Official Serum 2 thread!

[Ah_Dziz]

Waveshapers shape the input waveform by multiplying the values against a single cycle of a periodic function. That function is called the "transfer function". That's the most common usage of waveshaping I've encountered in all my years and years of synthesis and sound design. It covers clipping and a huge amount of distortion types as well as other interesting things. Using a waveshaper (as defined above) is one of the most common ways of adding nonlinearities to digital audio. When people get deep into analog modelling (not my area at all) they will use a waveshaper inside a zero unit delay feedback loop in a filter or amplifier circuit model.

The techniques I spoke of above can be used to create a waveform modulation that matches the interference pattern of two of that oscillator played at any volume with any detuned amount. I'd imagine that this could be extended to almost any number of stacked copies, but the math to figure out the transfer function would get far too complicated for me quickly. The added bonus to this technique is that it maintains unity gain with the input on it's own (when I do go for unison sounds, I'm not a fan of the way that the levels stack up in many synths, especially during constructive interference).

I'm not at all saying that this is what's happening in serum 1 or 2 or any other commercial synth. I'm nowhere near patient enough to code well or to understand the code of others well, but I like making things so I tend to stick with graphical DSP stuff. I'd guess that the most recent vector command sets (avx 512 etc) would make calculating a pile of similar waveforms simultaneously extremely fast, and are probably what most of the good coders out there use.

I do know that waveshaping can create all types of cool modulation including detuning effects and things like PWM or weird digital ring mod effects. Waveshapers are one of my go-to sound design tools. Reaktor and Pure Data are great places to play with all types of waveshaping from simple Boolean tricks to interpolating through a big ol set of transfer functions while modulating the scaling of the input as well as the phase of the transfer function.

Sorry for rambling on about waveshaping. I'll stop now. I'm still digging serum 2 a ton. The last update I grabbed seems like it improved performance on my Intel stuff a wee bit, but I haven't had any real performance issues in my use cases.

[jrwaltb]

I am in this conversation because I am trying to explain that the resource issue does not exist for me and for many other people. I have no idea why, but I know Steve has personally talked about it, and I am sure it is not being overlooked, as it must be a specific chipset issue. If you have a chance to get it cheaply, I would 100% recommend it. Every time I use it, I create something really good and I am not even using the multi-sampler or spectrum features yet, as they are overwhelming. People who actually know professional sound design must be making some incredible things with this synth, which I am sure they bounce to audio and then rinse and repeat.

Are you on Ryzen?

No, I am on an M1. My point was that even if I were on Ryzen, and even if it were a massive CPU hog only on Ryzen, I still think it is worth getting (especially if you can get it cheaply), as you can do things with it that you cannot do with any other synth right now in terms of modulation and DSP. All the resampling abilities are available, so you can just make a sound and drag it to the DAW, and the DSP is all cutting edge. The S2 filters are unmatched, but I am sure others will eventually catch up.

[Igro]

...

[Echoes in the Attic]

Is there any editing of the slices? Or does it just do it automatically? Seems like the sample OSC could use a full page zoom like the wave table view. Or is there one? Haven’t seen it.

[swilow11]

Sorry for rambling on about waveshaping. I'll stop now. I'm still digging serum 2 a ton. The last update I grabbed seems like it improved performance on my Intel stuff a wee bit, but I haven't had any real performance issues in my use cases.

This was actually very interesting. Curious if you've made any downloadable Reaktor presets using waveshaping in the way you mentioned. I associate the term solely with distortion but I guess it is more broadly adding harmonics via wave shape manipulation.

A very crude form of unison can be simulated using fm from something like a noise source. At very low levels of modulation you can simulate something like the many detuned voices of a supersaw. Even a very fast sample and hold modulator applied to pitch and modulating by a very small amount can create something similar to unison.

[Ah_Dziz]

Sorry for rambling on about waveshaping. I'll stop now. I'm still digging serum 2 a ton. The last update I grabbed seems like it improved performance on my Intel stuff a wee bit, but I haven't had any real performance issues in my use cases.

This was actually very interesting. Curious if you've made any downloadable Reaktor presets using waveshaping in the way you mentioned. I associate the term solely with distortion but I guess it is more broadly adding harmonics via wave shape manipulation.

A very crude form of unison can be simulated using fm from something like a noise source. At very low levels of modulation you can simulate something like the many detuned voices of a supersaw. Even a very fast sample and hold modulator applied to pitch and modulating by a very small amount can create something similar to unison.

I've never made any downloadable reaktor ensembles. I'm pretty poorly organized and have limited design skills. I just slap together stuff that I think might sound good. Same with all the other modular DSP junk I use. I like doing little one off stuff for something very specific.

[nusound mind]

Is there any editing of the slices? Or does it just do it automatically? Seems like the sample OSC could use a full page zoom like the wave table view. Or is there one? Haven’t seen it.

Go to 1:30 in this vid, shows you some sample osc slicing options.

[nusound mind]

Another good, short video on sample osc. for creative uses. Lot of fun to be had w/ Serum 2 sampler by the looks of it. Love that it's simple drag a drop too.

[Ah_Dziz]

I am not at home or I'd check. Does the sampler import rex files?

[concealed identity]

The biggest limitation with the sampler for me was having to do the chopping and editing in that really tiny window. For me, it was frustrating enough to not bother using it with anything but really short samples, which defeats the purpose of having slicing a little bit. I think someone earlier in this thread said that Steve is considering making that window expandable in the future, though.

[Bartone]

I bought it. I have been dreaming about Serum for such a long time and now I have it

For now, I am mostly using only version 1.
It runs smoothly and has everything I need
For now, and when I have time and need there is the Serum2. Perfect!

[kraster]

The biggest limitation with the sampler for me was having to do the chopping and editing in that really tiny window. For me, it was frustrating enough to not bother using it with anything but really short samples, which defeats the purpose of having slicing a little bit. I think someone earlier in this thread said that Steve is considering making that window expandable in the future, though.

In the Mr. Bill Podcast he talked about adding a unified oscillator view where you have a single oscillator view with all the usually hidden parameters like the advanced Unison or the granular shape options on display as well as having a blown up view of the oscillator itself.

[concealed identity]

Thanks for the details. That will vastly improve the workflow and functionality in my opinion, at least how I'm picturing it.

[tumface]

I am not at home or I'd check. Does the sampler import rex files?

No. But you can export the sliced break as a SoundFont from ReCycle, convert the .sf2 to .sfz, then drag and drop that onto Serum 2, and it works perfectly. It's what I'm doing.

[hcv242]

There are tons of serum soundbanks being released all the time, but most of them (if not all) are serum 1 compatible. That means that those patches are not using the new engines nor the new filters. Why is that?
Also, by now i was expecting to see Omnsisphere like soundbanks on serum 2, since it allows multisampling. Haven't seen one of those yet. Brands have tons of produced sampled content that they could easily import to serum 2. I don't see this happening, anyone knows why?