So what's next for U-He?

[pekbro]

It sounds like assembly might offer better performance if done smartly. Probably not worth the hassle.
It's too bad my dad isn't still alive, he could help you out with that, never ceased to amaze me what
he could do from basically nothing. He used to bring me the punch cards and univac one "swag"
from when they had it over at Sperry. Maybe it was Franklin Life, anyway one of those places had it.

[Urs]

@Toonertik I should have put "DSP capable" in quotes

Ideally, a DSP or FPGA comes in tandem with a Cortex A/M on the same chip, keeping the UI and communication on the ARM while offloading the audio processing to the DSP.

@pekbro When programming DSP SIMD (Single Instruction Multiple Data) in C/C++ it is typically a hybrid between assembly and C. You'd use Opcode-like statements ("Intrinsics") with C-style variables. It saves the developer from declaring and bridging to sections of inline assembly and it burdens the compiler with memory store/load management/optimisation. An advantage of doing so is reduced development time, and - if done thoughtfully - portability and easy implementation of unit tests.

I'll plant a seed.. it's about counting trigs/gates and outputting a gate at user_trigNo_count.. held for user_trigCount.. with maybe a volt value..

PS.. It's the sort of thing I have been trying to do with Doepfer A-162 TDEL and Erica's T->G but it's a bit messy and no exact sync, as you can imagine but still quiet fun

Hehehe, isn't that what pretty much any gate/CV sequencer could do? My beloved Moskwa and Tirana could probably do this. But I guess you want CV control over trigCounts?

I do have a thing for these kinds of processors though, like Tempi with adjustable pulse width. The absence of trigger/gate manipulation such as clock dividers in software synthesizers is something that has started bothering me.

We shall see.

[pekbro]

@pekbro When programming DSP SIMD (Single Instruction Multiple Data) in C/C++ it is typically a hybrid between assembly and C. You'd use Opcode-like statements ("Intrinsics") with C-style variables. It saves the developer from declaring and bridging to sections of inline assembly and it burdens the compiler with memory store/load management/optimisation. An advantage of doing so is reduced development time, and - if done thoughtfully - portability and easy implementation of unit tests.

Thanks for the explanation, of course I have no idea how that works, I just thought you could
bypass the interpreter layer and do everything at low level. That's the kind of stuff my dad used
to do. When he started out, they did not have the development environments we are used to these
days. Once when I was a kid, I saw him re-write the OS of one of those Tandy computers from
radio shack for a friend. It took him about 6 months, when he was done, there was no more
Tandy software in it, it was a custom orchid farm accounting system, you didn't load the software
once you fired it up, when you turned it on it was an accounting system and that's all it
could do after that. I was pretty young, so I am not exactly sure what all he did to it.

[lunardigs]

... The absence of trigger/gate manipulation such as clock dividers in software synthesizers is something that has started bothering me.

We shall see.

Is this something where gate signals could be pulsed or syncopated?

[Urs]

... The absence of trigger/gate manipulation such as clock dividers in software synthesizers is something that has started bothering me.

We shall see.

Is this something where gate signals could be pulsed or syncopated?

In the modular world there are numerous modules which take a stream of gate or clock signals and output one or more variations of that. In the most simple case you'd use a flip-flop to divide the tempo in half, e.g. turn a 1/16th clock into an 1/8th clock. Other modules combine gates or add randomization to create complex patterns out of simple ones. In the modular world, gates and triggers are often seen as equal to audio and control signals - whereas in most soft synths there's just Note On and Off and that's pretty much it. Arpeggiators and sequencers might add patterns, but they still just commonly result in NoteOn/Offs. This makes a lot of sense since people mostly play them with keyboards or produce music in a DAW.

I'm thinking, classic ways to animate soft synth patches is by using complex envelopes which resemble automation curves. They allow for a lot of control and detail, but they are also pretty much fixed after they're set up. Such things are a rarity in modular, since there's no good way to edit complex envelopes with a few knobs. Hence, people combine sequencers, simple envelopes and all sorts of things to evolve a sound. While this doesn't allow for much detailed control, the advantage lies in its dynamic nature: A few tweaks to a set of knobs and switches creates a whole new variation of that evolving patch.

I want to bring some of that into our stuff. Hive 2 was pretty much a first step into this. Any major version release of any of our synths will certainly reflect my affinity to these kinds of concepts, offering *options* (not a necessity) to embrace logic, randomness and more interaction between modules.

(not sure why someone voted your post down...? Accident?)

[Cepheus]

I want to bring some of that into our stuff. Hive 2 was pretty much a first step into this. Any major version release of any of our synths will certainly reflect my affinity to these kinds of concepts, offering *options* (not a necessity) to embrace logic, randomness and more interaction between modules.

The function generator and shape sequencer are very nifty additions to create interesting and unexpected sounds. For sanity sake and the Wife, I stay away from modular stuff but I like the randomness and following consequences of it for the timbre and note changes. What I would appreciate is that particular randomness in subtle way, where parameters like oscillator waveforms, OSC FX (hint ..), pitch, shape and timing of envelopes and LFO’s change or vary just a little bit. The human ear appreciate little variations in sound in physical instruments. That small variation in lip tension or air pressure in a trumpet. Not saying that soft synths should be capable of emulating those instruments but these aspects are what people appreciate in “unstable” analog synths. Also interesting to see that next to Urs also other developers like Steve Duda feel the urge to play with more flexible structures in synths. The automatable points in the LFO’s in Serum are a fascinating crazy development in sound synthesis. That means that LFO’s and envelopes are not so rigid anymore.

[lunardigs]

...

(not sure why someone voted your post down...? Accident?)

Yeah, I saw that and my dopamine levels dropped
Then it went back up when I saw an up-vote
Now I'm back to normal

Seriously though, thanks for the explanation.
I can see better now the gate/trigger relationship when using the BWS Grid.

[dangayle]

I want to bring some of that into our stuff. Hive 2 was pretty much a first step into this. Any major version release of any of our synths will certainly reflect my affinity to these kinds of concepts, offering *options* (not a necessity) to embrace logic, randomness and more interaction between modules.

Arturia's Pigments has a really nice implementation of this stuff. Extremely useable. No "gates" per se, but all the rest.

If I could trigger old school trance supersaw gates in Hive, I'd be so happy. If anyone knows how to do that, I'd love.

[Urs]

If I could trigger old school trance supersaw gates in Hive, I'd be so happy. If anyone knows how to do that, I'd love.

Not the kind if gates I had in mind , but that's actually easily done in Hive, nowadays most obviously with the Shape Sequencer that came with Hive 2.0.

But it has always been quite easy: The reason we made Hive's Note Sequencer run as "Mod" without triggering notes was also so that people could build trance gates and other rhythmic effects.

I guess people are expecting dedicated trance gate effects in the fx section... but obviously, those are very limited concepts. Both aforementioned modulators do the trick as they can even modulate things in the fx section itself, so you can do post-reverb/whatsoever trance gates without the necessity of a dedicated effect.

[Captain]

Thanks to you mentioning trance gates and supersaws, I suddenly wanted to make one in Hive (using the shape seq, was super quick & easy), and now I have spent the last 30 minutes jamming with classic trance riffs, and want to make trance music... Also noticed that adding a touch of Hive's compressor gives the gate a really nice punchy attack!

[pdxindy]

The automatable points in the LFO’s in Serum are a fascinating crazy development in sound synthesis. That means that LFO’s and envelopes are not so rigid anymore.

I haven't tried that in Serum yet, so cannot comment on that specific implementation, but generally making points automatable in such envelopes/lfos/msegs is tedious to work with and sounds more useful in theory than it turns out in practice.

I still find the combination of multiple simple modules to create complex shapes more interesting for evolving the shape over time.

[blue monk]

Is the Octave Cat off the table ?

[Urs]

Is the Octave Cat off the table ?

The idea died when many people made clear that monophonic synthesizer plug-ins were not anything they'd buy into. The Cat project was shelved in favour of the Repro-5 project, i.e. a polyphonic version of Repro-1.

We'll resurrect it maybe after we clear our backlog of releases and updates.

[toonertik]

We'll resurrect it maybe after we release and distribute CVilization and then clear our backlog of S/W releases and updates.

From the man himself...

[blue monk]

Is the Octave Cat off the table ?

The idea died when many people made clear that monophonic synthesizer plug-ins were not anything they'd buy into. The Cat project was shelved in favour of the Repro-5 project, i.e. a polyphonic version of Repro-1.

We'll resurrect it maybe after we clear our backlog of releases and updates.

I must admit I was thinking polyphonic. I was originally interested in the Voyetra Eight.
That said, I would also be interested in a monophonic Cat.