to get the required variation without sounding like an artifact (that fine line between "wow" and "aaaargh" that Martin refers to). I found that modulating the frequency with noise was better, but it's still a little too obvious, especially if you're listening to the plain oscillator.

I suspect (but I don't know for sure) that if I could set the modulation amount lower (I can't find a way to do this because the logic of Prophet '08's modulation system seems to disallow finer adjustments) I might be able to get the oscillator to sound pure and yet not have that flat and static sound.

Another thing I suspect is that the stability on a very fine level can affect how we subconsciously react to the oscillators even if we can't consciously put our fingers on it, especially after mixing with another oscillator, filtering, and enveloping.

[Jason]

There are probably advantages to having access to both very precise oscillators (like we have in the '08) and also oscillators that seem to have more natural movement, which is what I think I'm hearing in videos of some older instruments. For example, in the Porcaro video that I mentioned, he starts with a single oscillator starting point much like our Basic Patch, but I hear a more interesting sound coming out of his Oberheim Xpander than what we get with a single, unaffected oscillator on a '08. It could be that there are effects on the single oscillator demo that aren't being mentioned, and it could be that I'm hearing things (it's a copy of an old VHS tape). I think I also hear more movement in some of the videos of the OB-6 (https://www.youtube.com/watch?v=t81GYNGqO48). If there's a difference, I think there has to be applications for both; sometimes precision is better.

Thanks tumble2k, for taking the time to find and post the Mephistofele quote. I notice that he didn't specifically mention using the random wave of an LFO, although he suggests experimenting with different wave shapes. I'm interested in experimenting with adding trace amounts of other wave shapes to the oscillators, as well as trying his other suggestions.

[Sacred Synthesis]

Perhaps that is why Marc Doty recommended modulating the frequency of the oscillators when he created his video about making the Pro 2 sound "vintage"? I'm still trying to figure this out.

Yes, Doty's first Pro 2 video is representative of this view, and I would cite that demo as proof - in my opinion - that the destabilizing achieves nothing but destabilizing.  I thought the end result on the Pro 2 sounded worse than the "perfect" sound he started with.  It's like claiming that a good singing voice consists of the inability to create a smooth controlled vibrato.  Many singers have no, little, or an uncontrolled vibrato, and yet, their voices still sound excellent in spite of the fact.  But it's an entirely different matter to believe that, if you wreck your own vibrato, then as a direct result you'll have an excellent singing voice.

So, I'll have to part ways with the popular view here, and say that I'm never striving to destabilize the Prophet '08's sound; no, just the opposite.  If I add a quality to a Prophet '08 patch that sounds somewhat randomly modulated, it's actually an effect other than destabilization that I'm after.

I would agree that, if you play one oscillator of the Prophet '08, using a sawtooth wave form with a wide open filter and no modulation, the result is less than satisfying; it certainly does sound thin and rather unmusical.  But this is merely to admit that the P'08 is capable of creating a musically sterile sound.  Surprise, surprise!  It's an electronic gadget, so don't be disappointed if it can, at points, sound like a glorified fire alarm.  To me, it's not a concern that a single sawtooth with a bright filter setting sounds bad.  In my opinion, such a patch sounds bad on a Minimoog as well.  Personally, I would never use such a sound, so it doesn't concern me.  This leads to an entirely new topic, however - that of creating a nice warm and rich musical tone on the Prophet '08.

I think a single P'08 sawtooth with a slow attack (72), short release (32), soft to medium filter setting, and a generous amount of reverb, truly sounds beautiful.  It resembles the diapason stop on a pipe organ.  As a means of slightly rounding-off the tone, you can choose instead the sawtooth-triangle wave form - an option I've never seen discussed on any forum.  It has a character all its own, especially as you rotate through the cut off frequency.  Using the 2-pole filter setting, even with full resonance, will give yet another array of pleasantly music characteristics, even though this will attenuate some of the lower frequencies.  Another option is to add a slight amount of volume from the second oscillator - just enough to add a faint slow oscillator beating.  If you combine the first oscillator's sawtooth with, say, the second's square, or even slight PWM over a selected range, then you open up a domain of very subtle tonal characters.  This is an area in which I spend quite a bit of time - this searching for subtle differences from solo patch to solo patch.  I use a number of melodic tones that differ so slightly, one from the other, due to the combinations of pulse widths, sawtooths, and triangles.  Still another option for creating warm mono patches is to use the Output B or Layer options to double the identical patch.  You're still using the same sound, but panning it or differently processing it.  Don't underestimate the effectiveness of using substantially different amounts of reverb to each layer.  Using a patch with a moderate amount of reverb, combined with the identical patch having a large amount of reverb and adjusted down to a soft volume, can create wonderfully ethereal effects, rather like a natural echo.

Although the Prophet '08 can sound thin and sterile if you want it to, this is an advantage.  The question is, can it only sound thin and sterile in producing mono patches?  I would obviously answer, "By no means".  The approach I follow in designing musical mono patches never uses random destabilizing, but only vibrato or other forms of regulated modulation.  If I use the P'08's rather crude FM, it's for reasons other than modulation and more related to timbre.

[dslsynth]

Forum user "snowcrash" once wrote the following on the old forum:

I had lengthly discussions with people who are professionals in analog and digital signal processing and one guy mentioned, that the older synths have a kind of pink noise (or even darker) in the pitch-CV of the OSCs. So basically each OSC has a slightly different dark noise modulation (much different than what you can try to achieve with a single digital noise source for all OSCs) that makes the harmonics seem to sound "broader" and thus "fatter" if you want to stress that terminology.

So the idea with noise modulation is simply to try and mimic the older instruments in some digital way. Wonder if using lowpass filtered noise would make any difference? At least its a technique that deserves more exploration even though it may not work for everyone!

[Sacred Synthesis]

By the way, this thread shows the advantages of having instruments that have been around for a while.  The good old Prophet '08 has been fully vetted and its idiosyncrasies substantially fixed.  Now, it is what it is, and no changes will be made to it.  As a result, this thread is a decent mature discussion about effectively using the instrument, with no comments or complaints about software, bugs, updates, and so on.  It's so refreshing to be at this stage with a synthesizer.  My point is, it seems to be a luxury to reach this state in an instrument's life, when talk of changing it is long past, and only talk of using it is of interest.  I wish the Poly Evolver were still in production for the same reason. 

[Jason]

So, I'll have to part ways with the popular view here, and say that I'm never striving to destabilize the Prophet '08's sound; no, just the opposite.

I'm not sure that destabilizing an oscillator is a popular view. I would bet that most of us are going for similar qualities regarding sound design. (I, for one, consider the sounds that you are able to get to be the high watermark.) I think the clearest difference here is that you have probably experimented with and abandoned these ideas long ago, whereas, I have only experimented with them briefly and recently. As I said, my results weren't favorable, but there are still several things I would like to experiment with... including the suggestions you give in the second part of your email.

[Sacred Synthesis]

So, I'll have to part ways with the popular view here, and say that I'm never striving to destabilize the Prophet '08's sound; no, just the opposite.

I'm not sure that destabilizing an oscillator is a popular view. I would bet that most of us are going for similar qualities regarding sound design. (I, for one, consider the sounds that you are able to get to be the high watermark.) I think the clearest difference here is that you have probably experimented with and abandoned these ideas long ago, whereas, I have only experimented with them briefly and recently. As I said, my results weren't favorable, but there are still several things I would like to experiment with... including the suggestions you give in the second part of your email.

Jason -

I meant "popular view" regarding this forum in general and beyond it, not just regarding this single thread.  Related to the Pro 2, several of us have gone back and forth quite a bit about the theory that digital oscillators can be made to sound more analog simply by destabilizing them.  I disagree with this, while most others - such as Doty - agree with it.  I think the character and quality of an oscillator comes down to the fundamental and partials that comprise it, as well as the filter that attenuates it.  These two elements give you that starting tone which already has a personality.  All the other parameters only modify it.

I think this discussion should have some audio samples.  Let some one provide a sterile sawtooth from one P'08 oscillator, with the filter wide open.  I think we would agree this will sound sterile.  But leaving the filter as it is, then provide a sample of it destabilized by random modulation.  I say the result will be the same as Doty's - a sterile sawtooth that has been destabilized by random modulation, and not a tone that now sounds warm, rich, musical, or greatly improved.

[tumble2k]

I say the result will be the same as Doty's - a sterile sawtooth that has been destabilized by random modulation, and not a tone that now sounds warm, rich, musical, or greatly improved.

In fact you're quite right, the result IS a sterile sawtooth that has been destabilized by random modulation.

I agree with Jason. You're experimenting with some different and powerful approaches that I'd like to try out as well, especially adding small amounts of the second oscillator slightly detuned. Thank you for the explanation.

At the same time I'm interested in the idea that applying low pass filtered noise to the oscillator may broaden the spectrum in a pleasing way -- perhaps setting up two layers and sending layer B to the CV input can simulate this method (on one oscillator only). Very interesting!

[Sacred Synthesis]

Oh, I'm not suggesting that anything and everything shouldn't be tried.  Indeed, the Prophet '08 is a relatively simple instrument, and so, we should strive to get the most out of what is there.  I've only been addressing the view that fluctuations are at the heart of the analog character.  I say they have next to nothing to do with what really makes the character so musically appealing.  I find myself avoiding as best as I can destabilization, precisely because I'm after that warm rich analog tone.

Regarding your second comment, I should fill in a few details.  I find a lovely effect is created by designing, say, a flute or oboe type mono patch on the first oscillator, and then adding the second oscillator at about a quarter of the volume - just enough to create a slight oscillator beating.  I de-tune oscillator 2 to either 1 or 2, so the beating is very slow and gentle.  Combined with reverb, it creates a subtle acoustic instrument effect, since even they are constantly interacting with their environment - be it with other instruments or with room acoustics.  Even in a little practice studio there are such effects.

Another approach is to design two oh-so-slightly different single oscillator mono patches, and then pan the layers.  Since each sound is coming from a different side, they only slightly interact with each other.  The result is a slightly richer version of the above technique.

[dswo]

The modulation amount of 1 is still too high to get the required variation without sounding like an artifact (that fine line between "wow" and "aaaargh" that Martin refers to). I found that modulating the frequency with noise was better, but it's still a little too obvious, especially if you're listening to the plain oscillator.

I suspect (but I don't know for sure) that if I could set the modulation amount lower (I can't find a way to do this because the logic of Prophet '08's modulation system seems to disallow finer adjustments) I might be able to get the oscillator to sound pure and yet not have that flat and static sound.

Evan Valencia, who was on the old forum, shared two methods for getting smaller values:

"You can get less than 1 by routing the source through either the Mod Wheel or Envelope 3 if neither are used for other functions. I.e. say in the above example you want less than 1 LFO amount for Osc 1 to pitch. First get an LFO and assign it to Osc 1 Freq, LFO frequency and waveform to taste and LFO Amount = 0. Then you go to the Mod Matrix and assign source = Mod Wheel and Destination = LFO X Amount. The Mod Amount will be your top line of the extent of the range you want to cover, in the case you want less than 1, you set this value = 1. Now as you sustain a note, push the mod wheel up and you’ll hear the ‘warble’ get more intense. If you set the Mod Amount = 1 you won’t hear much of a difference, because now the extent of the mod wheel covers all values from 0 to 1 (which is what we want in this circumstance). The mod wheel has a resolution of 8192 or something like that so you get very precise control of the mod. Say if you set the Mod Amount = 127 now you hear the mod wheel affect Osc 1’s pitch in a much more profound way because now the Mod Wheel controls all values ranging from 0 – 127, makes sense?"

[DavidDever]

In an older 1V / oct synthesizer using a linear power supply (typically an EI-core transformer with a few thousand KµF of reservoir capacitance across all the rails), a noise level of 150 µV on the DC rails would definitely make a difference of a few cents here or there, keeping in mind that the fixed-voltage 78XX/79XX linear regulators manufactured 35 years ago aren't nearly as good performance-wise than their modern equivalents - so this effect may in fact be more pronounced in some cases. Note too, that this is all spurious noise, and gets worse as they age.

[MartinM]

Moinmoin

one thing about "sound philosophy" strikes me:

Indeed, the Prophet '08 is a relatively simple instrument...

Yes!

This is true for all analog synthesizers. They consist of only the very basic sound-generating and forming components:

• oscillator (maybe more than one, but something like 64 usually is the upper border)
• filter (usually very simple compared to anything involved in tone-shaping of "natural" instruments)
• envelopes (also very simple compared to let's say the possibilities of bowed or blowed instruments)
• some other "natural" (like keyboard velocity, ask classical pianists about a synthesizer's sensitivity), or "artificial" (like aftertouch or sequencers) modulation sources

And that's quite it. You simply cannot create piano, guitar or double-bass sound with an analog syntesizer, be it literally a classic or just designed that way (as P'08 is). You cannot even mimick the human voice.
Even string or choir pads are much simpler if compared e.g. to mellotron-sounds or modern samplers / romplers, not to speak of real choirs or orchestras. These pads only live by the fact, that "masses" are less precise than "individuals".

Let's put it straight: The only thing, an analog synthesizer does better - at least more distinct - than any "natural" instrument is filter sweep. Looking (hearing!) back into the 70s will teach You, that the most acclaimed vintage lead sounds have much in common with flutes, which are the most simple natural instruments regarding harmonic and modulational content. Some of them become lead sounds by heavily using filter sweep, all of them use the pitch wheel close to a rock guitarist's way of articulation, which in turn inherited a lot from violin-players. The "lead-character" lies more in playing style than in sound!
As a consequence of this simplicity, most synthesists used reverbs, delays or chorus-like modulation and some distortion in order to add more harmonics after the filter did its work. They did this because they regarded the "dry" synthesizer sound as to simple. These effects consequently were integrated into those synthesizers of the 80s, that now are regarded as best sounding of that age.
And most attempts to let VCOs "beat" against each other are merely mimicking chorus effects in order to "thicken" the sound.

This said, let's come to the positive part of it:
If You think about it, the very character of an analog synthesizer sound even lies in its simplicity. Concentration on a comparably small amount of comparably simple parameters is a veritable characteristic of real art. You can work out each single parameter and make it shine. Even if more than one parameter is changed, this principle is what the listener hears and feels.
And like in the old days of the 70s, the further processing of the mere VCO-sound is still crucial:
Did anybody really have a chance, to hear the sawtooth of a Moog VCO, or was it not always routed through a - maybe even wide open - filter and other components?

What I look for in a synthesizer is a

• precise

toolset with

• realtime parameter-control,

from there I may go wherever it leads me. And as with every craftsman's toolset, it is not the quality of one single tool, be it hammer or VCO, but the combination, that will do the work.
And did I say that I love the P'08? 

Martin

[tumble2k]

I just wanted to say that this is indeed a very interesting conversation that covers the artistic, technical, and philosophical. And we're pretty much only talking about the oscillators!

Fascinating what dslsynth said that the VCO frequency undergoes a slow random walk that gives it its characteristic sound. Playing around with adding noise to the VCO frequency in very small amounds (using the trick dswo kindly provided) I think I confirmed what Sacred Synthesis was saying: if you add enough noise to provide movement the oscillator doesn't sound realistic. I think the characteristic frequency of the noise matters, i.e., if it's pink or darker. At a low enough frequency the noise becomes Oscillator Slop, which at high numbers sounds to me like a bad VCO, just as Sacred Synthesis said. Small amounts of slop (up to 2) are nice.

What DavidDever says is even more interesting because he says the rail variation will appear as spurious noise, which I interpret to be broadband. I am imagining a VCO that never quite hitting the same trigger voltage every cycle. This can be mimicked on a Prophet '08 up to middle C (where the LFOs top out in frequency), and I'd expect the effect would sound quite different from slop. I would expect you could simulate this effect on a Prophet 12 at higher frequencies by FM modulating your oscillator with white noise. I don't have a Prophet 12, but I'll try on the Prophet '08 and report back.

Sacred Synthesis's method of mixing in a small amount of a slightly detuned oscillator is great. It does provide a subtle movement while maintaining the integrity of the original oscillator. I also agree with Martin that the Prophet '08 provides powerful real time control over a harmonically simple instrument. The other thing I might add is that an analog synthesizer can sound 80 to 90% like almost any instrument out there (okay maybe 50% of the human voice) which is pretty fun when you want to understand what makes instruments sound the way they do (up to the last 10 to 20% or so).

[Jason]

Playing around with adding noise to the VCO frequency in very small amounds I think I confirmed what Sacred Synthesis was saying: if you add enough noise to provide movement the oscillator doesn't sound realistic.

Sacred Synthesis's method of mixing in a small amount of a slightly detuned oscillator is great. It does provide a subtle movement while maintaining the integrity of the original oscillator.

I agree. Sacred has not yet said "I told you so," but he certainly could. I like to experiment and have had the time to do so this week. The methods of destabilizing the oscillators have not produced good results for me. The most effective suggestions are the techniques that S.S. recommends. One of my favorites how he uses larger amounts of PWM with slower frequencies. So thanks, S.S., and keep the suggestions coming!

[Sacred Synthesis]

One thing that I think spoils synthesizer sound design is a lack of subtlety and a tendency towards extremes.  In listening to many synthesizer videos and recordings, it often seems as if synthesists are trying to impress other synthesists with complexity, as if more is necessarily better.  The sounds are so over-packed with loud and caustic characteristics, blatant filter changes and modulation, that they better belong in a bank of company demo patches, rather than in a beautiful piece of music.  Good sound designers knows how to turn a parameter just one digit, or not at all.  They spend an hour or more searching for the finest minutest adjustments that are almost unnoticeable to the untrained ear, but that perfect a sound and finally attain the nearly unattainable - a sweetly musical character. 

There's nothing worse than a patch that screams, "Synthesizer!"  That's my view, anyways: get the electronics out of the sound and, to the degree that it's possible, make it seem acoustic, or like it sprang right up out of the soil, all green and leafy. 

Less is more, and in all things, simplicity.