|
Trance & impOSCar - A Love Story.
|
View this Thread in Original format
| Derivative |
This is a brief tutorial on how to build a few trance type sounds using this remarkable synth. Part 2 is coming tommorow or the day after or whenever I can be arsed and its a short addition to this tut which tells you how to build a sound very similar to the 'verse' in Binary Finary - 1998.
Heres a clip:
You need in no particular order:
A Digital Audio Workstation (I'm using floops for this tut)
ImpOSCar (obviously)
We will of course be building sounds from scratch because it puts hairs on your chest .
Load up ImpOSCar in your DAW and start with a totally blank patch. The default blank patch - template.fxp is bad because it turns every rotary down to 0 - even those which have bipolar values and those which cut out the volume completely or transpose it so low that its inaudible.
This means that in order to hear anything you have to turn the LFO amount, envelope, transpose, filter drive, octave shift, pitch mod, envelope 2 pitch and volume pots up or back to their respective 0 point values. This is lame so I have done this for you:
Start.fxp (EDIT 16 10 07: Link down - reupping)
When you go off and start programming ImpOSCar on your own, you should start with this new patch and consider this your 'blank slate'. The possibilities hereon in are endless.
Pluck me baby. Oooh yeah:
We are going to start with the oscillator section because this is what generates an oscillating signal.
We have 2 oscillators and a variety of waveshapes. The oscillators on ImpOSCar are powerful as - especially the Triangle, Square, Skinny Pulse, Pulse Width Modulated Square and 'Strong Lead' waveshapes.
The saw doesn't seem that powerful when you hear it buzzing and otherwise naked, but first impressions are deceptive and saw leads on this puppy are capable of tearing you a new one.
Anywho. I digress.
Step 1:
The default waveform is Triangle. Which is good for things like bass drums and sub basses but it contains only weak odd harmonics so its not really that useful until you start doing something like tuning multiple triangle waves against each other.
Instead switch OSC 1 to Saw. You will hear a buzzing sound. This is what we want - it contains all harmonics in relation to the fundamental frequency, both odd and even, descending linearly down the scale in intensity. Since we are building a harmonic instrument, it makes sense to use a carrier wave with lots of harmonic content.
Go and plink a few keys and get used to what it sounds like.
Step 2:
Yeah well, it sounds . But thats ok because we want to build a more interesting timbre and it involves 2 oscillators. Go to the MIX section and turn OSC BAL up until you have a 50/50 mix between both OSC 1 and OSC 2.
What you hear is 50% saw wave (OSC 1) and 50% triangle wave (OSC 2).
By shear coincidence I just noted that this sounds even worse than before.
Step 3:
Lets try changing the mix between oscillators. I want more Saw and less Triangle. Mixing oscillators is a good way of bringing emphasis on certain harmonics. Mixing in more square for example will exaggerate odd harmonic content.
What you have here is an 88/12 mix of a Saw and Triangle wave. This still sounds crap but you can make Megaman tunes with this puppy. Just not yet.
Serious note: Most of this sound is Saw but the extra bit of Triangle, in zero phase has strengthened the fundamental frequency of the tone we are building as well as the odd harmonics at lower frequencies. This has the effect of making it sound a bit bassier - some people would say 'warmer' but I didn't have to take my clothes off and fan myself during the making of this tutorial so I just don't know. At this point the only adjective that springs to my mind with respect to this sound is 'rubbish.' Sorry folks.
Step 4:
You can also detune the second oscillator against the first in order to change the timbre of the sound slightly. What this will do is change the period of the second oscillator - which in turn will change its frequency and therefore pitch. If you do this, you will notice a flanging/phasing kind of effect as parts of the sound of both oscillators disappear into anti phase and get destroyed. The extent of the detune determines the speed at which the oscillation takes place.
So thus far we have 3 variables with which to build a timbre - choice of oscillators, the mix between both oscillators and the relative tuning between oscillators. There is a fourth - Pulse Width (PW) of one or both oscillators but we aren't going there with this sound so ignore it. Lets get cracking...
I plinked some notes on my keyboard and waggled the DETUNE pot around and finally settled on 11% before my ears started to ache with how annoying this sound is. It sounds like an 8 bit saxaphone. That is to say, terrible. Perhaps I should quit now and save everyone the trouble of walking you through to the end of the proverbial tunnel, only to find at the end that someone has bricked up the exit with a wall of ...
Step 5:
We shall soldier on and start changing envelope shapes. Rome wasn't built in a day eh?...I feel a pluck twitching in my loins...
Go to the Amp Decay pot and lower its value so that it is around 1.2 seconds. I set it to 1.2075 seconds by pure accident. Well it sounds shorter now but its still a fuzzy plink. I then upped the Amp Release to about 0.05 seconds. Mousing over the pot shows that in fact I had raised it to 0.0328 seconds. This my friends, is accuracy. Shame the pots are tiny and so fiddly that it makes feck all difference have values that extend to 4 decimal places.
Notice that we have a slight tail off at the end. Its very, very short but noticeable nevertheless. Don't bother with Sustain. We are making plucks now and theres no place in plucktown for Sustains. You need to go to Padland to see more where they came from.
Leave the Amp Attack on 0. We want a very sharp, transient hit and ImpOScar clicks when the Amp Attack envelope is set low or on 0. Repeatedly strike a very short stacatto note and you will hear the click. This is one of the things that makes ImpOSCar better than the average Virtual Analogue bear, though you may not realise it yet.
Step 6:
Now we want to fiddle with the filter. The filter and the envelope are the 2 primary sound shaping tools on a synth. The oscillator section is where we generate timbres. Well we have that already, and a very sketchy staccato pluck sound - its time to make it more pluckish.
Increase the Filter Resonance. I set it to 0.31 Q. What resonance does is increase the amplitude (volume) of the sound near the Filter cutoff point. We haven't touched the Filter Cutoff pot and by default I have it set to 16,000hz (very high, on the verge of inaudible high frequency). This has had the effect of giving the sound more high frequency presence. Some would say it makes it 'brighter' but thats nonsense because I never needed to wear sunglasses even once when writing this.
Step 7:
We are approaching more of a plucking kind of sound now. I lowered the cutoff to 3327hz. Note that this filter (a multimode filter) is by default a 4 Pole Low Pass filter. A low pass filter allows all low frequency sound through but attenuates high frequency sound. Hence the term 'Low Pass.' The extent of the high frequency sound that is attenuated depends on the value of the Filter Cutoff. On a low pass filter, the lower you set the cutoff, the less high frequency sound is allowed through.
If you were to change the Filter using the Filter Type pot to a High Pass filter then the opposite would be true. Because a High Pass filter allows high frequency sound to pass through and attenuates low frequency sound. Therefore, the lower you set the Cutoff on a High Pass filter, the more low frequency sound you allow through.
For this I decided to drop the Cutoff enough to take out the extreme high frequency (which in a mix would share space with predominately high frequency instruments like hihats). It still sounds sharp and staccato. If you were to drop the Cutoff even further down you will notice that the resonant peak around the cutoff follows it. It will get bassier and sound like it is muffled or underwater. Eventually it will cut out all audible sound.
This is another feature of the synth which quite simply makes it the daddy. The filter sounds gorgeous and with the filter separation on full, the resonance can increase the amplitude around the cutoff to ear damaging levels. Its brilliant. Ill do a walkthrough for building screaming acid lines on this monster at some point but I'm too lazy right now.
Step 8:
There is an envelope for the filter too. This basically does exactly the same thing that an Amp envelope does with Volume over time except it controls Filter Cutoff over time. i.e. If you were to increase the Filter attack envelope, you would increase the time it takes for the filter cutoff to open/close up to its maximum (16,000hz) or minimum (16hz) point. Whether the filter opens or closes depends on the ENV -> Filter AMOUNT. A positive value causes the filter to open over time. A negative value causes the filter to close over time during the attack phase of the sound. The maximum/minimum value in question is determined by the extent to which you turn the ENV > FILTER pot.
A good pluck should have a convex filter attack phase - i.e. if you were to graph Frequency over time for the attack phase of this sound, we want it to be fast, inverse exponential. Why? Because we want the sound to suddenly emerge out of nowhere and...uhhh. Pluck!
However, for this sound, I am going to do the opposite and go for a concave attack envelope shape then increase the filter attack time alot. I cannot describe in words what kind of effect this achieves but its pluckish in an odd sort of way because the filter will snap back open when the attack phase ends.
We can do this by increasing the Filter Attack pot and lowering the ENV > FILTER Amount pot. This will cause the Low Pass filter to close over the time determined by the Filter Attack before snapping open again at the beginning of the decay stage.
I set the Filter Attack to 0.2993 seconds and dropped the ENV > FILTER AMOUNT to -21%.
At this point I noticed that I lost the Amp Release phase almost completely when I let go off a key. This is because the Filter Release is set to 0 to the filter closes dead shut when you let go. To stop this, increase the filter attack to let the sound ring out ever so slightly afterwards. I set this to 22.1520 seconds. The reason this can be so long is that the Amp Release is very short so the sound will die out very quickly once a key is released.
Its time to fiddle with the Keyfollow variable on the Filter. You can use this to make a sound more or less present in certain octave ranges but its very action on ImpOSCar is not unlike a second filter. I increased KEYB TRACK to 1.5x as it filtered out the low end buzz, but keeps the top end.
Step 9:
I'm going to screw around with the LFO now because, despite how Pluckish our sound has become, its pretty boring. I set the LFO Waveform to 'Guitar' and dropped the LFO > FILTER AMOUNT to -24.3%.
You won't really get what the LFO is doing until you play a cyclical pattern like this (EDIT: reupping) and let it loop around. So go and get a MIDI clip of Binary Finary - 1998 or make one yourself.
Do you hear what thats done to the sound? It is making the filter open and close along the selected waveform over time. Its cutoff is in fact oscillating along this magical 'Guitar' waveform. It can be set to oscillate along a number of wave forms - like a triangle wave for instance. Which means that the filter starts at a mid point, opens up linearly to a peak then closes linearly to a trough then increases linearly to a mid point. Then the cycle repeats.
The mid point is where the Filter Cutoff is.
The AMOUNT determines what point it opens up to and at what point it closes completely. Keep thinking of this in terms of the triangle wave example because its the easiest to understand.
A positive LFO > FILTER AMOUNT means the filter will begin to open up first. Then close.
A negative LFO > FILTER AMOUNT means the filter will begin to close first. Then open up.
The amount either side of 0 determines how much the filter opens up and closes. So setting it at -100% and the Waveform on Triangle means:
The filter will close all the way down to its minimum value (16hz) in a linear fashion. When it hits this minimum value it will rise up again, linearly to its maximum value (16,000hz). So setting it at 100% either way will basically make the filter open and close all the way. a sweep spanning the entire range of the filter (16,000hz) around the Filter Cutoff point.
Setting 50% either way will only make the filter open/close across half its range - a sweep spanning 8,000hz around the Filter Cutoff point.
You can increase the speed of this filter oscillation using the LFO RATE variable. Be careful with this as it is very sensitive on this synth. I didn't want it to oscillate freely so I changed the SYNC pot to MIDI so that it synchronises the LFO to my MIDI tempo - which handily is 140 bpm in floops - the tarnce tempo.
I then left the LFO RATE at 1 cycle per bar. Increase this to see what effects you can create. Whilst you are at it, change the LFO > FILTER AMOUNT too for some wild effects but remember to return them to the tutorial values I left them at or your sound will become totally gimped for the purposes of this walkthrough.
Important Note: Hopefully you can see that the action of the LFO is connected to the Filter Cutoff, which in turn determines where the Resonance pot has its effect. The entire Filter is linked to its own Filter envelope too so if you adjust one of these variable where it is being modulated by or is modulating another variable, it will have a knock on effect.
You have to be careful when programming a synth like ImpOSCar because once you have set up the LFO around a specific cutoff value, changing the cutoff will change the sweep range of the LFO. This is one reason why ImpOSCar is quite difficult to program at first and it is easy to turn any sound into trash if you just wing it.
Therefore, everytime you make a modification to the Filter Envelope/Filter/LFO section, I recommend returning the FILTER > LFO and ENV > FILTER pots to their zero point values. Then reset the envelopes. Once you get used to how each of these modules behaves you can start to wing it and do it all on the fly.
Get used to this kind of balancing act. It helps to know where everything is and what each variable does so you can easily identify what aspect of a sound starts to sound off when you change a variable.
Step 10:
Our pluck sound is beginning to sound alot like it came straight from 1998 so lets make it more so. Call Sony EMI, I think this one is a chart topper...
I thought the default portmento sounded ed up so I changed it to Auto Fixed (A) and set the GLIDE TIME to 0.05 seconds.
Step 11:
Its time for the effects. Normally I leave effects till last. I would usually sit this sound in a mix right now and work out where its clashing with other instruments and in what frequency range in what phase. Doing so is much easier before you add effects. I also tend to chop and change effects quite alot which will change where a sound is in a mix and you can totally gimp it if you aren't careful.
But since I'm not building an entire song, I'm going to skip the entire mixdown phase and just go straight for the effects. If you wish to experiment with this sound in a mix, then I suggest disabling the effects when doing so, then switching them back on when everything is sitting properly.
You can use Chorus to simulate the effect of a choir whereby many voices are sung in unison but each voice is slightly different, not exactly in tune and not exactly in time. What it does on this synth is to play back several copies of the sound, all slightly out of tune and triggering at slightly different times.
Some would say that this makes the sound 'thicker' which is twaddle since stupidity is a human concept and applies (most often) to human beings, not sound generating machines like this one.
I set the Chorus Depth to 1.10, the Chorus Rate to 1.49hz anf the Amount to 48% but experiment with these values. I just liked them with this type of sound.
Then I stuck a Delay on it. I cant be arsed to type variables anymore so heres a screenshot of everything:
Step 12:
aka the finishing touches:
I applied a little bit of widening to this channel in Fruity's mixer using VoxengoAudioDelay - its freely available off Voxengo's website and its decent.
I increased the Vel > Drive to -70%
Then I made a few changes to get it to sound more lively:
Increased Filter Drive to 92%
Increased Chorus Depth to 1.14
Increased Chorus Amount to 69%
Increased NOISE BAL to 99/1
Lowered Filter Cutoff to 383hz
Increased LFO > FILTER AMOUNT to +64.2%
Lowered KEYB TRACK to -1.0x
Here is the finished fxp:
(EDIT: Link dead. Sigh. I'm on it)
I'll detail my effects loadout later, and most of it is freeware so you should be able to follow it pretty closely. It is however, monstrously heavy on CPU cycles (because I use SIR). So be prepared to see a single sound cripple an A64 3200+.
Because I'm too lazy to write anything now heres the Screenshot for the 1998 'verse':
And heres the fxp:
(EDIT: Link dead. I know, I know) |
|
|
| Derivative |
This may be of interest to some people on the forum but impOSCar is also very very good at making those freaky Infected Mushroom squelchy noise effects provided you are good at stutter edits. If you are lazy you could probably use dBlue Glitch but if you get a good analogue Square/Skinny Pulse lead going with loads of resonance, you can often turn it into something straight out of IM the Supervisor.
The LFO is aggressive as hell and the most important aspect of that type of sound is the Filter Cutoff modulation by the envelope and LFO. |
|
|
| TaylorR |
OMG! you actually did it! :eyes: . thanks man. i'll read this when i have time. but now, it is off to school.
imposcar ftw.:D |
|
|
| KilldaDJ |
awesome tutorial dude, well written! :D
cool sounds too, but they lack the power of a V-station to my ears :gsmile: |
|
|
| ASFSE |
omfg this is ing great...havnt read it, but i plan on it because i just started using imposcar and i'm totally lost with it (even read the manual a bit)...
granted i dont really make trance, but this should help me understand how the thing works!! big ups!!:D :D :D |
|
|
| david.michael |
| What an incredible tutorial...thanks for this. |
|
|
| ASFSE |
ok just one stupid question...how do i load sounds into the imposcar, aka the default settings(zeroed settings)?? or your fxp of binary?? silly i know, but i've pressed all the buttons with no luck!:p
EDIT: ok i figured out i could just drop the patch into imposcar from the preset dir, but is thier an easier way to do this? |
|
|
| dj_alfi |
cheers man.. on behalf of all the noobs i thank u :P
i think the cygnus supersaw is lacking some punch tho.. more attack or perhaps some compression or something.. |
|
|
| DigiNut |
Added to the master list.
P.S. "pluck twitching in my loins"? :haha: |
|
|
| mysticalninja |
| weak samples. go back to your virus lol. jk i know ur just trying to help ppl. just needs eq. |
|
|
| Derivative |
| quote: | Originally posted by ASFSE
ok just one stupid question...how do i load sounds into the imposcar, aka the default settings(zeroed settings)?? or your fxp of binary?? silly i know, but i've pressed all the buttons with no luck!:p
EDIT: ok i figured out i could just drop the patch into imposcar from the preset dir, but is thier an easier way to do this? |
Yeah that was my bad. I made an fxp called Start.fxp which you can load into ImpOSCar and its a blank patch. But I forget to bring it into work yesterday on my memory stick (I don't have an internet connection at home). I've got it with me today so I'll upload that.
You just import a Cubase preset fxp. you don't need Cubase to do this (I do it in FL Studio).
-----------------------------
I have not used any compression or EQ for any of these sounds - only freeware delay and chorus where applicable. So treat the sound as raw material straight from the synth - It would be like loading up an impOSCar preset and playing the preset straight. However, this time you have built the sound yourself and you can tailor make every single part of the sound to fit in your mix..
You still need to 'produce' it and make it sit tight in a mix, but what you hear is just the synth - 2 raw oscillators, filter, envelope and LFO and the built in Chorus/Delay. Thats it. No unison.
If you want to use several instances of ImpOSCar you could probably make a bigger supersawish sound but the focus was really to get a good raw sound from a single patch that you can work with.
Also - I should probably have put a bit more effort into the samples. They are fairly quiet (-12 dB) but until I get my Dynaudio BM5as (3rd time lucky? :( ), I'm stuck with the free headphones I got with the Guild Wars special edition. I just gained up until it didn't sound all ed up on the phones.
Once I get the Sawish patches written up you can have a go and producing them up a bit and see what you can make of it. I happen to think its remarkable what this synth can do with just 2 oscillators.
Use an LFO to modulate the cutoff of a 4 pole low pass filter along a slow speed sine wave and you can almost perfectly replicate the lead for 1998. I did a little sound test yesterday.
I'll try to add a touch of widening and start panning notes around to give off a greater sense of space and direction (as the original lead seems to drift left and right periodically).
The only thing you won't be able to replicate is heavy filter saturation. The original lead crackles really nicely when the filter opens up but I can't get any digital synth to emulate that kind of effect. So thats more than likely the dead giveaway. |
|
|
| Derivative |
I encourage others to use this thread to ask questions about impOSCar or post their own patches, or beef some of these patches up thus turning them into absolute killer sounds. Just make sure to post what you did to the sound to get it to sound the way you do - as a learning exercise and for the benefit of others. Theres plenty we can all learn from each other.
If I use effects at all I usually make sure they are freeware so anyone can get them and use them. Whereas if you turned 1998verse.fxp into a monster sound but you used like Voxengo Analogue Flux effects, Cambridge EQ on your fecking UAD-1 card and an Avalon compressor then bollocks to that - I'm not spending 3 grand so I can copy it. Most people wouldn't and thats a great shame. :\
As far as Trance goes, I'm still living in 1996 to 1999 so alot of the sounds I tend to like building are modelled on really old songs - like 1998. So they sound...well...really old.
Cygnus Saw patch is coming soon. EQ it to your heart's delight. I avoided doing so because EQ is a dangerous tool to mess with on Guild Wars special edition headphones. Same with compression. Its just too easy to it up and make it sound worse. |
|
|
|
|
