Hmm… I do own a Hydrasynth, and cannot rule out this approach in their hardware (it’s definitely a valid way to get a wavetable sound), but doesn’t the Hydrasynth allow you to build the wavetable through a variable 8 waveform wavetable editor? If those interpolations are prebaked then they’d have to have one for every possible combination, as the order will matter if no interpolation is happening by the machine.
e.g. Sine Triangle Square would be a separate file from Triangle Sine Square, which would be separate from Square Sine Triangle, which would be separate from Square Triangle Sine, etc. With 219 waveforms, and 8 possible slots, I think that maths out to be 4,649,238,131,000,000,000 permutations, but I could be off.
This reasoning alone makes me skeptical ASM took this approach, but I cannot say for sure (my hydra has been under utilized for ages, but it’s about to find new life driven by the Tonverk)…
I don’t recall all of it but there’s a guy that looked into the HS’ wavetables that could be extracted using the Hydrasynth updater 2.0. It was shipped by ASM accidentally along with a debug tool that allowed the Hydrasynth’s flash, including wavetables, to be dumped.
After ASM pulled the 2.0 updater, this person went on and developed their own flash extracting software but I haven’t ever given it a try myself
Reading through this while never getting the point of wavetable compared to others forms of synthesis… I still struggle I think.
To my understanding “all” you get compared to other tones is a morphing between, as if you would cross-fade between 2(or more) other synths?
You essentially “freeze” presents with completely different parameters and seamlessly morth through them, and can add modulation on top of that.
While I do see the usability in some genres, for me it would probably be busy.
Thought experiment: setting up octatrack with 2 single cycle oscillators on track 1 and 2, and decreasing vol of one while increasing vol of 2 in scene 2 would mimic the morphing shown by @sezare56 .
(For a mono approximation, no idea if a poly setup with 1-4 and 5-8 would work, depending if OT is capable to split midi note data to different channels without retrokits cables).
You could also set different parameters for the tracks, so that not only volume changes but sound as well while morphing, making it even more comparable.
I think you’re underestimating how complex and shifting natural tones can be. It doesn’t have to be a wild morph between a sheep to a tiger, it can (and generally is) more subtle, but offers more dynamism, in terms of waveshape change, than a standard analog or digital oscillator.
You also seem to be vastly underestimating how many samples tend to comprise a wavetable. It’s not just two waveshapes being morphed between, it’s generally a lot more.
It doesn’t just crossfade petween two samples. It’s sweeping up to 64 waves and the you can mix two of these sweeps. And you can control the type, direction and speed of the sweep for both, independently. All this together brings the wavetable magic.
You are already explaining more features (like tonwerks implementation), I was thinking of wavetable synthesis on principle. You can stack and add to all forms of synthesis to get complexity.
Oh I’m not trying to undersell it. I just tried to grasp what it in it’s most basic/essential form does.
(And did go on a tangent how somebody could approximate it on an octatrack)
As stated, for specific genres I could see it’s utility.
(Taken as basis for evolving pads, walls full of synth sounds in a maximalist techno set, etc). For general use cases (bases, standard leads, tracks with a lot of different elements) on the other hand I think they could crash fast with the other sounds.
Then again the tonverk was “over composed” in most demos I saw. It’s predestined to saturate the auditory spectrum with all it can do. (Not that it’s not an amazing instrument where having the capabilities is always great)
(Then again I never understood all those EDM tracks with 150+ tracks…)
Oh and I fully understand the complexity of natural tones, ‘just sample it’ was never something I understood, especially since none of the usual samplers could do round robin and other options to at least approximate variability. TV is great in that.
Yeah I was just thinking. I don’t even have an OT to try it (price…)
Yeah interpolating and mixing is not the same (but imho similar enough if aligned, would it not be? )
It’s worth keeping in mind that wavetable synth came out in 1978 (PPG Wave), and we’ve been hearing wavetable synthesis in a lot of music in the 80s and 90s. It only came late to electronic music, and I wouldn’t say it has ever been wildly adopted since, to memory, the earlier units didn’t offer filters.
From Google’s AI (as I’m not a big pop/mainstream guy)
The PPG Wave (2.2/2.3) was a defining digital wavetable synthesizer of the 1980s, famous for its crisp, metallic, and “glassy” pads, basses, and leads. Iconic uses include Depeche Mode (“See You”), Rush (“Mystic Rhythms”), a-ha (“Take on Me”), Propaganda, Tangerine Dream, and hits by David Bowie and Trevor Hor
It’s not just Tonverk’s implementation. It’s just about every implementation called wavetable synthesis. They’re all based on Wolfgang Palm’s design in the PPG Wavecomputer (and the following Wave-series).
So it is a given that any wavetable synth mixes 2 of the sweeps? Ok that’s new to me.
Is it comparable to dual oscilator Minos? Then I would still not count it (most mono subtractive synths have dual osc or a sub osc at least, but that’s more design then principle)
What I meant: if I add more oscillators to subtractive it gets more complex. If I add more grains to granular it sounds more complex. Especially if I can control the parameters of those instances independently.
Then again, aren’t all synths just trying to emulate the addition of infinite sine waves anyway
I think you’re still under a misapprehension. A dual oscillator mixes two different oscillators. A wavetable oscillator is a sample of 64~2000+ waveshapes, of wildly different natures, and allows for a smooth morph between them (which means the in-between waveshapes too).
The Tonverk only does 64, but the sky is the limit theoretically.
At the end of the day it’s still just one oscillator, one waveshape (though the Tonverk is a dual osc version of this idea).
Check this out. Its’ the best example I know of:
It should start at 15:12 – if not, jump it there.
This shows off the Hydrasynth’s wavescan feature, which is a an 8 stage version of the 64 stage offering here in Tonverk land.
I was referencing the specific comment where if was stated “It’s sweeping up to 64 waves and the you can mix two of these sweeps.” → the sweep between waves AND 2 instances of that
thx nice video.