303 simulation

[3phase]

Here an other aproach that splits the Seq and soundengine part over 2 slots



G 203 split
 
A different approach to the original G-203 patch. This time the Sequencer is in an another slot (A) than the main 303 sound engine (B). This allows patternchanges while moving a knob on the sound engine. You can select from Slot B patterns in Slot A (Knob 7+8 A1).
 
 
Switching between slots via Focus. You also can assign all 2 slots to the keyboard but than you have stuttering triggers on the G203 soundengine in slot B…what can sound nice in some cases.
 
G-203 split manual:
 
Sequencer:
The sequencer in Slot A allows up to 64 step patterns. As in a real 303 you have a gate sequencer and a pitch sequencer that get programmed independently from each other. The gate sequencer has to be written by the G2´s panel knobs while the pitch sequencer allows alternately to be written by the keyboard and has only 32 steps. Each gate command written in the gate sequencer advances the pitch sequencer one step. A gate sequence with a length of 16 steps that only contains 5 notes uses only the first 5 steps of the pitch sequencer. If The Seq Length is setted to longer than 64 steps and synctrigger off  or later than 64 steps the sequence runs into silence.
The pitch seq runs in a cycle if more than 32 pitch informations are needed.
To Access the sequencers you have to use the patchedit mode.
(Patch button on the front pannel)
 
The GateSeq.:
In the 303 you write in "note"," legato" or " rest" commands to write a gate sequence. In the G203 you use the Knob/Button Row of the event sequencers to write trigger and gate commands. The Buttons set the trigger or single note command while the knobs set a legato command.
A single note command (Button act) creates a gate length of one 32th note plus the length you set in the editor with the GteXtns parameter. On a real 303 a gate length longer than a 32th is produced what sounds smoother than a straight 32th. When you set a legato command above a trigger command the gate length is extended to a 16th not or even longer when you chain legato commands ( knob full clockwise).  You should avoid writing legato commands that are not linked to a note command for overview reasons.
 
There is an “alt mode” button in the editor that changes the behavior of the gateseq. slightly to get closer to the original 303 timing. But the alt mode is a bit irritating to write.
In the alt mode a trigger (button) that follows directly after a legato (knob fully clockwise) is not played and adds instead to the length of the legato note with one 32th, leaving a 32th gap before the next trigger. In the normal mode there is no gap between a legato note and the next trigger.
 
The PitchSeq.:
The knobs write the pitch the buttons write a slide command on this pitch step. A Slide command keeps the gate open up to the end of the following note and activates a portamento  like on a real 303.
 An additional Transpose Sequencer, Knob row  B3 –E3 allows to set per step a transpose value (knobs) and Accent command (buttons)  that communicates with the Accent level of the soundengine in slot C and shortens the filter envelope time..
 
The main pitch sequencer can be alternately written with the keyboard. Knob 6A1 activates write in and disables transpose so you direct can overwrite a pitchstep in a running sequence. As in 3 real 303 a so-called pitch mode is provided. As soon it is activated ( 5A1) the pitch seq resets to its first step. You can run thru the 32 pitch steps by pressing any key on the keyboard. Press the keys softly to avoid triggering the accent circuit. When you stop stepping thru the pitch steps and activate write the next step is overwritten. Only 32 steps get written. The pitch mode is also handy to set accents and slides.
 
You think that is a weird sequencer? Maybe, but its pretty similar than in the 303.
 
Note: This version of the G203 seq has no swing…as the real 303…
You might wonder why the Swingparameter of the clock generator is addressed anyway… This is because it effects the resetcircuit of the Seqencer… and is causing on certain settings some nice glitches.
 
Other Slot A knobs:
1A1 the length of the pattern, 2A1 an delay to make the Seqencers later. 3A1 activates a sync trigger puls that appears every number of quarter notes you set with knob 4A1.  5+6A1 manual pitchvalue recording.
4A1 Sync divider in quarter notes:
Every x quarter notes the sequencers get a reset command. This is useful for polyrythmic structures but can be necessary to get tighter sync to external equipment as well. Knob 3A1 enables this Sync impulse.
Knob 1A1 sets the length of the pattern from 1 to 128 16th notes. When you set this to 12 steps the sync divider have to be set to 3 or 6 if you want a round loop ( only if syncpuls is enabled of cause)
 
7A1 deactivates keyboard transpose and 8A1 switches the Seq of for keyboard only control
1+2A2 activate set and activate the root note for the pattern (variation).
The Key the sequence starts playing after a variation change. You have to switch variations to get the effect. Is 2A2 off the key from the previous variation is used.
8 A2 is the swingfactor. Row A3 allows Clear and random for the Pitch and transpose sequencers. Slide and Acc commands stay untouched
 
 
The Soundengine in Slot B:
In earlier versions seq and soundengine were sitting in one Slot. The Seq in the VA controlling the Soundengine in the FX area via the 4 FX busses.
I changed this to allow to change the SEQ pattern without overriding the Soundsettings. You do that with knob 7+8 A1. 7 Selects and 8 executes the change.
 
Allmost all knobs are addressed but most of them are for finetuning the 303 behaviour. Its not really ment to use other knobs than in page A and E but of cause the other knobs can give nice results… But keep in mind that they do more ballistic adjustments and are no real performance parameters.
 
These you find in Page A. 1-4 A1, Filter CF, resonance, Env modulation and Accentlevel like in a real 303. The Decay ( a parameter that don’t does too much and an additional accent decay knob on 3+4 A2.
5A1 activates the Transpose Seq and communicates with the Note Quantizer on 5+6 A2. Range and notes are booth on 12 and 5A1 is fully clockwise the transpose seq in Slot A behaves like the octave  up/down attributes in a real 303 sequence.
Other settings give more exotic Sequence variations. 6A1 is a transposefunction.
Knob 1A2 allows to use the transposeseq output to modulate the filter
8A2 Waveform,
 
1+2A3 Tuning, Knob 3+4A3 unscale:
Almost every real 303 generates higher pitches on higher notes...they are not evenly tuned as any good bass synth
Set unscale off to get in tune with standard sounds Knob 3 gives the unscale amount...you have to readjust the main tuning when doing changes here.
 
I don’t explain all the tuning knobs now…its to complex…they all change the behaviour of the patch towards accents and evelope modulation.
I addressed the ones I found most usefull so its easier for you to do own tweaks.
When I ever have a version that is optimal tunded I will remove the knobs and stay with the ones in A.
 
Have fun,
Sven R. aka 3phase
 
 
 
 
 
 
 
 
 

---

G203 Split 1.prf2
Description:		Download (listen)
Filename:	G203 Split 1.prf2
Filesize:	15.45 KB
Downloaded:	1748 Time(s)

---

[mosc]

Great way to do this - slitting the sequencer from the voice.

Great documentation. You are setting the standard here.

Looks like this attachment got duplicated. If so, would you please delete one of the copies. You have to be pretty patient when posting attachments.

[monobass]

Hey 3 Phase,

Nice patch

Did you work out any way of doing the correct length notes in the end? i.e. making them 3.5/96ths long.. The Gate Xtns thing works fine, I'm just wondering if there is a way of actually doing it accurately which reacts dynamically to tempo.

This is more of an academic question than a criticism

thanks.