Still no decent 303 emulations for iOS...

brambos

@Matt_Fletcher_2000 said:

On the 303 sequencer the glide is different in two aspects:

it doesn’t start when the second note is hit, instead it actually finishes when that second note is hit. It can do this because it is a sequencer, not a live keyboard, hence no extra sensory perception or premonition is required by the synth to guess if you’re going to glide from one note to another, it already knows it is going to happen, it is written, in the sequence data.

The other aspect is quite strange, in that it is a constant time, not constant gradient. The time to glide is the same no matter whether it is a note and the neighbouring note, or a note and a quite widely separated note – it takes the same duration in time, therefore the slope will have to become the variable.

Thank you. That's really interesting. I wonder if @brambos could enlighten us on how it works in TroubleMaker?

The above is not almost, but not fully, correct. The 303 sequencer has no knowledge of what's coming (which is a very common misunderstanding); slides are triggered as soon as the next note gate is opened. There is nothing fancy like a lookahead going on. The 303 is made out of disappointingly simple components, no crystal balls telling the future ;-)

When a note has a slide and the next note triggers, its control voltage is sent through a first-order lowpass filter circuit causing the voltage to slide up or down exponentially in a fixed time of just under 60ms. At high tempos it may not reach its end point and slide to another voltage from whatever point it was when the gate next was opened.

This is not only the easiest way to make the electronic circuit (requiring only bare bones sequencer logic), but is also fortunate for us because it lets us mimic the exact same behavior relatively easily using MIDI.

Troublemaker (both the sequencer and the sound engine) follows the exact behavior of the 303 when it comes to ties and slides. Which is also why the first note doesn't slide back to where it came from when you release the second note before releasing the first; it requires a new gate trigger to tell the circuit where to go.

Matt_Fletcher_2000

@brambos said:

@Matt_Fletcher_2000 said:

On the 303 sequencer the glide is different in two aspects:

it doesn’t start when the second note is hit, instead it actually finishes when that second note is hit. It can do this because it is a sequencer, not a live keyboard, hence no extra sensory perception or premonition is required by the synth to guess if you’re going to glide from one note to another, it already knows it is going to happen, it is written, in the sequence data.

The other aspect is quite strange, in that it is a constant time, not constant gradient. The time to glide is the same no matter whether it is a note and the neighbouring note, or a note and a quite widely separated note – it takes the same duration in time, therefore the slope will have to become the variable.

Thank you. That's really interesting. I wonder if @brambos could enlighten us on how it works in TroubleMaker?

The above is not almost, but not fully, correct. The 303 sequencer has no knowledge of what's coming (which is a very common misunderstanding); slides are triggered as soon as the next note gate is opened. There is nothing fancy like a lookahead going on. The 303 is made out of disappointingly simple components, no crystal balls telling the future ;-)

When a note has a slide and the next note triggers, its control voltage is sent through a first-order lowpass filter circuit causing the voltage to slide up or down exponentially in a fixed time of just under 60ms. At high tempos it may not reach its end point and slide to another voltage from whatever point it was when the gate next was opened.

This is not only the easiest way to make the electronic circuit (requiring only bare bones sequencer logic), but is also fortunate for us because it lets us mimic the exact same behavior relatively easily using MIDI.

Troublemaker (both the sequencer and the sound engine) follows the exact behavior of the 303 when it comes to ties and slides. Which is also why the first note doesn't slide back to where it came from when you release the second note before releasing the first; it requires a new gate trigger to tell the circuit where to go.

Right. Thanks.

TroubleMaker certainly sounds great by the way. I've been enjoying just messing the standalone and I'm looking forward to being able to save out audio sequences to Gadget once it gets its audio tracks. The 4 bar export should make that nice and simple.

Richtowns

Just as a side note, the only ios synth Ive seen that has this level of control over the portamento is z3ta. It gives you the option of fixed/ variable time (Variable divides the time between notes) plus always on vs ligato. I know its not been supported but its still pretty neet.
I was never able to make it sound much like a 303 though. For us novice patch builders its much easier and gratifying to pick up TM to do our dirty work. My, oh my its good!

brambos

@Richtowns said:
I was never able to make it sound much like a 303 though.

Nothing about the 303's architecture is conventional, so it's almost impossible to mimic its behavior using general purpose synths. That accent alone is so complex it will drive you mad at first (not even Roland gets it right in their TB-3 and TB-03) ;-)