Sound Design in Reason

FM synthesis is a method of generating sounds using 2 or more signals. One modulates the other and by definition it would return a different sound. FM Synthesis is probably also one of the difficult methods to explain in just a few words. Just as a reference to a complete math book about it: standford documentation on FM Synthesis

To back up my own thoughts on the whole FM Synthesis, I understand most of this while programming the FM-Pac (or FM-Pack, or OPLL) made by Yamaha in 1986. The chipset was based on one of its keyboards (DX-2 keyboards at the time) and I got around understanding most of it what FM-Synthesis did by the time. Even while I have dropped the whole FM generation from around 1997 and moved on to wavetables and such, I still have some common knowledge about what makes FM tick.

Personally I find the explination about FM-Synthesis made by propellerheads themselves (An introduction to FM Synthesis - part 1, and part 2) and interesting read, however I feel there is one minor issue with this story: it only focusses on the FM Oscillator to sell copies of Reason 4. For some reason it makes you feel like FM synthesis was introduced when Reason 4 came, however this is far from the truth. It was already there when the subtractor was introduced.

To demystify FM synthesis, I will take a different look then just looking at the FM Oscillator. Because there is one thing I am missing inside the picture of FM-Synthesis: Waveform!

When you take any of the oldschool generation game councils (like the Sega Genesis for instance), it uses a sound chip that is based of FM Synthesis. The synthesis is defined by 2 different parameters: Carrier wave and the modulator. This is also refered to the 2 operator FM. Like the FM-Pack, it had 2 operators and both parameters have a series of definitions:

Waveform, Attack, Decay, Sustain, Modulation Amount, LFO and so on. This was originally designed for both the carrier and the modulator.

When we look at the 2 devices that originally have FM, we can see 2 of them that have this: the Subtractor and Thor. Most common thing that people will 'talk' about is the Thor FM Oscillator. Yet there is something much more advanced inside thor to really emulate 100% fm-synthesis. This is the part where it usually goes beyond understanding because FM is hard as hell to understand in just a few minutes.

Ok, enough for the intro, lets get started.

FM and the Subtractor

To understand the subtractor FM it isn't really hard to get things going. The only thing you need to know right now is that FM uses 2 oscillators (the carrier and the modulator). So, in a way, we need both oscillators to make FM effective.

In this setup I am using a single sawtooth wave form and then use that to make FM with it. However, in this setup there isn't much happening with the FM knob, since it needs a modulator. Which isn't here yet.

RNS Download: No FM
MP3 Download: No FM

The only way that frequency modulation works inside the subtractor is to have a second waveform which is different then the original. In this case one could use a Sawtooth, but lower the octave (or increase the octave). In a way, the waveform becomes different (smaller or larger in size), thus it modulates the frequency.

RNS Download: Frequency Modulation Subtractor
MP3 Download: Frequency Modulation Subtractor

Modulation on the sine form

The biggest problem with FM-synthesis is that it can be radical in changes. Because we are chaning the frequency of the first wave form, depending on what the second wave form is. In most cases waveforms suchs as Saws and Square forms will often make a sound "harsh", roughed edged and totally out of sync with the kbd (eg: they don't play along with the notes that nicely). In most cases the safe spot using FM is below the setting of 32. Everything above it will make it go 'off note' at some point. This is an avarage take, since it depends on the waveform.

The sine form is most likely the most secure one put FM on it, since it is clean and mostly harmonic (no gradual changes since it slightly goes up and down).

RNS Download: Frequency Modulation Subtractor
MP3 Download: Frequency Modulation Subtractor

To fully understand what makes FM tick, one could use a long note instead. Notice the changes of the tone while the waveform goes up per octave:

RNS Download: Frequency Modulation Subtractor
MP3 Download: Frequency Modulation Subtractor

One thing I hope will be noticing in this setup is that as soon as the modulator waveform gets closer to the original waveform, the fm-effect becomes lesser noticable. This is because the modulation goes faster and thus it becomes more stable. Another setup you could compare it too is using and LFO and control the pitch while you are at it.

RNS Download: Pitch changing Subtractor
MP3 Download: Pitch changing Subtractor

See the similarties over here? One might think there is a misconception here. It might seems like we are controlling the pitch of the waveform. However, we are not. We are modulating the frequency of the waveform (the speed of the waveform), so in a way, the faster the waveform goes... it will lead into a difference in pitch. This is confusing... I know.

Lets try a more simple aproach. The square wave form is something that just goes on-off-on-off. You normally wouldn't recognize it, because it sounds like a harmonic tone. However, when you go really low, you will hear what a square wave form does.

RNS Download: Square wave Subtractor
MP3 Download: Square wave Subtractor

Just a simple low square form... like I said, it just goes on and off. Almost like a ticking sound. Ok, what if we add a sine form to that and throw that agains the Frequency modulation while we are at it. This will make the Square form change accordingly to the sine form.

RNS Download: Square wave Subtractor FM
MP3 Download: Square wave Subtractor FM

There, now you will definitly notice whats happening over here. The square wave form changes shape accordingly to the second oscillator (which is in this case a sine wave). The FM knob will tell how much the frequency is going to change. So, if you would do it really drastic it will more or less become an LFO set on sine form and connect that to the pitch at the same time.

The difference is: if you would an LFO to control the pitch, the LFO rate will remain the same and thus the pitch effect remains the same, yet with FM the rate changes as soon as you going to trigger different notes. Because different notes also means that the oscillators changes at the same time. Therefor: FM is hard as hell to understand in just a few minutes.

Anyways, I will just leave at this for now. Maybe it may give you some ideas to play around with. Before I will close this topic down, here is a different aproach which I will be discussing in the next part(s):

RNS Download: Square wave Subtractor FM
MP3 Download: Square wave Subtractor FM

Here is a minor bit of a tutorial about making real instrumental approaches. The problem with real instruments is that you have to have heard them, otherwise you can't make a conclusion about what an instrument would sound like. This is also a bit of a tricky part that comes along with it. Since alot of people will make an accusation that it does not sound real, or it does not come close to the real deal. Synths (eg: the subtractor, malstrom and thor) are pretty powerfull stuff to create synthetic sounds, however... that is what they are. The synths are fake. Thus creating real sounds isn't usefull at that. However... one could always use a synth for a backup to make the sound itself a bit more 'unreal', while still having a real sound on top of it.

It sounds cheap, but the only way you are able to create realistic sounds is by using samplers. Most of the NN-XT patches that are inside the orchestral bank are live recorded instrumental sounds. Thus, they come as close as they can get when making realistic sounds. However, the real problem with the NN-XT is that they often lack something. It lacks a certain touch.

I personally could debate about every single instrument that is out there, yet this is something I am not willing to do yet at this point. Since this would lead into an article discussing the different instruments and would probably get real boring at the same time. Just to give you a heads up on this one: here is a list of instruments that is made on Wikipedia. The technique which I am going to discribe right now probably works with a lot of different instruments, and that is more important to write down, then just explaining the differences between each single instrument.

Ok, long intro... Lets get started.

Orchestral Bank

Inside the Orchestral bank you may find alot of nice instrument patches which can be loaded up inside the NN-XT sampler. These patches have everything to do with orchestral sounds. They are arranged inside different categories, woodwings (instruments made out of woods), strings, Mallets and Brasses (mostly coper related instruments).

The challenge about making it sound good, is to layer them properly. And that is often the real big issue with it also.

Lets just start of with a simple example here, a marc violin. The Marc Violin is short, has a real sharp tone inside it, and works well with short notes. However... well.. Let's just hear it.

RNS Example Download: Violin Session 1
MP3 Example Download: Violin Session 1

To make the sound itself more interesting at some point, we can add this one inside a combinator patch and multiply the same violin for a couple of times. To make this work, we'll be needing a combinator, a mixer (can be a mixer 14:2, or a mixer 6:2), an NN-XT patch and a spider audio splitter merger.

With the spider audio splitter merger we can split the sound into different signals to process. In the real world, you will often hear that a sound has a delay, or is being bounced of through a wall, or ... has some form of accoustic. In this setup one could 'fake' that enviroment just a tiny bit. By using 2 delays (as an example) one can put the same signal in a minor delay. Which kind of makes it sound like a chorus effect, yet at the same time it doesn't. The reverbs I am placing inside this example are for shaping the sound output just a tiny bit different. Otherwise the sound itself would come out a bit 'metallic' like.

At this point it makes the sound just a tiny bit fuller, since we are kind of like playing 3 times the same violin at the same time. More or less. The panning in this setup is pretty much an important factor at this point, since having a delay, and the sound dead center as well, it will make the sound 'hollow' at the same time. And that is something you probably want to avoid here. So... Just to demo this setup:

RNS Example Download: Violin Session 2
MP3 Example Download: Violin Session 2

Layering

Layering your sounds is often the key to get the right tones, or sound you probably need at the time. At this point the violin session we created shounds short and snappy. To emphasize the short and snappyness there is a Violin called Violin Stacc Dwn_Up. This articulates the beginning of the violin stabbing sounds. However, at some point it does come pretty close to the Marc violin at the same time. So, one might throw in a filter or two just to blend it in just a tiny bit different.

RNS Example Download: Violin Session 3
MP3 Example Download: Violin Session 3

To make it sound a bit more 'hall' and longer, one could add a FatOrchString. This one can be found inside the Reason Factory Soundbank (the default thing), and is located under the NN-XT patches, Combo and Extras. And alternate setup would be using the SuperString (inside the same folder). The only problem with the last 2 mentioned sounds is that they are very powerfull in volume. So, one might want to reduce the volume a lot with these 2 in combination with the other Violin sessions coming from the orchestral sounds.

RNS Example Download: Violin Session 4
MP3 Example Download: Violin Session 4

The real cool part with the FatOrchstring in this case, is that it can be usefull to play longer notes. The Marc and the Stacc Dwn Up are for playing short notes mostly. So it kind of blends well making short / long transitions at the same time.

RNS Example Download: Violin Session 5
MP3 Example Download: Violin Session 5

Flutes in Reason

Just as a second example, lets take a look at flutes. Flutes are mostly wooden sounds and are often very high pitched in tone. Inside the Orchestral bank one can find a wide range of different flutes. However, one could also shape it back and forth to create a new one. Personally I often go back to the combination of using a straight flute and combine it with a clarinet at the same time. The flute itself is high pitched, has a more 'vibration' feel to it, while the Clarinet is more stable in tone.

Only one minor problem when combining the two... it sounds like a detuned sound.

RNS Example Download: Flute Session 1
MP3 Example Download: Flute Session 1

To sort this main issue here we can add either a bit of reverb or delay in here, so it doesn't sound that much of a detune effect anymore. It's all about finding the right 'sweet spot' and it comes with trial on error.

RNS Example Download: Flute Session 2
MP3 Example Download: Flute Session 2

The Fake Vibration

In this perticular example, one might find a trembling undertone going on inside the flute itself. When listening to it for a long time, you might recognize that the sound itself almost sounds like a sine form that just trembles up and down in volume just a tiny bit. Just to emphasize this trembling feeling, we could use something 'fake' in here. When we add a malstrom to it, and throw it on an LFO that is connected to the volume, it just goes slightly on and off at the same time while the flute is playing.

I placed the Scream unit in between it to put some dirt on the matter. Even though it is minor, it is there... and thats what matters.

RNS Example Download: Flute Session 3
MP3 Example Download: Flute Session 3

Only one minor issue with the setup is that default the malstrom doesn't listen to the velocity settings just like the NN-XT patches usually do. So, in that case we might want to use a velocity setup just like that to correct the problem.

In this case the volume will increase or decrease accordingly to the velocity of the notes themselves.

RNS Example Download: Flute Session 4
MP3 Example Download: Flute Session 4

Puff...

Puffing sounds are common in a flute combination. It is the sound that is being 'triggered' at the point you breath in, and starting the first note on the flute. It is hard to explain just in a few words if you never heard a flute before. But trust me... it is often there. Only the good artists that have +30 years of experience on playing a flute will not have this problem. For this I often throw in a Voice Puff. This one can be found inside the Reason Factory Sound Bank under the NN-XT patches, then the Reed and Pipe.

The only thing you need to 'fix' with it, is that the default puff sound is often very long. Since we only want to add that extra touch to the beginning of the sound, we can drop the release just a bit.

Another option would be using the Malstrom in this case and use a waveform called "breath". But I would be a little bit carefull with that one, since it can also sound very fake at the same time.

RNS Example Download: Flute Session 5
MP3 Example Download: Flute Session 5

Right, at this point I hope you are starting to figure out the big picture with it. It comes down to 'layering' and layering some more when you want to create these realistic sounding instruments. Often the biggest problem, or dillema, is that some arrangements want to 'fake' the accoustics. This is often a different story and it requires alot of knowledge about how accoustics works. But on the short term, I hope this helps a few people out who are dealing with the problem with having sounds that often sound too weak, or just don't come as close as the real deal.

The only way you are really going to get close to real sounds, is probably by recording it yourself... But that's probably a whole different story.

Last week I didn't post any sound design topic due to the issues I had with the webserver. But, I had this topic in the back of my head to do. TB303 sounds inside of Propellerhead Reason.

A common thing that alot of reason users have been asking me in the past. It is either the question: how did you create the TB303 sound effect. Or how do I use rebirth in reason. Problem is, using rebirth inside Reason is in lamers terms a bitch. I have no other way to describe it. Problem is, rebirth acts as a rewire host, and there is no wait you can control it using it inside reason. There are no tracks to create, and the only way you will get the sounds from rebirth inside reason, is using the sequencer of ... rebirth.

Therefor, I have always skipped rebirth and tried to examine what posibilities lies ahead from creating those TB303 sounds inside reason. Some of this will be 'fake', since we are just trying to emulate the TB303 series. Other stuff are 'thoughts'. Since when we have a TB303 sound created we can eventually expand on that sound and try to create something new from it again.

Since I have had the discussion with someone called "h4x0r" in the past on youtube about what makes a TB303. I will through once again these specs: TB303 specs

The basic point of discussion I have had in the past is, what makes the TB303 tick... Well, to be honest there are 4 different things: Waveform, Filter, Resonance and X Mod (cross modulation). The first 3 we have, the last one however, we don't have. But we can fake that part most of it. Also, the TB303 has some form of distortion trigger when looking back at Rebirth. Other then that, the TB303 plays notes inside a single sequence (which is handled by a 16 step sequencer).

Just to give you an idea on how the TB303 sounded like, here are a few examples:

Originally the tb303 module was introduced to become a snythetic bassline replacement, but since it didn't get much attention from live band artists back then it became more or less a famous techno or acid synth.

Most of the common things inside of the tb303 sounds is that it often sounds pretty 'crispy'. There is a somewhat harmonic tremble tone inside the tb303 from time to time, and that is where the xMod comes in. It modulates the waveform internally. While reason only has the Ring Mod to modulate internally using the subtractor at this point, there are some ways to use a subtractor to emulate the thing. So, lets get started by looking at the subtractor to 'emulate' the tb303.

The waveform

The discussion I have been having alot of times is the waveform. Looking back at the specifications it uses 2 waveforms: square and saw. So we can start picking those to start with.

At this point one might also have the polyphony set to 1. The TB303 is only able to play one single note at the same time. This setup will also help is in the near future when we want to emulate the trigger called 'slide' (I'll be getting into that later on).

So far, the sound itself is pretty straight forward and sound more or less like a nintendo game orsomething like it.

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Phase and modulation

The trick we want to go for is to make the sound a bit more harsh and tremble a bit more. To emulate that effect we can use a thing called Phase and subtracting. This means that the one of the 2 oscillators will subtract the sound from the original waveform. To make the sound 'kill' itself during progression, we could enable the Ring modulator for this. This will mean that the shape of the sound is changing with the Phase, yet it kills itself using the Ring mod. Making it harsh...

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

This sound might start to sound 'weird'. But at some point you might understand what I am trying to do here. Since the 'tremble effect' is definitly not noticable on the higher ends, one might to set it up really low to notice the difference with phase / ring mod.

Setting it up more appropriate

Now that we found the 'trembling sound effects' which the TB303 provides we can tune it a bit more to make it sound more like a tone, rather then this aggressive pulse. So, next stop would be increasing the oscillators back up, with minor decrease of 1 octave.

With this setup we can go really low, and have an aggressive pulse comming back. Still... it sounds pretty clean at the same time. But we'll be fixing that in just a few seconds.
Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Resonance

One of the things that makes the TB303 tick is Resonance inside the sound. In a lot of cases you will find these sounds coming back, and that is also one of the reasons why a TB303 sound is more distinctive and recognizable. Since the TB303 was originally made for 'emulating' basslines, you might want to keep the cutoff frequency pretty low, however go totally nuts on the Resonance of the filter.

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Slide

To emulate the slide feature of the Roland TB 303 one needs 2 settings. First the note trigger set to Legato. This means that the filter envelope and Amplitude envelope will continue its course while the sound is continues. Second we need a minor bit of portamento. This means that the notes will slightly go pitch up and pitch down from note to note (rather then doing it 'all of a sudden').

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Accent

The Accent is something that comes with the TB303. The only main difference with the Reason velocity and the TB 303 velocity is, reason has 127 states (value 0 till 127) yet the TB 303 has 3. Off / On / Accent. So at some point it gets close to what the redrum has with its 4 state (using the pattern editor), Off / Soft / Medium and Hard. One could discuss when one would stick with it or not. Personally I prefer the overal 127 values of the velocity since it allows more control. But basically it comes down to one simple principle: the velocity controls the filter amount or the loudness of the sound amplitude envelope.

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

Setting up Thor

For the Thor approach it becomes a bit tricky at some point. Problem is, there is no real ring modulator as in the subtractor. There is however a method to modulate different oscillators with eachother. But it requires a lot of insight at what you are trying to do here. With the current setup I am going with, I want to modulate the Square form with a Saw form. This would be similar as having an LFO attached to the square wave, however... there is one major difference. The LFO is usually a stable pattern (unless you tell it to be otherwise), but the waveform changes shape when playing different notes. Eg: the lower the note, the longer the waveform gets, the higher the note the shorter the waveform gets. So we'll be using that to our advantage in this case.

The initial setup I am going for is using 1 Square pulse with almost a straight pulse with (just dropped it a bit to the left). The third oscillator is a Saw wave form that makes the sound a bit more harsh. The 2nd Oscillator (also a Saw wave) is going to be our pulse width modulator to emulate the X-Mod.

To make our 'fake' x-mod work, we need to tell the saw wave to plug inside the PW setting.

And there we go. I have set the amount around 50 in this case, because I feel everything above the 50 it modulates too heavy (sounding rather harsh and off-note). By using the OSC2 -> setting, it allows us to use the oscillator itself to become a parameter by default. So, in away we can connect any waveform to anything. But that is more or less offtopic in this case. But it is a nice method to get some crazy modulation going on when an LFO just simply does not do the job well.

Last part, the audio out setup. As OSC 1 is being modulated by OSC2, we don't want to be using OSC2 as an audio source (since it just acts as a modulator). Osc 1 goes in this case through filter 2 and Osc 3 through filter 1. One could decide to put them through both the same filter. But while we are emulating it, but not use the best of both worlds at the same time.

The only thing that is a bit troubling, is that the sound still sounds a bit too clean. To put some real dirt on it, and make it really sharp I am using a BiPulse Shaper setup with maximum drive.This will result into a sharp pitched trembling tone combined with a lot of resonance inside it.

To conclude this whole setup, it sounds as followed:

Reason 4 RNS download: TB303 initial setup
mp3 file download: TB303 initial setup

The Matrix

Eventually it just comes down to one simple addition. If you really want to have the look and feel of the TB303 sound setup, use a Matrix pattern sequencer. Since it allows you to play only single notes (just like TB303) and it allows you to slide really easy by using the section called Tie.

Reason 4 RNS download: TB303 Matrix setup
mp3 file download: TB303 Matrix setup

Reason 4 RNS download: TB303 Matrix setup
mp3 file download: TB303 Matrix setup

Reason 4 RNS download: TB303 Matrix setup
mp3 file download: TB303 Matrix setup

However, we can also use some cool tools from reason... for instance, throw in a bit of distortion while we are at it.

Reason 4 RNS download: TB303 Matrix setup
mp3 file download: TB303 Matrix setup

I do know for sure that there will always be a kind of discussion weather this is sounding like a TB303, or not. However, with Rebirth out of the loop, this is probably as close as we can get creating TB303 sounds. However, one might also combine old school and new school with eachother. Call it progressive TB303, or new style TB303. As long as it reminds us of those sounds that we loved so much, it doesn't really matter if it is created by a single wave file, a piece of software or the real thing.

Have fun, and till next week!

A discussable topic is the usage of a Double bass, because it is tricky at some point. Alot of sound engineers will most likely say: "It is better to compress a single bassdrum, rather then playing 2 at the same time". At some level I agree on this, since you have to know what you are doing when it comes to the principles of double basses. The tricky part comes from the sine form. A lot of bassdrums are analog sounds (well enginered kickdrums made on a single analog synth). The tricky part comes from having two sine wave forms and have them inverted. When a waveform becomes inverted, as in the complete opposite of the original waveform, it has the intention to make the sound disapear.

To make this part noticable I am going to this part yet again, one more time to present this thing. So here goes, a series of RNS files that show that I mean with it (incase you missed it).

Download RNS files:

Normal Sine NN-XT
Inverted Sine NN-XT
Inverted Sine Subtractor

That is problem number one, second problem is, if you play the exact same sine form on the sequencer, it will mean that the sound output will increase, not decrease. It is like adding a single thor patch, put the analog oscillator on sine wave, bring in a note on C3, and copy paste the same patch and play the same note on C3 again. It makes it louder in a way.

Loudness can be good from time to time, however, make sure you don't ever cross the 0db level while doing so.

However...

There are methods to solve this issue. Most of the common methods would be controlling the pitch, Eq it differently and bend it differently. But I feel like jumping forward, so let us just take it one step at a time

Redrum Computer in Reason

Lets just assume that the Redrum will be used in most cases to layering your drums. Other artists would prefer the NN-XT rather then the Redrum. The NN-XT is a more advanced work horse to layer your drums, so I might want to stand still at that topic later on. The reason why the Redrum is often choosen for drum layering is because of its simplicity. One can prototype a drum really fast. The ingredients for this perticular thing would be: Redrum and a bass drum loaded up inside channel 1.

For this example, I am going for something more Techno / House like drum. It all works similar like with genres as HipHop, Trance etc. At this point the bassdrum sounds nice. It has a nice bunch, crisp feeling to it, yet it lacks some heavy undertones (deep bass).

Download RNS file: Single Redrum channel
Download Mp3 file: Single Redrum channel

When we would load up a second sample inside for instance channel 2, we could just bring in the same notes to make it play at the same time. Just copy paste the sequence on to channel 2. However, there is an easier workaround for this part.When you toggle to the rear of the Reason rack, you will find a couple of audio slots and CV slots from the redrum.

The audio cables are in this case not the once we are interested right now, the CV parts does. They are marked with the numbers 1 and 2 in the above picture and they have the following meaning:

1. CV Gate out
2. CV Gate in

The Gate out is being triggered each time we trigger a note on that specific channel. So, when we would connect the CV Gate out of channel 1, and connect it to the CV Gate in of the second channel, this will mean the each time channel 1 is being triggered, it will do the same with channel 2. We don't need anything on the sequencer to solve this part for us.Eventually when tying them together, it should look similar as the following.

The only thing that remains is that we need a sample to play on the channel 2. This is where I am throwing in the BD_Boomer of the default xclusive bassdrums.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

At this point, it still does not really sound 'perfect'. But it does add a bit of a low tone next to the XTC5 drum that we already have been playing on channel 1. However, when looking closer to the mixer, you might also notice that the sound itself has become louder. That last part is the tricky part.

Where does the loudness come from

Like I said earlier, when triggering the same sine forms, it makes the sound louder at some point. Even while we have been choosing 2 different samples, it still has the intention to make the sound louder at some given point. Most bassdrums trigger the frequencies around 500Hz and below. Most common issue is that there is a range around 80Hz and below which is not noticable by the ear. Bassdrums are often for the feeling, and thus they tend to go more lower then they usually would do. Since both bassdrums are triggering around the 60Hz and below, they will emphasize this specific frequency and make that part louder. Thus the mixer inside Reason does 'hear' this, it is not always noticable.

The only method you could solve this issue is by either mixing down the two bassdrums volume wise, or.. pitch wise. Increasing or Decreasing the pitch can have a big impact on how the drums will sound like. Especially with this boomer bass I have loaded up. This drum is by default already pretty low by tone, most of the 'hearing' presense is done by the XTC 5 drum. So to keep the whole drums intact. One could decrease the volume of the XTC 5 in this case, which reduces the presense of the drum, yet decrease the pitch of the Boomer drum, which makes it even more lower.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

This whole concept comes with trial on error. You will most likely have to listen to the same drums for 5 minutes straight, adjust something and feel the changes are made are right. There is no real golden rule on this part (I wish it where so, since it would make it easier to explain it in just a few words).

High Tones

Another twist around the whole subject of the double bass, is that you could reduce the high tones of a specific sample. Channels 1, 2 and 10 have this option by default inside the redrum, the others don't. This is also one of the main reasons why I prefer to use these channels to shape the bassdrum. The tone is a setting that is similar as a HighPass filter kind of thing. It cuts of, or increases the high tones of that specific channel. I prefer to decrease it with this specific one, since I personally find the high pitched tone on this XTC5 a bit annoying in the long run.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Tripple Bass

Now for the juicy stuff. Since we have been layering 2 drums at the same time, we could for instance add a 3rd layer to all this. However, like I said, this would also mean that the sound itself becomes louder in some way. However, there is also an option to seperate channel 2 and channel 10 (which I am about to use) over left and right. But first, we need a way to use the CV Gate out to connect to two different channels. To do this, we'll be using a Spider CV Merger/splitter.

In this specific example, the Gate out of Channel 1 is going inside the Split A section. Then the signal is being routed back to the Channel 2 CV Gate in and the Channel 10 CV Gate in. Which means that now channel 2 and 10 are playing the same thing as channel 1.

To make the final adjustments on this, Channel 2 goes to the Left, Channel 10 goes to the right. While channel 10 has the exact same drum, yet set on a minor different pitch and tone reduction. This results into the following setup:

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Last part for now to finish it up, we can add an additional Compressor unit to spice things up a bit. It all depends on your flavor and style.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Just sum up all fears... here is a quick example file that I did really quickly:

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Double Bass and the NN-XT

Like I initially stated, some people prefer the NN-XT over the Redrum computer. While the redrum is 10 channels specific, the only reason why people would prefer the NN-XT over the redrum is the layering part and fine tuning the sample. Since the NN-XT can shape each layer differenltly using the Amplitude Envelope (which the Redrum does not have at all), Modulate pitch or add an additional LFO to the mix. In most cases, it is the ideal drum machine. Yet its workflow is also a bit different and harder to go by.

It all starts inside the remote editor of the NN-XT. One can loud up more then just one sample inside it, and give each layer something different. Lets keep it simple, and add a bassdrum inside the NN-XT. I think in this attempt I am going for a more Hip Hop kind of beat, where the tempo is going to be around 90 beats per minute. Most of the common hip hop kits inside Reason use percussion name like: Riot, Rare or JC. So I think I will stick with those, just to keep it easier to track.

In this case, I have loaded up the BD1_Riot from the Redrum xclusive drums inside the NN-XT. The reason why I am picking this perticular drum since it has this little high pitch snap in the beginning. Which is like a cymbal kind of thing.An it has a nice ring to it.

The amplitude of this perticular part might be important to mention in this case. Default the samples has a little release to the sample. In this part I want the release to be all the way up, so the sample does not cut off when the note stops. If you have a long bassdrum that almost never quits, then it would be ideal to reduce the releaase setting in this case. However, the bd1_riot is short and picky.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

For the next segment I am using a BD5_JC. This is the type of bassdrum that goes really low, and adds a bit of punch to the whole thing. Notice that the Pitch is reduced by using the semi setting of the sample

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Next part is kind of similar as I did with the redrum. The sample goes to the left, and we load up a second sample that goes to the right.

While adjusting the fine tune setting of each drum, you will find some parts where the drum becomes really present on the low ends.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

From this point there are probably tons of different ways to reshape the whole drum. For instance apply a little bit of pitch effect on top of the second and third layer.

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Use a bandpass filter on the first layer to make the first layer a bit shaper (and more distinctive).

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

Just to top it all off, here is a minor example what you could do using a bunch of NN-XT samplers:

Download RNS file: Double Bass Redrum
Download Mp3 file: Double Bass Redrum

As a request coming from the forumboard, I am just going to skip the drumming part (which I initially intended to do). In this case I first want to stand still at the pads used in Drone music. Drone music is refered to a minimal genre that often is build up by pads. Sometimes a few instruments are playing in the background (like vibraphones or short picked tones), but the idea behind is to make the sequences long and monotomic at some point. Due to that the sequence itself is pretty much monotomic, the main trick into ambient drone music does not lie in the sequence itself, but in the instruments that are being used. In this topic I want to stand still at the parts of the ambient drone pads. Might I also point out, I am not an expert on this genre, however, I will give some guidelines that might help you to get things going and build from there on.

The main elements

To sum up the main elements one would need to get a pad going in the Ambient Drone universe one would need a few things:

- Layering
- Dynamics
- Effects

Step 1 Layering

Lets start of with the first part, and go from there. Layering is a technique that I often have been discussing. It usually starts with a single oscillator, then bringing in a second one, and then one can start adding additional devices. Since it is all about stacking stuff inside the rack, one would probably prefer a Combinator patch to make a pad of this magnitude. In most cases I start of bringing in a mixer 14:2. Reason why I prefer the 14:2 over the 6:2 is that the signal strength of the devices attached to them come out more accurate (visual wise). It is not the additional Aux system I am interested in in most cases.

So lets start of with that. Add combinator, attach mixer inside the combinator and we would have the following starting point.

At this point there is nothing special about it. From this point on we could start choosing our synth as a basis. Personally I prefer the Malstrom or Thor in this case, since they are more versatile when it comes to the inital pad build up. In a later stage one could add subtractors to fine tune the layers and such. But then again, it probably depends on what the preferences are. One could stack up a bunch of subtractors to get the same effect.

To start up, I am going to use the Malstrom. Since it sound is often a bit thin, and that is what one would need at this point. Something not fully present since we are going to fix those issues in a later stage. As a wavefrom, I am picking the additive Wave 2 in this case. Since it is pretty light and bright. Other wave forms that one could use as a basis would be: the basic analog, square, saw, triangle, sine, the male choir (for something more deeper) and the additive wave (which are awesome for dynamics). There are probably tons of other waveforms to pick from, but I will just keep it short.

By decreasing the motion of the additive Wave 2 it will shape the sound differently. The main problem with the additive wave is that it often does not sound right. Its native note is C3, and everything above or below in octaves it starts to sound messy. By reducing the motion of the additive Wave, it already gives it a more synthetic feeling to the sound surroundings. And with Ambient Drone pads, that is something most definitly you want to go for... because it should not really represent a realistic sound.

To initially set up the second oscillator one could use the same motion ssetting, or make a minor difference. In the initial case I am going for I have set the motion for OSC A on -23, so having the OSC B on motion -21 or -25 would be a good thing to do. Since this makes the sound just a tiny bit more 'detuned', but using a different approach. Second we can re-arrange the tune settings of the 2 different oscillators, where one goes up, the other goes down. That is your basic detuning setup.

Download RNS file: Drone-setup1.rns
Download Mp3 file: Drone-setup1.mp3

At this point the biggest problem is the starting point of the additive wave. Since it starts with a pretty big hightone and then becomes more subtle. I think it depends on the flavor, but to fix certain issues with the waveforms one could go through the index of the malstrom and try to find the starting point of the samples to play. With some waveforms the index does not have a big impact, only with the waveforms that constantly change over time (with is the case with wave forms such as the additive wave, thunder, metal cans and so). I have set the index to 36 with both oscillators, eventually the start point will change since we are going to change the envelopes as well.

To make the ambient pad more effective one could setup an amplitude envelope with the following:

- attack pretty high up
- decay, all the way up
- sustain, all the way up
- release, pretty much high

Which would look like the following setup:

This will mean that the volume slightly kicks in over time, and when the note releases, it takes along time to fade away. It does not really mater that the notes are overlapping with eachother. That is, at somepoint, the key to this type of music. But please make sure: to check the polyphony count! Problem with long release is that the polypony will increase while doing it. So, if I would play a 3 finger chord session, and place a 3 finger chord session after it, it will mean that the polyphone count at that specific moment is 6, and not 3. Also make note, the more polyphony you have the more CPU usuage it takes to blend all the different sounds. A safe spot, while using 3 finger chords, have the polyphony count set on 9.

Download RNS file: Drone-setup2.rns
Download Mp3 file: Drone-setup2.mp3

So at this point we have started to layer our initial sound, but... we are not done quite yet. At this point one might find that there are cetain tones pretty much present while playing a single long note. It kind of becomes pretty much present when looping it and listening to it over and over again. That is also one of the keys with ambient music and the time consuming parts with it. When listening long enough, you will find certain elements inside the pad become to much present, and such elements don't belong there. With the additive wave I made, it is something that goes low and it goes like "ta du ta du ta du".

One method to find the Frequencies of what is causing this issue is to add an EQ and then increase the frequency... (ehm what?!)... By adjusting the Frequeny knob accordingly (from high to low, it makes it easier to find that specific tone that is causing the annoying ta du da du sound.

Download RNS file: Drone-setup3.rns
Download Mp3 file: Drone-setup3.mp3

Using the EQ on its max is often a technique I use to find the problem areas, since it gets really noticable when increasing a band and float around the different Frequencies while looking for the problem tone.

It get annoying doesn't it... Since we are here to fix the issue one could easaly get rid of that annoying tone by decreasing the EQ of that specific band.

Download RNS file: Drone-setup4.rns
Download Mp3 file: Drone-setup4.mp3

Okay, at this point I think we have setup our initial layer nicely. Next part is the dynamics. If you wandering right now why I am not adding new layers yet... hold that thought.

Step 2 Dynamics

Because ambient drone music is often subtle in changes, you probably want to create the changes inside the patch that go subtle. While the additive wave form is already pretty dynamic by itself, it could always use a minor touch up for the dynamic parts to increase. Some ideas that often go through my head when it comes to dynamics and the malstrom are items like the following:

- Modulate the filter
- Modulate the Volume
- Modulate the index
- Modulate the shift

Modulation is basicly the same as attaching an LFO to any specific parameter (just in case you are using Thor). It comes down to changing a parameter while the tone progresses. However, the trick with ambient drone pads is: do it subtle. You probably want to keep the tone changes long as in 8/4 long, or even better yet... not have them tempo syncronized at all. Since that is what makes it more dynamic. If you have for instance a beat in your ambient track that subtle goes thumbling in the background then it is even better to have the mod A/Mod B with no tempo sync.

To focus on the filter parts, one could turn them on and use LowPass (LP12), BandPass(BP12) or Notch. Ok, the malstrom does not have a notch by default... But we can fix that. It depends on what kind of layer you are working with. Ambient basicly focusses on mostly everything... the lows, mids and highs. In this case, I will call this the mid layer. Since its tone is already much present there. So using LP12 in this case would be ideal.

To find the right frequencies really fast, we could use a Rate setting which is pretty high, use a bit of Frequency changes to find the right spots.

Download RNS file: Drone-setup5.rns
Download Mp3 file: Drone-setup5.mp3

You will notice, while the frequency goes down, that it does not have much impact on the filter what so ever in the beginning stage. Yet, while the frequency goes down, it becomes more noticable. This is probably why in this case, having the LP12 set around 60-80 would work probably best. Now that we found the right frequencies and such to tweak with the modulation, the next thing we can do is decrease the ModB rate, remove the automation and set the frequencies accordingly.

Next stop would be changing the volume. While we do have a ModB and a volume parameter, one could easaly use that same setting. However, since we want to make it more dynamic in this approach, it would be wise to flip the rack around and attach a second LFO parameter that only controls the volume modulation.

The output volume of the second malstrom is not being used. It is just there to be an LFO. The ModA parameter is connected to the Level, which is simular as having a ModB using the volume setting. By decreasing the Level knob just a bit, we can determine how much the level goes up and down. Since we don't want to really over do it just yet, I am setting it to just a tiny bit (17). So this means the level changes gradually... not too much.

Download RNS file: Drone-setup6.rns
Download Mp3 file: Drone-setup6.mp3

Step 3 Effects

Now that we have most of the basic elements for this single malstrom up and running, we can start layering all the different effects.A few good once would be: reverb, delay, chorus, phaser and detune. Ambient pads are probably one of the few instruments where I don't use the Scream 4, nor distortion (huh?! there's something you don't see me typing everyday). Since the pad is already pretty much long on the release, the reverb should be subtle. However, one can use both the dry and wet signal to make the changes. This is a method I personally prefer and it requires only 2 things. 1. A Spider audio, 2 a Reverb.

Download RNS file: Drone-setup7.rns
Download Mp3 file: Drone-setup7.mp3

The really tricky part lies with in the RV7000 in this case. Since it is just a reverb... however, one use the same dry signal from the malstrom (read, spider audio) and attach that to as many devices as we can. In this case I have hooked up the dry signal to the main mixer of the combinator inside Channel 1. The reverb goes in channel 2. However, we could create a second RV7000, and attach it to channel 3. Using a different Reverb and then seperate channel 2 and 3 by using panning. This is one method to go for, and it works really well. However, the cool part about the RV7000 is that it also allows us to shape the EQ of the reverb. So in a way we could apply stereo seperation on reverb level... Sounds confusing... I know, lets just take this one step at a time.

First we take the current RV7000 we have in our song. To enable the EQ on the reverb, we need to select "EQ Enable" (highlighted in the above picture). Since the reverb is set to wet, meaning it only uses the reverb sound, it will become really noticable what we are going to do with it. Keep that in mind. Next, we'll open the remote programmer of the RV7000 and select the EQ switch on the left panel.

\With the EQ selected, we can use it just like you would normally use a 2band EQ, or MClass Mastering unit. But again, this only has an impact on the Reverb.

So setting up an EQ like the following will mean that the reverb is controlling mostly the mid/high range. Not the Lows. We can split the signal yet again and attach the same reverb to the signal. But using a different EQ setup. Which should eventually lead up with the following setup.

Download RNS file: Drone-setup8.rns
Download Mp3 file: Drone-setup8.mp3

For our dry signal we could eventually apply a minor delay in here too... etc.. etc..

Download RNS file: Drone-setup9.rns
Download Mp3 file: Drone-setup9.mp3

Resuming the "Hold that thought part"

Before I ended the paragraph about the "layering" part, I said: Okay, at this point I think we have setup our initial layer nicely. Next part is the dynamics. If you wandering right now why I am not adding new layers yet... hold that thought.

Now the thought continues. Because what we basicly have been doing is focussing on one level of the sound, namely the mid tones. The lows are pretty much gone, the highs are somewhat present. But most of the stuff is focussing on the mids. Why I prefer to focus on the mids first is because they are most noticable. Shaping that part is kind of vital. From this point on we could stack stuff pretty high and focus on the lows and highs. Maybe add an additional mid tone and a swapping notch that ties the lows and highs together. Or go nuts like the following muppet.

Layering the Lows

For the lows I am going to attach a sawtooth*4, based from the malstrom. The nice part about the sawtooth*4 is that it uses a minor sound modulation inside the sound itself. So having a minor motion going on would again be ideal in this case. Since the this patch is focussing mostly on the lows one might bring in a LP12 in this case with the frequency pretty low. For the amplitude envelope it is going to be the same drill.

At a certain point you would get a sound like the following as in solo.

Download RNS file: Drone-setup10.rns
Download Mp3 file: Drone-setup10.mp3

and as a complete setup:

Download RNS file: Drone-setup10-2.rns
Download Mp3 file: Drone-setup10-2.mp3

Layering the highs

For the highs we could use either a bandpass or highpass in this case. High pass would be ideal in this case, but that would require some nifty routing inside a thor patch.However, I am going to stick with a bandpass filter this time and set the frequency pretty high, so it comes close to what a high pass filter should be. The wave form is going to be more roughed edged. The sawtooth 16 should be ideal in this case.

Basicaly it comes down in to the 3 same things I have already been writing about: layering, dynamics and Effects. So, here comes another malstrom.

The main idea I am going for with this setup is that the octave is setup with one octave higher. The other 2 were reduced with one octave. This makes the sound more distinctive since it is more high pitched. By using the Bandpass, with a minor modulation, it starts to sweep in pretty. The attack is just a bit higher, so it takes a little longer before the high tones start to take effect.

Download RNS file: Drone-setup11.rns
Download Mp3 file: Drone-setup11.mp3

The Bridge Layer

From this point on we have 3 different layers. Each sound has its own purpose. It kind of sounds pretty harmonic in a way. Sometimes, one could need a 4th layer. This is what I refer to as the bridge layer. It connects the 3 together and hops over from one layer to another. The phaser unit is ideal in this case, or using a notch filter that goes up and down, covering the highs and low frequency.

If you are going for a phaser unit, keep the original sound clean. Analog sounds usually work best in combination with the phaser when it comes to its harmonics. Having it subtle in the background will probably not have much of an impact. But knowing it is there is half of it. Taking the sound away, you will probably go like "huh?! Where did that sound go?!". And that is the key with layering the sounds.

So, in a nutshell... it comes down to the following example file

Download RNS file: Drone-setup12.rns
Download Mp3 file: Drone-setup12.mp3

Or when you want to go for the notch version instead:

Download RNS file: Drone-setup13.rns
Download Mp3 file: Drone-setup13.mp3

Anyways... I think this topic kind off sums up the process of creating ambient drone pads. Just keep in mind, one could keep things simple just by creating just one layer, but the more layers you use, and the more dynamics one would create, the more interesting the pad can become. Since these kind of pads become rather long in length, layering will be the key to success on this part. Hopes this helps, and see yah next week!

Often I get the question about: can I use the Mastering units attached to my instruments? Answer: yes you can. The mastering units were originally designed to be attached to the final mix, thus the mastering suite was born. However one can use multiple units inside the rack. A common example I personally often use are the MClass compressor and the MClass EQ. The other 2, the stereo imager and the maximizer, I rarely use as a output for a single instrument. Problem is, if I want to create a stereo effect, I know how to get it. I don't need a stereo imager to solve my 'issue'. The maximizer I prefer not to use as a instrument processing unit. Since it makes the sound louder and clips the output volume. Since I prefer not to audio clip at all, I personally don't have any issue with the levels of the individual instruments. But that is all side tracking the intro part by now...

Since I feel the need to look into some usage of the stereo imager, one might think "you don't use it, why write a tut about it?". Because I have seen some people use it totally wrong... thus, I need to make a point here by emphasizing what the stereo imager does, and how it could be aproached when using it as a instrument processor (rather then using it in the final mix).

Just to demonstrate how wrong the stereo imager can be interpretated... here goes one example file:

RNS Download: stereo-imager-wrong
Mp3 Download: stereo-imager-wrong

One might still not get the clue on this part. So... lets throw in a next example file, and notice what happends:

RNS Download: stereo-imager-wrong
Mp3 Download: stereo-imager-wrong

RNS Download: stereo-imager-wrong
Mp3 Download: stereo-imager-wrong

Notice the similarity in the second and third example. What the heck just happened? Well, let me try to explain. What the stereo imager has been doing for us in the first 2 examples, it cuts of a specific frequency (X-over Frequency) and it tries to seperate the frequencies over left and right. Depending on the frequency we are triggering. The amount of stereo seperation that takes place is determined by the amount of "Wide" and "Mono" we can apply on the Lows and Highs.

However... It does not work normally on mono devices. And that is basicly the point I am trying to make. Since, it also needs a seperated signal from left and right to do anything with stereo seperation. And that is the part where it usually goes wrong. A really easy method to solve the issue would be adding a Detune device like the Unison or Chorus of Thor. Since these 2 have make the sound stereo, and adds more width to the sound itself. One really important visual aspect of the stereo imager is that it will light up around the areas it has an impact on.

So, if that thing (where I am pointing the arrow at) is not showing any 'lights', it often means that the thing is either mono, or not working properly.

RNS Download: stereo-imager-better
Mp3 Download: stereo-imager-better

You might be wondering if this would be the ideal aproach to use the stereo imager, I would have to say: no. Since it makes it chaotic in a way. But this is the only method I can come up with to make the effect 'noticable' to the ear. You might also notice a total different sound output when the High Band of the last example file is turned to 'off' (put the knob in the middle so the High Band Width is set to 0). The chorus effect of Thor has some stereo effect, it seperated the left and right channel just a tiny bit.

Anyways. I just wanted to make this part pretty much clear before I would even concider to continue to discuss the thing I had in mind to work on. Since in the last few parts I showed a couple of tricks you could do to make it stereowise effective by detuning, using 2 different EQ's on both channels etc.. etc. However, the stereo imager is probably an easy mechanism to add a bit oomph and aaah on the ears. There are some features that some might no be aware of.

Since a pad is often long in lenght, we want to have as much tricks and tips we can pull out of sleaves to get some interesting stuff going on. That is why I am working on the third part on this pad madness. To get myself started, I am going to start with a single thor patch. This patch has 3 analog oscillators in there, with a minor detune. It uses 2 different Comb filters and filters up and down accordingly to the filter envelope. With in the end there is a Low Pass ladder filter, which also has a minor tremble in it since an LFO2 is connected to the thor patch. That in combination with a minor EQ drop on the lows, it sounds as the following:

RNS Download: stereo-pad-initial-step
Mp3 Download: stereo-pad-initial-step

At this point, most of the stereo effect comes from the RV7000. Since the reverb engine is always a tiny bit stereo effective. From this point of, we are going to go nuts on the stereo effectiveness...ses...

When we take a close look inside the Stereo imager rear plugs, one might notice that it has a default audio input, and output. However, there is also something called Seperate out. This one can be used to bypass just to highs or lows. A similar method would be using the solo button on the front of the thing (which has 3 options: both, solo high or solo low). So what we could do is use 2 times the same signal, use a low and high seperation and inner connect those to the mixer. But we'll be needing a spider splitter to do that (since we are using the same output of the thor patch twice). To make the stereo imager actually do its thing, I am going to plug in a Unison Detune in between.

As in the above picture it basicly explains on how the signal is being split into 2 seperate sections. Both go to their own stereo seperator, and then they combine again at the mixer. So with all this 'jazz' going on, what would be the big advantage about this setup? Well, let me put it simple. At the back of the rack we can flip around one for the High Ends, and the second for the low ends to stereolize. The X-Over frequency will determine the frequencies above or below which we want to put into stereo. Just a minor example file, I will present you the following (please note, in the beginning I am putting it all to mono, and then go heavy on the Effects so you can hear the differences):

RNS Download: stereo-pad-initial-step
Mp3 Download: stereo-pad-initial-step

Ok, with all this stuff going on, we could make it even more interesting. The next step I am going to undertake is make the lows wobble around just a tiny bit from left to right. To get this effect rolling we can simply add a single LFO parameter, put that thing in good use while routing it to the pan setting of the mixer 6:2 (the mixer 14:2 works the same way).

RNS Download: stereo-pad more completed
Mp3 Download: stereo-pad-more completed

This was just a simple way to add stereo mix to your pads when you really need them. In most cases you would probably leave the stereo images along and adjust the left/right output signals manually (at least, I would). But these setups can have a nice addition when you are struggling with the points on "I am missing something". Adding a little bit stereo can widen up the whole song just a bit. Especially with pads. Since they mostly tend to travel over alonger period of time. So that might make things just a bit more interesting.

In the next part I just want to stand still at a similar technique but then spicing up the drums just a little bit by doing something really 'simple'. Till next week. And a happy new year!

When you open up the remote programmer of the NN-XT, you would probably go nuts around the amount of knobs that are inside the NN-XT. However, it isn't that hard to understand its full capacity. From the bottom right to the bottom left you have the basic settings such as amplitude envelope, filter, pitch, modulate envelope, LFO2 and LFO1. When looking at a normal synth like the subtractor or thor, one can control these using the edit automation feature. While the NN-XT is lacking this option, there are ways to take control of them using a different method.

The basic controls we have are the

- frequency and resonance (filter settings)
- attack, decay and release (amplitude envelope)
- modulation envelope decay (modulation envelope)

So indirectly, we already have the options to control most of the important aspects of the NN-XT sampler. When loading up a patch, we could use these knobs to control the volume, how the sound evolves over time, and we could control the filter settings. Just as a minor example, I am using a jovian pad with a minor adjusted attack on the amplitude envelope.

Download RNS file: Jovian Pads
Download MP3 file: Jovian Pads

At this point theres not much special going on. The special part would be that we could make our own pad from the ground up using an NN-XT sampler. Yet, this would also mean that we need to understand what the NN-XT thus, and so forth. The basis of the Jovian pad in this case is a sample called MKS_EUPHORIUM. It uses 4 different layers per note. The first layer is the default sample. The second layer is the same sample which are 2 octaves lower, the third layer is the same exact sample yet with semi set to 7th (meaning it goes 7 notes higher) and the fourth layer has a semi set to 7 and the octave set to 2 octaves lower. Other then that, most of the layers are the same yet the panning is somewhat different.

At some point we could look at the NN-XT as our mini combinator patch, since we can layer as many samples as we like, do all kinds of internal routing and add some automation inside the patch itself. I know, this part will make your head spin probably. So, lets just take it one step at a time.

Euphorium

Let's just create a similar pad, based of the same sample and go from there. We initialize the patch, open up the programmer and load up the samples MKS_Euphorium C2 till C5.

Download RNS file: Euphorium Mk1
Download MP3 file: Euphorium Mk1

Default it would sound like that.

Problem is at this point that the sound is somewhat layered wrong. So we need to fix that.

Download RNS file: Euphorium Mk2
Download MP3 file: Euphorium Mk2

To get the similar approach on the Jovian pad with the 2 octaves, we could play our NN-XT sequence as the following way:

Download RNS file: Euphorium Mk3
Download MP3 file: Euphorium Mk3

Or we could duplicate the samples inside the NN-XT and move the octave up with 2 on the sample level. However, one could also use different samples inside different zones. Like I have mentioned before, it is a layered thing. So we could combine as many samples as we like.The only limit is the cpu in this case.

In the next setup I am adding a new sample. This one comes from the Jp4 series (pads) and is the FltrPad1-03_D3. Since the sample is recorded on a D (notice the D3 mentioned inside the sample name) we need to adjust the root knob to correspond with the notes. So, root knob goes to D3 in this case. However, the only problem with this perticular sample is that it tends to jump right on top of the rest of the pad we created so far. One way to solve this issue is by adding a little bit filter cutoff to this specific sample.

Download RNS file: Euphorium Mk4
Download MP3 file: Euphorium Mk4

At this point we could copy the sample sample (the fltrpad1-03_d3) and make some minor adjustments in the panning and tuning as well, or setup a different filter etc.

Download RNS file: Euphorium Mk5
Download MP3 file: Euphorium Mk5

Even though the adjustment is just 'minor', it is there. And that is usually what counts with layering this kind of stuff. When you would remove it, you will notice it that it is 'gone'. That is why 'micro managing' an instrument is sometimes important to do.

Last part I am going to add right now is a Saw tooth form.This one comes from the snyth leads -> M20 raw saw. It is a bit tricky to just kick in a raw sawwave inside the sound that we already at this point. It kind of distinct itself from the rest once again. To solve the issue here, I am using a modulation envelope which is connected to the filter setting. Which means that the filter goes up and down (since I am using a minor attack on the modulation envelope).

Also in addion, I have setup the samples to connect to the velocity level. This means that the harder the noted are being triggered, the louder the sound is going to be. Again, this can be set differently using the different layers. One could use one layer to be controlled with the same velocity either have a couple of diverse settings where each sample triggers a different velocity.

Download RNS file: Euphorium Mk6
Download MP3 file: Euphorium Mk6

RPG 8

Just to go offtopic on the matter, or at least somewhat on topic at the same time. We could shape our pad differently using different routing things etc to create a dynamic part inside our part. Often a LFO would be idea in this case, since it gradually changes the progress of the pad. However, there are other ways to shape it. One method would be using a Matrix pattern sequencer and create a curve. However, I am going to go for a total different approach on the matter.

By adding a RPG 8, without any connection, we could use this one to create a pattern and make something rhythmic from it. In most cases the RPG 8 is being used to make arpeggios. However, what most people don't often know, is that it can also be used to shape the sound. Just to explain this part a bit: the gate from the Arpeggiator creates an CV output that goes from 0 till 96 in most cases. This is based on the idea that we have 8 octaves, and each octave has 12 notes (C, C#, D, D# etc..). So, with some adjustments we could use the RPG 8 cv gate output to recreate a pattern and connect the pattern to any setting as we see fit.

To have something really noticeble going one, one might prefer to use the octave settings of the rpg 8 and use octaves 2,3 or 4 to get most of the noticable effect (since the setting will jump up and down more then just 12).

So when we create a random synth, or sample... we could use the gate output of the RPG to create a single pattern and connect it to for instance to a filter setting. This will mean that the notes act as a filter setting, and are not just sounding like notes anymore.

This would result into the following effect:

Download RNS file: Euphorium Mk7
Download MP3 file: Euphorium Mk7

Anyways, have a merry Christmas and an outstanding new years eve. And with a little bit of luck, I am adding a new chapter after the first day of 2010. Until next time!

Pads are often used in a very wide range of music genres. They are often tend to be long, sweeping and long lasting in length. It totally depends on the the wave form, envelopes and filters on what makes the pad. However, the whole pad thing will be the one that is most likely one of the most time consuming parts to set up (next to the leads). A lot of times a simple pad works, however, one might find it daring from time to time to stack a whole bunch of layers on top of eachother just to make the difference between a simple, and complexed pad session.

In this topic I will start of simple, give a couple of ideas, and then go to the more complexed approach. One might get lost at some times when I make certain discissions inside this topic in specific, however, I think it is also about using your own ideas and experience to make the perfect pads for your own productions. And to add, one might notice the amount of love I will put in this topic, since I just love pads and leads... since they are a real important part for the production parts.

What instruments to use?

It is often a question that comes to mind. In a lot of times, I prefer to stick with analog as a basis, however one could also use real life instruments. In most cases one could stick with the malstrom for thin layering (since the malstrom isn't that thick in sound), or use thor to get most out of the synths. Other setups would be sticking with the samplers in alot of cases. Even though samplers are limited at what they can produce (since it is based of a single sample) one could use multiple ideas and layers to get different sounds from them. In a lot of times using a combinator patch would be the ideal approach.

Synths

Lets gets started with a simple synthetic kind of pad. Like I said, I prefer analog sounds. So one might find me using the wave forms such as saw, triangle, square and sines very often. Since these are the most basic waveforms to use inside a complete arrangement. And they tend to be mostly harmonic in most cases. Just as a simple form, the first example I am going to show works pretty straight forward. I am using a thor patch, stick 3 different analog oscillators in there and use a minor bit of detuning (different tune settings). Second there is low pass ladder filter involved. With using a high attack in the filter envelope, it means that the filter will gradually climb up, until it reached a specific time. Next to this is the amplitude envelope. This setting makes the volume gradually gain momentom. In a nutshell, these are the settings I have changed:

And it basicly sounds like the following example file:

RNS Download: setting up a basic pad
Mp3 Download: setting up a basic pad

For the next example I am using a similar method. Difference is there is a different wave form used to shape the pad. Also there are some other minor changes made inisde the Wave BandWidth.(BW) to shape the waveforms a little bit different (rather then using its default tone).

RNS Download: setting up a basic pad
Mp3 Download: setting up a basic pad

While using for instance the Malstrom, the basic princible is still the same thing. At this point, it is a different waveform (string with minor motion), and attached a bit of attack on the amplitude envelope.

RNS Download: setting up a basic pad using a malstrom
Mp3 Download: setting up a basic pad using a malstrom

When it comes down to the clean subtractor (with no effects attached to them), it often tends to become a somewhat dull sound. However, there are some ways to even get some stuff done with a few subtractors left and right. However, in a lot of times one might want to use them for background or support orsomething in that specific form. In other setups, the subtractor is more ideal for creating basslines or leads in this case, since most of the other synths are more versatile to create some great pad sounds.

RNS Download: setting up a basic pad using a subtractor
Mp3 Download: setting up a basic pad using a subtractor

Last but not least, the samplers. Personally I prefer the NN-XT over the NN19 since there is more control from the initial sampler (remote editor). However, when you prefer to go heavy on the automation then at some point the NN-XT might be limited to use (which totally depends on the knowledge on how to set the remote editor up).

At some point the samplers are pretty much straight forward to use, since one only needs to load up an NN-XT patch or NN19 sampler, change in most cases the filter envelope/ amplitude envelope and one could be done. Just as a quick example:

RNS Download: setting up a basic pad using a nn19
Mp3 Download: setting up a basic pad using a nn19

So till this point most of the stuff I have gathered here are pretty much straight forward stuff: I pick a waveform, select a filter, create a filter envelope next to it (most likely with a high attack, since it sweeps), and last I have the amplitude envelope settings where the attack is often a little bit more to half / high, and the release high. When it comes down to a more complete approach, I personally stick with combinators or stack a whole bunch of different pad like instruments on top of eachother. The reason behind the 2 is: layering. There is a huge amount of difference between just having a single oscillator playing a single sawwave compared to having all kinds of different wave forms joined together. Also, even if it is just a minor twist inside a complete patch, it can make a difference.

Stereolizing a pad inside Reason 4

Just to demonstrate the whole layering approach on an easy level, I have setup 2 different combinator patches based on 2 of the patches I already mentioned. The first one is based of the basic pad using the subtractor, the second one is based of the NN-19 sampler patch. The minor difference is, that there are 2 subtle changes. First, there is a exact copy of the patch and duplicated inside the combinator. With having a minor detuned setting (tune up, or tune down) it already sounds a minor bit more present. Second, the 2 different layers are stereo wise more effective. One patch goes to the left, the other goes to the right. So, in a way, this is somewhat as a replacement for the Unison, yet difference is: we are at full control here.

With the second patch (nn-19), I took it one step further. The sample itself is detuned, however the next difference is the filter setup. Both nn-19 patches uses a Bandpass filter. Yet, the settings are somewhat 'off'. Therefor, it creates this minor bit of additional space inside the sound itself once more. The same trick applies to using EQ-ing for instance, subtracting different bands and moving them around will definitly give them just that little amount of space inside the patch itself.

RNS Download: setting up a basic pad using a subtractor stereolized
Mp3 Download: setting up a basic pad using a subtractor stereolized

RNS Download: setting up a basic pad using a nn19 stereolized
Mp3 Download: setting up a basic pad using a nn19 stereolized

Layering pads inside the combinator

One specific part of the whole process will offcourse be the layering inside the combinator. And this is also the frustration part about the topic itself: since the amount of layering is unlimited, and the amount of different effects is almost unlimited. So, the first though might be, what to do... Often when I get around into the same stuff inside the combinator, making pads and such, I often tend to start of simple. It just depends what the purpose of the pads are. Do they need to be dark, or bright and light.. Or, do they have to be very dynamic in the long run, or can they be pretty long and boring. There are all kinds of different discissions one can make during the process of making just a single pad. That is probably one of the reasons why pads and leads are most often the sounds themselves which take most of the time just to tweak. Since these will often set the tone of the whole music project (while basslines are more about the feeling).

One of the processes that could be done here, is the same setup I just did to go from 'simple' to 'stereolized'. One starts with a simple instrument, builds it up.. and thinks: Ok, this is the fundament right there. And builds on top of that. Using this process, I will start with one example. In this case I am starting of with a simple thor patch based on the multi oscillator (since these work pretty well as pads themselves).

Just to get myself started with this one, I have created a combinator, added an inline mixer in here (since I tend to use more then just one channel), and added a single thor patch. The initialized thor patch contained a analog oscillator, so I replaced it with a multi oscillator, increased the amount of detune just a little with and removed the initial low pass ladder filter. The control of the pad is in this case handled through the amplitude envelope only (so no fancy filtering yet).

RNS Download: Basic pad based on thor multi osc
Mp3 Download: Basic pad based on thor multi osc

From this point on we could use the same cheap method of using a chorus which applies a bit of stereo amount into it. Personally I prefer to make my own stereo sounds. So the next step would be doing the same thing as I already did before. Recreate the same thor patch (copy patch, add thor then paste patch on the second thor patch). Make some adjustments and send the to thor patches to left and right.

This should look somewhat similar as the following setup:

I had to minimize the 2 thor patches, since the screenshot would become a bit to big. The only main difference in this case is that the detune amount is somewhat different. With the first thor patch the detune amount is set to 33, with the second it is set to 28. In the file itself it looks (and sounds) like the following:

RNS Download: Basic pad based on thor stereolized
Mp3 Download: Basic pad based on thor stereolized

At some point, it would work similar when using the programmer of thor in this case. Since it also an option to use OSC1 to Left in filter 3, and OSC 2 to right in filter 3. However, this method also comes with a minor issue which needs a workaround again (Amplitude envelope is skipped). In this setup, with 2 thor patches inside the combinator we don't have this problem. Yet again, I just wanted to point out there are mulitple ways to solve all the different patches / internet routing.

Next difference I am going for with this setup, is use 2 different chorus flanger effects. Notice that the first thor patch is going to the left channel, and the second thor patchs goes to the right channel. With this setup, it is also an option to use 2 different chorus setup. This is almost similar to the stereolized chorus which default comes with thor. Difference is, we can control the amount of stereo seperation by hand if we like.

RNS Download: Basic pad based on thor stereolized chorus
Mp3 Download: Basic pad based on thor stereolized chorus

From this point on we could use all kinds of different filters, EQ's, reverbs and what not and attach that for instance to each individual thor patch or inner connect it to the final mix of the combi. Ofcouse we could change the inner routing of each thor patch as well. By just using a couple of LFO parameters and connect it to the filters and such.

To make most of the usuage of stereo, one could use 2 different settings on the waveforms for left and right channel. Other methods would be using a different notch filter. Notch filters are great for widening the sound stereo wise. One could use the default state veriable filter and use the Notch setting. However, problem is that it does not really show much what the filter is doing. Preferably one could use an EQ (Mclass unit or 2 parameter EQ) instead to change the shape of the notch filter. The basic idea behind a notch is that you cut down the middle frequencies. And the gap of the Notch is determined by the amount of "Q" applied to it. The bigger the gap, the more you will loose from the original sound itself (which also depends on where you are cutting the frequencies). For pads it is most likely about the mid range of the sound (500 Hz till around 6kHz). So cutting it there will definitly make a (noticable) difference there.

RNS Download: Basic pad based on thor stereolized EQ
Mp3 Download: Basic pad based on thor stereolized EQ

Movement inside the sound while it travels

From this point it is also an option to create movement inside the sound itself. A few options to play with is to change the output of the left and right channels while the sound progresses. The easiest method would be using an LFO of somekind, since these are often based on basic pattern that repeats itself (and does not require much automation). Other methods would be tieing a matrix pattern sequencer and use the curve instead, or use other type of methods that have a cv output.

Just as an example, we take an additional malstrom graintable synth, and we'll be using the mod A (modulator A) just for signal processing only (not for audio output).

in this case we'r conntecting the LFO to a CV Spider merger splitter. The original signal is going inside the CV in, and we'll connect the CV out (slot 1) to rotary 1, and CV out inverted (slot 4) to rotary 2. This means that the rotary 1 and 2 will rotate inverted from the original LFO signal. The only thing that remains is to connect it to something to 'rotate'.

In this case I am using the Parametic 1 Frequency from the 2 different mclass equalizers. Which means that the frequency (notch in this case) will change along the time line accordingly to the rate of the LFO. In a lot of times you might want to find the right ranges to be the minimum and maximum though.

RNS Download: Moving sounds
Mp3 Download: Moving sounds

The mixdown of the sound

From this point on we could take the pad still anywhere just by mixing it down on the final output signals. One could choose to use the indepent channels (left and right seperatly) or use the final mixdown which goes towards the main input of the combinator (which is in this case a mixer 6:2). A few options to throw in here would be the following: reverbs, delays, additional EQ, distortion, phasing etc. As long as they shape the whole sound to make the final tweaks.

RNS Download: Mixdown of a pad
Mp3 Download: Mixdown of a pad

As everything discussed above basicly applies the same way as with other devices (even though I only used a single thor patch in this case). One could stack up stuff that high, that it could become one big fat pad with a whole bunch of different stuff going on. For instance, one could focus one instrument to be the center point, add 2 more instruments which go for left to right and have a forth one do a little bit of phasing for the dynamics.

As an example:

RNS Download: Padzorized mini song
Mp3 Download: Padzorized mini song

RNS Download: Some minor pad with some mixing down
Mp3 Download: Some minor pad with some mixing down

Granual synthesis isn't something new these days. Just throw in the words like "Malstrom graintable", and one would already know what I am refering to. Another tool, which comes with Fruity loops is the Granualizer. The benefits of the granualizer of FL is that is uses samples rather then a synth with a fixed amount of presets. I already explained somewhere in the past that it is an option to re-create the grains effect by using an NN-XT sampler in combination of the RPG-8. With this topic, I just want to focus a bit more about the series of effects we could create with using grains... since at some point, they can be rather interesting. So, let's hook up, grab some grains and start throwing them towards the speaker.

The Malstrom

As I initially stated, when thinking grains, we would think Malstrom inside Reason. Since it holds most of the keys on what granual is in this case, and it is easy to use and setup. With the amount of 'weird' looking voices, which we probably wouldn't be using most likely as a lead, we could easaly create some effects using grains. Just stack up a Modulator A, or B and throw some parameters to them and we are done.

One might need to understand what the grains do, what index is, and what motion does. Granual synthesis is like taking a complete sample (wav file, or AIFF file) and slice the whole thing up in smaller sections. At this point it is similar on a Dr Rex loop thing. However, the difference is the way that granual synthesis plays.

The synth will play the slices in a sequence, yet with different 'pitch changes' to correct the waveforms. Making it sound more gradual. Each "slice" is called a grain. Each single grain can be controlled by using different envelopes or effects. For the malstrom graintable we have effects such as shift and motion to control the grains. The motion controls the speed of each grain (and how fast the grains play in a sequence), yet shift pitches up a single grain.

To control the different grains we could use lfo parameters to change the grains position (which is in this case refered as "index"). It is kind of comparable with having a sampler and controlling the start position of the sample on the fly.

With this basic understanding about what granual synthesis is, we could use this to our advantage to make instruments, or sounds on the fly as we see fit. Since granual synthesis is a very powerfull tool to create something 'weird' from time to time. And that is probably why granual synthesis distincts itself from a normal way on playing a sound.

Just as a minor example, I have setup an example file where I am using a waveform called "additive wave #1". This sample usually sounds more like a pad kind of thing when playing it with a normal grain sequence. However, when reducing the motion (thus slowing down the sequence time on how the grains play), it starts to sound more like a cheap c64 sound effect.

By using an additional Modulator A, we could for instance control the index or shift (since those are the 2 parameters which in this case the Mod A can control). The index can be speeded up or slow down (thus increasing and decreasing the speed on how the grains play in a sequence). The shift makes the sound pitch up and down more or less. The benefit of shift though is that it does not slow down the sample as it usually would do with turning down the pitch knob for instance.

RNS Download: granual-movement.rns
MP3 Download: granual-movement.mp3

To create space inside the malstrom, it is an easy way to use the filter A and filter B slots and connect them to Oscillator A and Oscillator B. In this case Oscillator A will travel through Filter A (which is the Left channel) and Oscillator B through filter B (which is the right channel). To emphasis the space effect we could use methods suchs as tuning, or a different amount of motion to create the space inside the sound. As long as the sound is almost similar, yet somewhat different on the settings (not too much, since this will probably lead into a total different sound).

Yet, while we are at it, we now how a single ping-pong sound based on an additive wave inside the malstrom. There isn't much hot-wiring in this patch, and it is pretty straight forward. However, we could make things just a little bit more alive by using the same LFO (or modulator as it is called inside the malstrom) and connect it to the panning of the embedded mixer.

RNS Download: granual-movement2.rns
MP3 Download: granual-movement2.mp3

Another option would be using an stereo imager instead. However, I personally always find the stereo imager more dull since it is often very straight forward when it comes to the output (since it is controlled by the cross over frequency, and not by a pattern).

A different method would be using different patterns, and make them go faster or slower. It just depends on the type of effect you are going for. In most cases, I prefer to use the 'rhythmic' patterns work best when the pattern plays fast (high rate), while the smoother patterns (such as sine form and such) play better on a slow rate. Since these are more gradual and are often best to make really long and dynamic sequences for effects such as these.

RNS Download: granual-movement3.rns
Mp3 Download: granual-movement3.mp3

So at some point a granual synth would be an ideal method to shape something rather weird, dynamic and sur-real. Another method would be using a wave generator and slide shift the position of the starting point. At some point, Thor is most likely idea in this case, and specificly the following:

- Analog Oscillator and the Square form in combination with the Pulse Width
- The Wave form in combination with the Position change
- The Phase modulation and the Phase modulation amount

These mentioned types have a similar dynamics when it comes to shifting the sound along over time. For instance, connect an envelope to the pulse width, position change or Phase modulation amount woud be one method.

When we would look into yet another method to make granual sounds, we could for instance go for the Arpeggiator -> NN-XT method. Where the sample start point from the NN-XT is being connected to the Arpeggiator Velocity. Eventhough, I have explained this method before, I never gotten really in depth about it. To setup the Arp/NN-Xt method, one could use the Velocity of the apreggiator to make the sample start point of the NN-XT change over time. One setup would be doing it manually (Edit automation) either hook it up with some dynamic pattern (LFO/Modulation/Matrix pattern sequencer).

In this instance, I am using a longer sample from the Other samples section inside the reason factory soundbank. Most of the samples inside the other sound samples are often long sound effects, stabs or other effects. And will sound very common when hearing them. However, we could 'recycle' the sounds from any given bank, and throw in some grain effect such as the following 2 example songs:

RNS Download: Granual NN-XT.rns
Mp3 Download: Granual NN-XT.mp3

With this example, the sound is more contant in the background, and serves pretty well for a chill out tune sort of. It is kind of like a random piano banging sound in the beginning which can not be placed as being a piano... but it sounds 'odd' at some point. Sound like these may sound odd from time to time, since these kind of sounds are often used as a background noize effect yet rarely take the uphand, but would fill up the track very nicely (just mute the 2 combinator patches for instance and it would give a whole different meaning to this piece).

As the granual effect could also be manipulated using automation features from the combinator (or directly from the patches inside the combi), it is most likely that a custom sound that you are creating will often be changed during the progression of time, or re-used differently inside another production.

Just as a test case, I have setup the following song to demonstrate this:

RNS Download: Granual NN-XT2.rns
Mp3 Download: Granual NN-XT2.mp3

The method I am using in this song is the same, yet using a different sample attached inside the NN-XT and I am fiddling around with the frequency knobs and velocity. It gives a whole different meaning to the song when one would compare it with the previous song. First of all, the drum track is different, setting up a total different mood in this case. Next, the granual sound changes more often over time (and more noticable in this case).

Granual effects are probably also good for sweeping effects, combined with an lfo parameter connected to the Velocity changes (as long as the velocity is connected to the sample start point that is, otherwise this won't work). For instance if you would take a longer sample that is based on a synthesizer sweeping pad sound. One could use the sample start point of the grains to move up and down, rather then using a filter knob in stead (which also works offcourse). The difference between the granual method, or the normal play sample and connect it to the filter method would be that the sound itself could become more metallic either more surreal. Since we are basicly repeating the same notes, with a longer release while the start point of the sample changes over time.

In this example, I am using the pad sounds form JP4 filter pad (nn-xt -> Pads -> JP filter pad) section. By loading up two times the same sample, we could for instance create depth inside the sound itself by using a different tune setup for each different sample. For instance, the first sample goes left with a minus tune, while the other goes right with a plus tune. For instance:

Left channel (first sample)

Right channel (second sample)

As a stable sound effect it would come out as the following effect:

RNS Download: Granual NN-XT3.rns
Mp3 Download: Granual NN-XT3.mp3

What also is an option here to do so, is to create a difference into the starting point of the sample which is in this case connected to the velocity. This would definitly mean that the 2 different channels are seperated from eachother. This would be a similar effect when using a malstrom and create a bunch of spread on the master out, and use a different motion setting on Osc A and Osc B.

RNS Download: Granual NN-XT4.rns
Mp3 Download: Granual NN-XT4.mp3

However, the only sad part is that there is not really a volume glide part with a technique like this on the using the velocity, since the velocity is connected to the arpeggiator in this case. The only few workarounds would be using a LP12/LP24 filter cut off and connect that to the final mix. And use the programmer of the combinator to sligthly move the threshold of the filter up (so it filters in, and thus this will result into a gradual output increase in volume aswell). Or, one could tie the master volume knob of the nn-xt and connect it to one of the rotaries inside the combinator (using the programmer part). In this case the rotarie could act as a volume balancer.

So there are multiple way to solve that part.

Setting up a granual effect using the NN19 and the matrix pattern sequencer

The cool thing about the nn19 is that this device has a sample start point as a single knob. What we could do is use this single knob to be inner connected to a matrix pattern sequence curve. To set this method up, we will have to create a combinator, add a nn19 sampler, add a matrix pattern sequencer (without its default connection), connect the curve from the matrix pattern sequencer to the rotary 1 of the combinator.

Then open up the programmer of the combinator and connect the Sample start of the NN19 to the rotary 1.

In this setup, it will mean that the matrix pattern curve is now inner connected to the sample start (and it will visually show this inside the rack aswell). The only thing that remains is the curve itself. In the next example file, I am using a simple method of changing 5 different patterns where most of the patterns are 4 bars in length. I just randomly picked the pattern, so there is no "thought" behind it, but it is the idea that matters. Since at some point we could create something rhythmic again, by using a single solo sample which plays a complete sequence over time.

RNS Download: different-method.rns
Mp3 Download: different-method.mp3

The cool part about using this kind of method is that it works as well as having a lead, or something that plays in the background. It is all about how the EQ'ing is handled I guess, and how present the sound would be. But in most cases it will probably sound pretty much glitchy at some point. Which also has its charm.

However, when it comes to granual synthesis, one might get most benefits from the Malstrom. However, from time to time, one might need a single sound sample to achieve the same sound effect one is looking for. Just to sum things up, I am going to conclude this part with a more or less "complete" sequence. This one is more based on Granual synthesis coming from the malstrom itself. However it just shows how powerfull it can be to throw in a bunch of grains and throw them all up in a sequence. Notice that I am using sounds as "Metal Cans" and "Thunder" to create these effects ;)

RNS Download: granual-effects.rns
Mp3 Download: granual-effects.mp3

In addition here is a piece I have worked on this afternoon(12th december), which uses a lot of malstrom grains and moving around on the grains while the song progresses over time:

RNS Download: granual-malstroms
Mp3 Download: granual-malstroms

In the last section I focussed mostly on the noise oscillators of thor and give a couple of examples on how we could approach on to creating sweeping sound effects using that. In this perticular part I want to focus more on the electronic music style of effects. These are kind of random when it comes to its sound design, yet that is also why they would distinct themselves from the other sound effects (such as reversed rides, crash cymbals and wind effects).

electronic music effects are often generated by an analog type of sound. This is probably most likely why the sounds are different then any other 'real life' instrumental effect. They work pretty well in most electro house type of genres (since the music style itself is already in that type of sound), however one could for instance also try to experiment with the basic concepts and use them inside trance, dnb or other type of music genres.

The basic understanding of LFO patterns

LFO patterns are often one of the key features to setup a sound effect, time it right, and get the parameters to change as you want them to change. As an overal tip, I can say the following: if you have no clue how a pattern inside any LFO would work, you might concider connecting the pattern to the Pitch of the oscillator/waveform. For instance, when you take a malstrom, it has over 32 different patterns in there. By using the Modulation A of the malstrom, one might start to understand just a little bit on how the pattern changes the complete waveform when applying a little bit pitch. Just as a simple example one could learn understanding it with the following basic template:

RNS download: Pitch LFO
MP3 download: Pitch LFO

Making some rhythimc effect

Most of the patterns inside the LFO can be used as rhytmic patterns. So, in a way we could make a trembling pulse generator and slightly let it fade in. This could be some form of replacement for a sweeping in kind of sound effect and quite interesting from time to time. Since there are 2 things changing at the same time: the volume (by the amplitude envelope) and the amount of trembling due to the fact that the volume is increasing, thus the effect becomes more noticable while the sound progresses.

Just as another simple example:

RNS download: Pitch LFO Fade in
MP3 download: Pitch LFO in

The idea behind an effect is just simplicity by itself, it repeats the same pattern over and over again and in this case the sound is being changed with the help of the amplitude envelope attack (set on pretty high) and second the attack of the filter envelope. Where the filter is attached to a comb + filter setting.

To make the sound become just a little bit more rhythmic based, we could for instance apply a mod B to A set up in this case (since the malstrom allows us to do so). When we use the Mod A amount inside the modulation B, this will mean the the parameters change along with the amount of modulation which we apply to it. This method is often handy when one wants to create patterns based longer then a 16/4 pattern base.

Just for instance, I am using a half square lfo for the modulation B, and this results into a following effect:

RNS download: Pitch LFO Fade in
MP3 download: Pitch LFO Fade in

Even though this almost sounds more like a lead, since it all based on a sawwave form. I thought using a sawtooth just to make it noticable what we are trying to do here. But there are some descent voices that we can use by using the above mentioned technique just to make something sound like a soundeffect. For instance, take a voice called throat. Throat by default has the tention to just be a kind of voice sound effect. However, by slowing down or speeding up the motion settings, one could create some cool effects with them.

RNS download: Pitch LFO Fade in
MP3 download: Pitch LFO Fade in

and imagine it with a tiny bit of drum and bass thrown agains it...

RNS download: Pitch LFO Fade in
MP3 download: Pitch LFO Fade in

What is happening here is the following, the Mod B act as a modulation on the Mod A. Which means, while mod b is changing, it will change mod a at the same time. Yet, keep in mind that mod A is changing as well. While mod b is set on 3/16th, it means that the modulation on mod A is also being changed on a 3/16th part yet, the change amount is determined by the Mod A amount on the Modulator B (that just rolls of your tongue right there...).

The effect gets some emphasis because I threw in a little bit volume and a little bit filtering on the modulation B. Because it is just a minor setting (minor increase and minor decrease) these changes will be subtle, yet noticable while the effect gains volume over time. So at some point it is a rhythmic thing that is going on, while on the other hand it is just a random sound effect due to the pitch changes that take place inside the modulation A.

A similar setup we could make inside Thor, yet it requires a bit programming since default there is no Mod B to Mod A. But what we could do is tell thor to use an LFO 2 and connect it to the Amount (or Scale) on the LFO1. Which means that the LFO1 is changing amount by the amount of LFO2 again. Which can be a cool effect from time to time.

Just like the following example, where I am using the programmer to change the lfo amount using the scale and connect it to the position of the WaveForm (Wave oscillator):

RNS download: Pitch LFO Fade in Thor
MP3 download: Pitch LFO Fade in

It could be used as some form of lead, however it would probably sound odd as a lead. But I do think that effects such as these are often the key between having something complete as a sequence. At least, there are some points when someone might miss something inside the sequence and often something 'simple' as this might just fill up the bits and pieces to tie certain parts together in the song.

In a lot of cases the LFO will be a vital part with setting up constructions such as these. Since they determine the progression of the sound. Also make use of some descent envelope settings, since these determine the starting point and the ending of the sound. For sound effects such as these, I do believe that 'gradual' gain works best, since having them thrown in in an instant, it might scare the listeners socks off.

I also believe that these kind of sound effects work like a charm in the background, so make sure to go easy on the gain on these. Even though I haven't done it yet, since I am just here to demonstrate the noizy effects and such... but it is something one could keep in mind when making patches such as these. On the other hand, if the producer finds it too loud, one could just easaly drop the gain by using the mixer or the overal output level of the instrument.

Combinator

The combinator is probably the ideal solution for making complexed patches such as these. Since the combinator allows unlimeted amount of instruments to stack, unlimited amount of CV wiring, and ehm.. did I mention the word unlimited already?

For instance, lets make something weird here. To start off, I will add a mixing panel (it can be 14:2 or 6:2, we are only going to use 2 different channels). Next, we'll connect two times a subtractor. These 2 subtractors will have to play the same sound effect, so we might start working on one (mono) and copy/paste the patch when we are done.

Lets just start of with something simple as this. We'll mute channel 2 of the mixer panel (since in this case it just plays the same sound twice). And other then that it is the default router. The first subtractor goes into channel1, the second one goes in channel 2. I will explain why I am using 2 subtractors soon (ps, I am using a subtractor in this case, but the idea can be applied to any instrument or synth).

Next up, lets shape it to become somewhat like a strange sweeping synthetic effect thing. Initially, I am going for a bell shaped kind of sound effect this time, and make it swell during the progression. Wave form is set to 10, using osc 1 on octave 4, and osc 2 on octave 2 (so it is 2 octaves lower for the second oscillator). I am using a bit of tuning to make the sound tremble a bit (+8 and -8).

From this point on we could still take it anywhere. So, I am going for a modulation setup to kick in the phase modulation. To make the Phase workable, we'll have to use Phase mode set on - (subtract) or on X (multiply). Subtract usually works best, multply can often have some strange pitch events making it sound offnote.

Next part is the modulation envelope. In this case I am connecting the envelope to the FM. I am using FM this time, since FM works very well having 2 oscillators on 2 different octave settings. With having a low octave setting, and a low attack on the modulation envelope, it means that the amount of FM is increasing over time. It is kind of like editing the automation on the FM, where the FM knob increases. But in this case, it goes automaticly without having to touch any additional knobs.

This all results into the following sound effect:

RNS download: Combinator Effects
MP3 download: Combinator Effects

Another interesting effect inside this specific setup would be using LFO2 of the subtractor and inner connect it to the phase. This makes the Phase knob go up and down accordingly to the shape of a sine form. If the effect is subtle, it could work... but in most cases I would go easy on the setting. When you would apply a little bit delay to the LFO2, it will make the phase modulation kick in just a little bit later.

Now for the juicy stuff... From this point on we'll just apply a bit of filter to it with a bit of envelope.

RNS download: Combinator Effects
MP3 download: Combinator Effects

In this stage we have some descent swell effect going on. Still we also have the ability to apply an lfo 1 for the subtractor and could apply that aswell. In this case I am going for somewhat like a Rhythmic LFO pattern, which is connected to the filter. The amount is minor in this case.

RNS download: Combinator Effects
MP3 download: Combinator Effects

So, by now we have some form of the basis on how the sound could travel like.At this point, it is probably a good time to bring the second subtractor and go wild on that one aswell. First I am going to copy and paste the exact same patch inside the second Subtractor (right click on the first subtractor, Copy Patch, then right click on the second subtractor, Paste Patch). Next I am going to unmute the second channel on the mixer, so both subtractors play the exact same sounds. Last part, the mixer channel of Channel 1 goes Left, and the Channel 2 goes right.

From this point the sound effect is starting to sound just a bit more spacey. And from this point on we could go totally mental. To emphasize the space effect I will be using an additional LFO setting. This one comes from the Malstrom, but it could be anything. I am adding a Malstrom, but I am not using the audio signals. Since it will in this case serve as an additional LFO setting.

The LFO parameter I am going to use is the Mod A this time. And use the signal twice, we'll stick this inside the CV merger splitter.

Notice inside the CV merger that the output signal slots of the CV spliter has a normal and a inverted slot. What we could do is use the normal CV and stick this inside Subtractor 1, yet use the inverted version and stick this inside Subtractor 2. Which means that the parameter we use is inverted. But, because we are using Subtractor 1 on the LEFT, and Subtractor 2 on the RIGHT, it will mean that the signal goes from left to right, and right to left. Which can be a cool effect from time to time.

RNS download: Combinator Effects
MP3 download: Combinator Effects

Anyways, this was just another idea on how to setup a long lasting sweeping effect of some kind. Even though, I ended up with the combination of 2 subtractors and 1 malstrom, keep in mind that it works with any instrument. As long as you setup the envelopes right, and use the correct wave forms to search for the sound effects you are looking for. Just be creative, experiment with it etc...