Tone2 Rayblaster: OUT NOW! (demo version available)

[aMUSEd]

I was wondering how long it was going to take for a developer to make a synth based on such principles. This is an area that interests me greatly; when I was in grad school not so long ago, I did an independent study on granular synthesis and psychoacoustics, so what they are implementing here has my attention.

But isn't this something that can already be done in synths that allow you to resynthesise any waveform such as Alchemy or Harmor?

[C-note]

I was wondering how long it was going to take for a developer to make a synth based on such principles. This is an area that interests me greatly; when I was in grad school not so long ago, I did an independent study on granular synthesis and psychoacoustics, so what they are implementing here has my attention.

But isn't this something that can already be done in synths that allow you to resynthesise any waveform such as Alchemy or Harmor?

Or Krishnasynth abandoned from long defunct Devine Machine

[Sendy]

I actually really liked the "vomit enducing" sounds in the video, but then I'm weird (edit: Until we got into generic dubstep territory)

This seems like a cross between granular synthesis and pulsar/pulse train synthesis. I'm liking that the "pitch noise modulation" and low cut filters from Saurus have carried over. Those two features go a LONG way.

[Ingonator]

I was wondering how long it was going to take for a developer to make a synth based on such principles. This is an area that interests me greatly; when I was in grad school not so long ago, I did an independent study on granular synthesis and psychoacoustics, so what they are implementing here has my attention.

But isn't this something that can already be done in synths that allow you to resynthesise any waveform such as Alchemy or Harmor?

No, it's not the same.

IMS works differently. An example: You do a sample of one of your favorite synth (maybe at around 30-50% of the maximum Cutoff) and from that sample you create a single cycle waveform with zero-crossings at the beginning and the end.
If you load that waveform into Rayblaster it recreates the impulse response of the filter of the original synth. The Formant knob could then be used like the Cutoff knob in the original synth but within certain limits. At very high Formant settings the filter will sound like a BP filter.
This also works for resonant waveforms. While in a normal sampler the "resonant peak" (inside a Signal Analyzer) of the sample always stays at the same position and you could even cut it off with e.g. a Lowpass filter. In Rayblaster the "Resonant peak" while turning the Formant knob moves like in the original synth.

you could then shape the waveform further by using "Osc windows" which could result in a sound that sounds like e.g. Comb filter of like frequency modulation.
Also PWM with different modes could be done. Let's say you got a impulse response based on a Sawtooth and switch to "Square<PW>Peek" PW mode you suddenly get a Square waveform with adjustable Pulsewidth.

With the "Harmonic" value you could "zoom" in a nd out of a waveform, either resulting in multiples of the original waveform (for bright sounds) or just parts of the original waveform (e.g. for Bass sounds). When you have a quite big waveform and use a low harmonic setting you could "scan" trough the whole waveform by modulation of the start phase parameter.

With ha custom made waveform you'll also receive a custom made filter!!


Those are just a few things that are possible...


Ingo

[xh3rv]

Sort of a PPG with faster internals, dynamic memory, some convolution-ish operations on wavetables?

(In the U.S. we're just about over with the longest election ever, my brain's fried on spin analysis)

[Ingonator]

Sort of a PPG with faster internals, dynamic memory, some convolution-ish operations on wavetables?

(In the U.S. we're just about over with the longest election ever, my brain's fried on spin analysis)

When using a big waveform including multiple waves you could do some kind of wavetable synthesis indeed (with the right use of the harmonic value and the start phase modulation). I am currently working on that...

In each oscillator a simple waveform morph could be done by loading different waveforms to Wave1 and Wave2 and use "WaveMix".


Ingo

[Dunc]

My post got lost in the last page so I am repeating it here:

Are you able to answer these questions Ingo


How long can a short sample be ? Are we talking in the order of a few seconds or merely a couple of thousand samples ?

Is the resynthesis capability in the RayBlaster from the get go, or is an "paid" update down the line ?

[zerocrossing]

Seems interesting on paper but that demo video didn't have a single sound I found interesting on it.

[Ingonator]

My post got lost in the last page so I am repeating it here:

Are you able to answer these questions Ingo


How long can a short sample be ? Are we talking in the order of a few seconds or merely a couple of thousand samples ?

Is the resynthesis capability in the RayBlaster from the get go, or is an "paid" update down the line ?

There is a difference between loading a waveform in the oscillators directly and using the Resynthesis feature AFAIK. A waveform loaded directly is up to 65536 samples AFAIK and a sample imported with the Resynthesis should be between 0.5 to 5 seconds AFAIK. I have worked so much with my own waveforms (also based on DNR Wave Designer) that i rarely used the Resynthesis feature so far.

A very long sample imported with the Resynthesis could lead to a very high pitch which at some point becomes unusable.

Anyway this does not work like a usual sampler as each imported waveform also creates it's own filter impulse response. With using the Formant knob you could get lots of different timbres when using a complex waveform.


Ingo

[stanlea]

Seems interesting on paper but that demo video didn't have a single sound I found interesting on it.

Same feeling here : this video doesn't reflect at all the technical explanations. Looks like an amateurish youtube demo, and the sounds aren't appealing. I hope there will be more sounds soon.

[Ingonator]

Seems interesting on paper but that demo video didn't have a single sound I found interesting on it.

Same feeling here : this video doesn't reflect at all the technical explanations. Looks like an amateurish youtube demo, and the sounds aren't appealing. I hope there will be more sounds soon.

Audio demos will be published after the deadline is over which will take some time. Currently we are not allowed to publish demos.

Anyway there is lots of informations available now and in my posts above i explained several things that could be done with Rayblaster.

The text, also in the detailed explanations will show much more than the video does.

Besides this we don't talk about "vaporware" but about an almost finished product here...


Ingo

[zvenx]

I too enjoyed the 'vomit-inducing' sounds in the video..... has some nice low end too, something I find missing in gladiator and saurus....
looking forward to this.... just when I said I wasn't going to buy any other synth unless it is made my spectrasonics.
sigh
rsp

[KBSoundSmith]

I was wondering how long it was going to take for a developer to make a synth based on such principles. This is an area that interests me greatly; when I was in grad school not so long ago, I did an independent study on granular synthesis and psychoacoustics, so what they are implementing here has my attention.

But isn't this something that can already be done in synths that allow you to resynthesise any waveform such as Alchemy or Harmor?

No, not really.

First, I apologize, I was a touch ambiguous by saying "granular synthesis." I should have said "microsounds", a larger family of sounds of which granular synthesis is actually a specific technique within (check out the book "Microsounds" by Curtis Roads, printed by MIT press). Microsounds exist in what Roads designates as the Micro Time Scale, with the lower limit being the threshold of human perception and the upper limit maxing at approximately 100 milliseconds. Within this time scale, perception is affected a little bit differently than other time scales.

Within this time scale, there is the assertion that continuous tone is an illusion, and that what we hear as continuous tone is actually discrete sonic events which fool the ear in to believing it is perceiving a continuous tone. In different bands of frequency (ie, low frequency vs high frequency), a certain length of event and a certain number of events need to occur in order to induce the illusion of continuous tone. With even these tiny sonic events, timbre is still distinguishable.

In any case, granular synthesis is a specific technique within the larger umbrella of microsounds. Which specific technique/techniques Tone2 is using I don't know (my guess would be something along the lines of pulsar synthesis, FOF, etc), but what their synth attempts to do (or does, rather) is provide the illusion of continuous tone (known as the forward masking effect).

Now, this stuff isn't new, and if you use environments like Csound, etc, you may have attempted such things yourself. But it is uncommon in commercial synthesizers, and I welcome any attempts at using such methods.

[ariston]

Well, yeah, the video is kind of... wannabe-cool. The sounds have that typical Tone2 hyper-real high-gloss sheen, which is fine, if that's your cup of tea. The bold claims are, as usual, a smidgen farcical.

I'll just wait for the demo. Not going to impulse-buy this time, though, as I was burned by Saurus.

[LeVzi]

Well i'm not entirely sure what that video was about, started off like a dolphin talking.

But believe me, Rayblaster is something different. I can't get my head around it at all, but guys like ingo and himlaya etc are working on sounds for it to show it off.

I am just amazed on how fat this thing is.