Official Serum 2 thread!

[swilow11]

I pretty new to Serum and a bit confused as to how the wavetable engine works.

The wavetable editor seems very much like an additive oscillator editor similar to Zebra although I'm 99.99% certain the wavetable engine is not an additive engine (probably what the spectral engine does).

So what is Serum using at runtime? Does it convert a single cycle from the editor o imported PCM audio into like a math formula/equation? Or is it a different format more like vector graphics bezier curves and points?

I checked the manual and couldn't find much information about this. Hoping someone here is an expert and can clarify how it works internally!

Serum wavetables have 2048 samples for the waveforms. So they are basically arrays of short samples. When you are using the editor, you are manipulating the harmonic conent in real-time. However, the waves are not generated by partials. Vital and Peak have wavetables generated via partials.

In the context of additive synths like Razor, partials refers to discrete sine waves. Vital doesn't use that sort of oscillator.

Sorry if I remembered. I thought Vital used some kind of spectral wavetables.

It uses FFT for the spectral effects but the wavetables themselves are a different matter. The underlying concept isn't so much that the harmonics represent discrete sine wave partials, but more that sound can be mathematically decomposed into individual harmonics/sinewave-like frequency bins.

I really love some of Vitals spectral warp modes and hoped that Serum 2 would offer more stuff like that. Like the random amps warp mode. You can do the same with Serums wavetable editor by randomising bins but it's static and either on or off. Although you could theoretically create 2 frames with one full randomised and the other not and then create interpolation via crossfade or spectral morphs to cycle through it.

It would be awesome if the warp modes for the spectral oscillator could be used with wavetables. Although quite a lot of these new modes do similar sounding things.

I really love using complex wavetables in the harmonic filter of the spectral oscillator. Similar to Phase Plants Filtertable. You do need to keep your eyes on your meters and use a clipper or limiter because sometimes this filter creates massive amplitude spikes. It would be good to have built in clippers similar to what Current uses.

[kraster]

I'm mostly interested in the "clips" feature in Serum 2, offering polyphonic sequencing, like having a daw within a synth. I noticed there's a button for "gate" in "clips", but I'm not clear if it works the same way as in most arpeggiators, in which it can sharply snip off the end of notes to achieve a staccato effect. There's no knob to adjust it. Is this "gate" just for snipping off the end of a phrase rather than individual notes? Has anyone out there played with this and can comment?

What do you see as the advantage over simply using another seq/arp or your DAW unless you are selling one finger presets?

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

[pierb]

Serum wavetables have 2048 samples for the waveforms. So they are basically arrays of short samples. When you are using the editor, you are manipulating the harmonic conent in real-time. However, the waves are not generated by partials. Vital and Peak have wavetables generated via partials.

Thanks but what I mean is what is Serum using at runtime (not how the wavetables are stored).

If Serum was just reading PCM samples and interpolating those depending on the pitch then there would be no difference with a sampler (or the sampler engine in Serum 2).

[swilow11]

I'm mostly interested in the "clips" feature in Serum 2, offering polyphonic sequencing, like having a daw within a synth. I noticed there's a button for "gate" in "clips", but I'm not clear if it works the same way as in most arpeggiators, in which it can sharply snip off the end of notes to achieve a staccato effect. There's no knob to adjust it. Is this "gate" just for snipping off the end of a phrase rather than individual notes? Has anyone out there played with this and can comment?

What do you see as the advantage over simply using another seq/arp or your DAW unless you are selling one finger presets?

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Nice post. I must admit, I have largely ignored the clip launcher basically cos it felt superfluous given DAW/Bitwig but you've actually explained its potential really well.

One issue I have is that none of the shortcuts/key commands work for me. I'm in Bitwig and have to always right click to do everything. This just makes it much more of a pita than simply using the piano roll in bw.

[Uppity]

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Thanks for the great overview! Much deeper features than I imagined.

[DrGonzo]

my thoughts:
first of all, Steve should tell his instructions to the preset makers ! Seems there went not enough love into their work in that respect.
2nd, add a CPU% load reference -of some sort- to the preset names.
I think this could help the userbase alots for now.

Or you can just create a simplified CPU friendly version. It's not that difficult to make a big reduction and keeping the sound somewhat similar.

I had a similar problem a lot in the early days of Omnisphere, mixing FM and Granular really pushed the CPU consumption back then. Today everyone can run the original presets.

/C

[swilow11]

my thoughts:
first of all, Steve should tell his instructions to the preset makers ! Seems there went not enough love into their work in that respect.
2nd, add a CPU% load reference -of some sort- to the preset names.
I think this could help the userbase alots for now.

Or you can just create a simplified CPU friendly version. It's not that difficult to make a big reduction and keeping the sound somewhat similar.

I had a similar problem a lot in the early days of Omnisphere, mixing FM and Granular really pushed the CPU consumption back then. Today everyone can run the original presets.

/C

Its actually also pretty good sound design practise too. I've been guilty of making super complicated patches with macros changing multiple parameters and LFO shapes and whatever else I could modulate. Great fun at the time and got some awesome sounds, but sometimes after returning to the patch a few weeks later, I'm absolutely baffled by it.

My main hardware synth atm is an Argon 8. You get 8 freely assignable modulation slots and 4 preconfigured. This feels like a huge limitation but the restriction leads me to being more concise and focusing on making more meaningful decisions. Basically I can return to a preset and I'm not scratching my head with confusion. I actually often find I don't even use all the modulation and instead focus on playability with aftertouch and expression pedal. It's more about being musical than a programmer.

Imo, the best presets I've got for Serum tend to be simpler. I bought a sound bank from Protoculture (trance artist) and these just play so well and sound good without being swamped in indiscernible bells and whistles.

[kraster]

Serum wavetables have 2048 samples for the waveforms. So they are basically arrays of short samples. When you are using the editor, you are manipulating the harmonic conent in real-time. However, the waves are not generated by partials. Vital and Peak have wavetables generated via partials.

Thanks but what I mean is what is Serum using at runtime (not how the wavetables are stored).

If Serum was just reading PCM samples and interpolating those depending on the pitch then there would be no difference with a sampler (or the sampler engine in Serum 2).

I think it’s helpful to clarify the terminology around “samples", "frames" and "tables".

A sample can be either a short recorded snippet of audio or a single discrete amplitude value at a moment in time.

In Serum’s wavetables, each frame is 2048 discrete amplitude points that make up one cycle of a periodic waveform. So each frame is itself a table consisting of discrete points.

A phase accumulator reads through this table at a rate controlled by the desired pitch and audio sample rate, looping back to the start to create a periodic wave when it reaches the end.

Because the read position can land between discrete points, the synth has to interpolate between them to keep the waveform sounding good.

(Tthere is essentially no difference at this point between the precalculated frame and a sampled snippet of 2048 samples consisting of the same information.)

The key to wavetable synthesis is that you have multiple frames, and the synth smoothly interpolates (blends) between these frames as well, letting the sound morph over time.

So essentially, a wavetable is a table of tables—a collection of waveforms , where each waveform is itself a table of PCM samples (frames).

At runtime, Serum performs interpolation both within each frame (table) and between frames to generate evolving sounds.

The wavetable editor in Serum is just a way of visualising the 2048 samples as a kind of additive oscillator. A wavetable can be derived from anything since ultimately the only important thing is the array of values within each frame.

In serum you can make wavetables/frames from formulas, samples (snippets) and even images.

At the end of the day, all output from a synth is PCM samples since that's the only things that DACs undertand. What marks the difference is how those PCM samples are generated.

Even simple tones like sine waves are output as PCM samples, but these can be generated mathematically on the fly without needing pre-calculated tables.

A sampler, on the other hand, just plays back fixed audio snippets. A wavetable synth, blends multiple waveform tables to create its distinctive sound.

So you're not wrong to say that it's PCM samples but the implementation is what distinguishes it.

[kraster]

I'm mostly interested in the "clips" feature in Serum 2, offering polyphonic sequencing, like having a daw within a synth. I noticed there's a button for "gate" in "clips", but I'm not clear if it works the same way as in most arpeggiators, in which it can sharply snip off the end of notes to achieve a staccato effect. There's no knob to adjust it. Is this "gate" just for snipping off the end of a phrase rather than individual notes? Has anyone out there played with this and can comment?

What do you see as the advantage over simply using another seq/arp or your DAW unless you are selling one finger presets?

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Nice post. I must admit, I have largely ignored the clip launcher basically cos it felt superfluous given DAW/Bitwig but you've actually explained its potential really well.

One issue I have is that none of the shortcuts/key commands work for me. I'm in Bitwig and have to always right click to do everything. This just makes it much more of a pita than simply using the piano roll in bw.

Are you sure that the keyboard shortcuts option is enabled in Serum's Global settings?

In Live I can use the clip operations with keyboard.

[Funky40]

my thoughts:
first of all, Steve should tell his instructions to the preset makers ! Seems there went not enough love into their work in that respect.
2nd, add a CPU% load reference -of some sort- to the preset names.
I think this could help the userbase alots for now.

Or you can just create a simplified CPU friendly version. It's not that difficult to make a big reduction and keeping the sound somewhat similar.

I had a similar problem a lot in the early days of Omnisphere, mixing FM and Granular really pushed the CPU consumption back then. Today everyone can run the original presets.

/C

Its actually also pretty good sound design practise too.

I do understand these points.
But i can counter: do the work once, or have it be donne a 1000 times again and again by your customers ?

To me, there is a rule vs. SW: things need to work fast...and fluff. Or we deal with stuff just adding a new level to the game. I´ve not seen that yet with Serum2. ( but i gladly buy great presets for it once out.....and clearly labeled so i can see if it is for me or not)



In my case, "my sound outcome" is too huge degrees based on the post Instrument processing anyway. So, my time goes into that ! But that´s just me.

[kraster]

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Thanks for the great overview! Much deeper features than I imagined.

It's even more mind boggling when you consider that you can run all of that into an arpeggiator that has its own custom pattern mode!

[swilow11]

I'm mostly interested in the "clips" feature in Serum 2, offering polyphonic sequencing, like having a daw within a synth. I noticed there's a button for "gate" in "clips", but I'm not clear if it works the same way as in most arpeggiators, in which it can sharply snip off the end of notes to achieve a staccato effect. There's no knob to adjust it. Is this "gate" just for snipping off the end of a phrase rather than individual notes? Has anyone out there played with this and can comment?

What do you see as the advantage over simply using another seq/arp or your DAW unless you are selling one finger presets?

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Nice post. I must admit, I have largely ignored the clip launcher basically cos it felt superfluous given DAW/Bitwig but you've actually explained its potential really well.

One issue I have is that none of the shortcuts/key commands work for me. I'm in Bitwig and have to always right click to do everything. This just makes it much more of a pita than simply using the piano roll in bw.

Are you sure that the keyboard shortcuts option is enabled in Serum's Global settings?

In Live I can use the clip operations with keyboard.

Oh, right. I didn't even know there was such a setting... Thanks for the suggestion!

[swilow11]

my thoughts:
first of all, Steve should tell his instructions to the preset makers ! Seems there went not enough love into their work in that respect.
2nd, add a CPU% load reference -of some sort- to the preset names.
I think this could help the userbase alots for now.

Or you can just create a simplified CPU friendly version. It's not that difficult to make a big reduction and keeping the sound somewhat similar.

I had a similar problem a lot in the early days of Omnisphere, mixing FM and Granular really pushed the CPU consumption back then. Today everyone can run the original presets.

/C

Its actually also pretty good sound design practise too.

I do understand these points.
But i can counter: do the work once, or have it be donne a 1000 times again and again by your customers ?

.

Totally agree, customers shouldn't have to do this. I was more just weighing in on the educational value of optimising ones presets as opposed to justifying the insane CPU hoggery of some S2 presets.

I should just add that many synths have very hungry presets. I've found patches that are basically unplayable in synths like Phase Plant, Current, Pigments, Falcon. In fact, the only synth where I haven't encountered that (or accidentally stumbled into it with my own sound deisgn) is Hive 2. Which sounds as good as any of the aforementioned which raises a few questions imo...

I should also add that many S2 presets are totally fine. In fact, most of the few I've sampled work perfectly. We all may be getting too caught up focusing on the minority and overlooking what is more typical for this synth.

Veering to a different topic sort of. A feature request- can we copy oscillators and mod mappings between Serum 2 instances? The easy work around is save a copy of a preset with whatever features you don't want switched off but copying between instances would be cool. I like the hybridise preset feature but would also like to have more control over it by being able to directly drop a sound source into another preset. I get that this could all get confusing if you're copying oscillators that are being modulated by other oscillators but maybe there could be a workaround to that.

[pierb]

So essentially, a wavetable is a table of tables—a collection of waveforms , where each waveform is itself a table of PCM samples (frames).

So let's ignore for a second the actual frames, interpolation between frames, etc.

Say you have a single frame with a single waveform cycle and a .wav file with a single cycle load into a sampler.

Are you actually saying that given a single cycle both a sampler and wavetable engine will do exactly the same thing? Eg: interpolating PCM samples for a particular pitch/speed rate?

Can anyone who has a technical knowledge confirm this?

[swilow11]

I'm mostly interested in the "clips" feature in Serum 2, offering polyphonic sequencing, like having a daw within a synth. I noticed there's a button for "gate" in "clips", but I'm not clear if it works the same way as in most arpeggiators, in which it can sharply snip off the end of notes to achieve a staccato effect. There's no knob to adjust it. Is this "gate" just for snipping off the end of a phrase rather than individual notes? Has anyone out there played with this and can comment?

What do you see as the advantage over simply using another seq/arp or your DAW unless you are selling one finger presets?

The clip player is really powerful. It's not just a piano roll for the sake of it.

You can have the sequencer triggered by key presses and can be transposed. You can play it in mono or polyphonically and it conforms to the keyboard scale quantization setting.

The sequence can run freely so pressing a key enables it as it cycles through or you can have it reset on a key press. The start of the sequence can be quantized to a value to make sure it's locked to the clock.

It has numerous play modes like forward, backwards, pendulum and a bunch of random modes that slices up the sequence into divisions of the selected timebase

You can also have different sequences with different setting play simultaneously.

You have a mode that offsets the start of the sequence and modulate that position with a modulator.

Additionally you can sequence note expression, macros, midi ccs.

There's a static mode that allows you to save notes or chords and/or macro settings, cc values etc. that can be recalled by pressing a key

There are also probabilty settings as well as Octave probability and even velocity randomness.

You can also just use it for storing common patterns and having a preview for user presets.

You can also just drag and drop sequences from the sequencer into the daw if you like.

I don't see these things as separate since they integrate heavily with the synth. Doing all of this from a DAW would take forever.

Nice post. I must admit, I have largely ignored the clip launcher basically cos it felt superfluous given DAW/Bitwig but you've actually explained its potential really well.

One issue I have is that none of the shortcuts/key commands work for me. I'm in Bitwig and have to always right click to do everything. This just makes it much more of a pita than simply using the piano roll in bw.

Are you sure that the keyboard shortcuts option is enabled in Serum's Global settings?

In Live I can use the clip operations with keyboard.

Oh, right. I didn't even know there was such a setting... Thanks for the suggestion!

Quoting myself but I'm still not able to use shortcuts in Serum. I'm thinking it's probably a Bitwig issue more than anything else.

Any other Bitwig folks run into this?