Dawesome MYTH

[Borbolactic]

Someone here says HALion7. But, 'sample-based'? What does that mean? Doesn't look like it does additive/per-partial though. Pigments? Spectral engine any good? Does it do resynthesis?

FWIW HALion7 has what they call a "spectral oscillator". You load a sample into them and then it uses a resynthesis algorithm on it to transform it at the oscillator level that you then run through the synth engine

You get a lot of control over how that algorithm works which means the raw sound of the oscillator changes before you run it through the actual synthesis engine

https://youtu.be/qEpA0GpBVk8?si=4yZK-kIo6rK4OSJV

Thanks but probably not. Reminds me of Iris 2. If I don't pursue an AI-related option being considered, where I can create my own voice models, rather than them being hard-coded, I might be leaning toward UVI Falcon... but $350? Yikes.

[databroth]

"MSEG (Multi-Segment Envelope Generator) granular synthesis is a technique used to create complex textures and sounds by manipulating short audio segments, or 'grains', in real-time. This process allows for the creation of unique and dynamic sounds that can be used in a variety of musical contexts.

I really don't understand what you're getting at here
are you implying anything with these quotes or trying to make a point?

what I find strange here with this quote specifically is that it starts with the words "multi segment envelope generator" then just continues on with a completely unrelated definition of granular synthesis

MSEGs and granular are two entirely different things
I'm guessing this was a typo? or are "MSEG granular synthesis" meant to be placed together like that?
if so, that's sorta like saying "ADSR FM synthesis"

I'm sorry, Idk if you're even trying to say anything here, but I feel a responsibility to respond as you've quoted me saying Myth isn't granular, which I will repeat, Myth is not granular.
I'm only trying to clear up any misconceptions

[Borbolactic]

"MSEG (Multi-Segment Envelope Generator) granular synthesis is a technique used to create complex textures and sounds by manipulating short audio segments, or 'grains', in real-time. This process allows for the creation of unique and dynamic sounds that can be used in a variety of musical contexts.

I really don't understand what you're getting at here
are you implying anything with these quotes or trying to make a point?

what I find strange here with this quote specifically is that it starts with the words "multi segment envelope generator" then just continues on with a completely unrelated definition of granular synthesis

MSEGs and granular are two entirely different things
I'm guessing this was a typo? or are "MSEG granular synthesis" meant to be placed together like that?
if so, that's sorta like saying "ADSR FM synthesis"

I'm sorry, Idk if you're even trying to say anything here, but I feel a responsibility to respond as you've quoted me saying Myth isn't granular, which I will repeat, Myth is not granular.
I'm only trying to clear up any misconceptions

"Played back more quickly, each grain overlaps with the next one, or some grains get skipped depending on how the software works. To avoid clicks and glitches, each grain is faded in and out with a volume envelope, a process known as 'smoothing'." ~ https://www.soundonsound.com/techniques ... -synthesis

----

When we have a 'static sound' (Myth's user manual quote) broken up into ~500 MSEGs, that also seems rather granular and, coincidentally, at about the same number of partials one would expect a good additive synth to have.
You said yourself to the effect that each iris functions as a volume envelope, yes? That's exactly what's in my above Sound On Sound quote.

Dawesome also seems well-known for granular.

[databroth]

"Played back more quickly, each grain overlaps with the next one, or some grains get skipped depending on how the software works. To avoid clicks and glitches, each grain is faded in and out with a volume envelope, a process known as 'smoothing'." ~ https://www.soundonsound.com/techniques ... -synthesis

----
When we have a 'static sound' (user manual quote) broken up into ~500 MSEGs, that also seems rather granular. You said yourself to the effect that each iris functions as a volume envelope, yes? That's exactly what's in my above Sound On Sound quote.

Dawesome also seems well-known for granular.

MSEGs aren't granular though, those are two different things, like adsr and FM
mseg is just a modulation curve, granular is sample playback, the iris are not playing back from a sample

Your quote is about granular, the Iris are not doing granular, yes envelopes are used in granular to fade the grains in and out, but this is not what the MSEG in the Iris is doing

you seem to be mixing up a few concepts here, it's sorta like how photos and paintings are both contained within a rectangle, but this does not make a painting a photo just because it is on a rectangle.

the volume envelope in this case is being applied to a series of bins (like partials, but spaced evenly, not harmonics mind you), this is not a sample though, and these are note small layered window envelopes like they would be in granular.

I have a better understanding of what you are getting at now, but I must say that you are mistaken in believing Myth to be granular. I also wouldn't say Dawesome is "well known" for granular, he has one synth that is granular (Novum) and one effect with granular processing (Love). Abyss, Kult, Chop Suey, and now Myth are all not granular

that is besides the one grain delay block in Myth, but that is unrelated to the Iris

does this help clear anything up and give a better understanding what is going on under the hood?

[Borbolactic]

"Played back more quickly, each grain overlaps with the next one, or some grains get skipped depending on how the software works. To avoid clicks and glitches, each grain is faded in and out with a volume envelope, a process known as 'smoothing'." ~ https://www.soundonsound.com/techniques ... -synthesis

----
When we have a 'static sound' (user manual quote) broken up into ~500 MSEGs, that also seems rather granular. You said yourself to the effect that each iris functions as a volume envelope, yes? That's exactly what's in my above Sound On Sound quote.

Dawesome also seems well-known for granular.

MSEGs aren't granular though, those are two different things, like adsr and FM

I realize that, but they are apparently used for granular.

mseg is just a modulation curve, granular is sample playback, the iris are not playing back from a sample

The irises, as per the manual, are playing back a 'static sound'. What does that mean and is that kind of like a sample? Can we not operate in 'granular' fashion on a 'static sound'?
MSEGs are multi-stage/segment envelope generators and there are 500 of them, operating on 'static sounds'...
While not necessarily a strict definition of a 'sampled sound', that doesn't necessarily have to preclude 'granular', at least in some senses, does it?

Your quote is about granular, the Iris are not doing granular, yes envelopes are used in granular to fade the grains in and out, but this is not what the MSEG in the Iris is doing

While maybe not outright samples, I'm unsure how much that matters.
Chopping up a sound into 500 pieces seems to function similarly and bound to have some granular/granular-type effects, logic and/or workflows.

From what I've heard of Myth so far, it does seem to have granular attributes, and that could be what some people might be picking up WRT what has been called its 'core' sound.
I wonder if most or all of Dawesome's related work have a similar 'core' sound, FM and whatever else thrown in, notwithstanding.

you seem to be mixing up a few concepts here, it's sorta like how photos and paintings are both contained within a rectangle, but this does not make a painting a photo just because it is on a rectangle.

Unsure about that analogy.
If we want to use it, however, I'd suggest it's more along the lines of what's the art about; why does it look the way it does; what is its canvas, mediums and/or technologies, what is its nature or message, how does it compare with the artist's previous works, what does the artist have to say about it, and so forth. That kind of thing.
Oh, and of course, what happens if that artist's particular setup is pursued by another and in what capacities, etc..

the volume envelope in this case is being applied to a series of bins (like partials, but spaced evenly, not harmonics mind you), this is not a sample though, and these are note small layered window envelopes like they would be in granular.

'Bins' only seem to help reinforce my 'granular' suspicion. I wouldn't get too hung up on it not being a sample and so not 'classically' granular. That's not really the point.

So what is it then? Where does it fall in the research? There a name for it?

I have a better understanding of what you are getting at now, but I must say that you are mistaken in believing Myth to be granular.

I must say you are mistaken if you believe that I believe Myth to be granular.

There are beliefs and then there are suspicions and questions and the like.

I also wouldn't say Dawesome is 'well known' for granular, he has one synth that is granular (Novum) and one effect with granular processing (Love). Abyss, Kult, Chop Suey, and now Myth are all not granular

that is besides the one grain delay block in Myth, but that is unrelated to the Iris

does this help clear anything up and give a better understanding what is going on under the hood?

Uh-oh, 'one grain delay block'... the plot thickens...

WRT Dawesome being 'well known' for granular, if you re-read my comment, I wrote, 'seems'.
I also seem to recall using something granular from Dawesome a long time ago that may not be in your list.

[databroth]

idk what to tell you, you seem to be operating on such a loose definition of any term it doesn't matter what we call it, it's just simply not granular, it's not chopping the sound up at all, the static sound is generated via oscillators, it's not sample playback, it's not grain playback at all.

yes, granular uses envelopes too, does this make a moog granular as well? moogs have envelopes

I don't understand how bins could possibly reinforce your interpretation that myth is granular either, would you call spectral FFT effects "granular" because they split up the frequency into multiple parallel bands? Granular in regards to synthesis isn't anything and everything that's clustered, it's specifically in relation to sample playback. This isn't really an opinion, it's just a matter of operating on some level of definition here.

The msegs aren't the bins, the mseg is a volume curve, they aren't overlapping in any sort of granular way at all. It has absolutely no relation to granular synthesis, it's closer to additive.

"Don't-You-Dare-Call-It-Anything-Resembling-Granular-synthesis?"

I mean it's not a matter of dictating your speech, I just worry that you are misunderstanding what is going on, and I honestly can't tell if you are being sincere or simply trying to argue for arguments sake. I'm not saying "don't you dare call it granular" I'm saying "calling it granular is incredibly misleading and not in any way an accurate representation of the sound engine"

"There are beliefs and then there are suspicions and/or curiosities and the like."

ok, do you "suspect" Myth to be granular? if so, am I really that mistaken?

"WRT Dawesome being 'well known' for granular, if you re-read my comment, I wrote, 'seems'.
I also seem to recall using something granular from Dawesome a long time ago that may not be in your list."

ok, you said "seems" that doesn't really change that I'm clarifying that Dawesome is known for far more, I apreciate that you are so keen on the specificity of your words, in which case, I hope you understand why it is important for me to clarify that Myth is not granular.

if you can point to me any dawesome product that I didn't list, please let me know because as far as I'm aware, I listed all of them. I understand you said you "seem" to recall.

I don't mean to harsh your vibe here, I hope I'm not coming across as rude or mean spirited, but you seem to be completely mistaken in your understanding here. I'm only trying to inform you about Myth and granular synthesis in general.

end of the day, it comes down to how much we enjoy using each synth, and I do suggest downloading the demo and trying it out

[IvyBirds]

Thanks but probably not. Reminds me of Iris 2. If I don't pursue an AI-related option being considered, where I can create my own voice models, rather than them being hard-coded, I might be leaning toward UVI Falcon... but $350? Yikes.

HALion7 is really nothing like Izotope Iris 2 which I have owned since the day it came out as I also owned Iris 1.0

With HALion7 you take a sample and you can play it back through the synth engine, or you can take a sample and turn it into a wavetable and then play that wavetable through the synth engine. Or you take that sample and run it through a granular engine and then run that through the synth engine, or you take that sample and run it it through the resynthesis engine and then run that through the synth engine. It comes with a giant library of samples, you can import your own, or use it to record your own

You can layer up to 16 of these sample based timbres into a single patch however you want, you can also use it's deep 8 operator FM engine and/or it's deep VA engine in layers as well

The resynthesis of the Spectral Layer is just one part of a much larger system, but it's very powerful and has a lot of hands on controls if you so choose

To me it seems like MYTH is a step backwards from that and is significantly harder to get the results I am looking for. If you own Kontact or Omnisphere you can also buy it at a deep discount. If you wait for a sale like I did it's $199 so it's not very much more than MYTH, and offers much more.

Of course YMMV

[foosnark]

I'm glad I read the manual because I finally really understand what Myth is doing!

Chopping up a sound into 500 pieces

That's not what is happening. The machine learning algorithm identifies four layers in the sound according to their harmonic content. Then for each of those layers, it creates a volume envelope.

For example, feed it a plucked string sound, and it might identify:

1. the sharp, noisy attack
2. the long, slowly decaying fundamental and a few additional harmonics
3. an additional string resonating sympathetically, taking a little while before it's excited and fading out faster than the main note
4. some other inharmonic partials due to string stiffness, miscellaneous resonances etc. which fade out pretty quickly

For each one, it approximates it with an additive oscillator (we can guess it's additive because the "pure" control attenuates quieter partials). That oscillator maintains the same static timbre over time, but it has a volume envelope for its overall level. That envelope happens to be an MSEG, which allows it to match the changes over time very well without using a ton of data.

Compare that to typical spectral resynthesis, where the synth analyzes, say, 512 frequency bins and stores an array of levels for each bin. That method is going to have more accuracy in the frequency domain, at the cost of a lot more raw data (and some loss of accuracy in the time domain). And it doesn't give you the opportunity that Myth does to change the balance of those different layers, because it makes no associations between different frequency bins.

In terms of sound reproduction accuracy, sometimes Myth's approximation happens to nail it just fine, and sometimes it doesn't because it would have needed a lot more of those static spectra. But again, the point of Myth doing things this way isn't accurate reproduction, it's making it easy to manipulate the results and feed all those transformers and modules.

...

None of this is granular synthesis.

Granular synthesis is literally sample playback, either from a saved sample or in real-time out of a delay buffer. It plays back overlapping "grains" with control over the fade in/out, start position, rate and direction of each grain.

Analyzing a sound for its spectral content is not granular.
Analyzing the volume of a sound and creating an MSEG from that is not granular.
Playing back small segments of a sample buffer in an overlapping way is granular.
Myth does not play back samples.

After Myth is done analyzing a sample it doesn't need it again. Save the preset, and it's very likely smaller than the sample was (unless you used a really short sample). Email the preset to someone but not the sample, and they can play it.

[Borbolactic]

idk what to tell you...

"Dawesome MYTH is a resynthesis synthesizer... Instead of a bunch of static waveforms that you then shape with filters, you can drag and drop samples into an Iris, of which there are two. MYTH then resynthesizes that audio into a complex waveform.

This process, we are told, uses machine learning to deliver this waveform. Quite what this does and how is not immediately clear, but it feels somewhat akin to what a granular synth does. ." ~ Gear News, https://www.gearnews.com/tracktion-dawe ... esis-wheel

Me: "Is myth synthesizer granular?"

MS Copilot/Bing:

"Indeed, Myth Synthesizer is closely related to granular synthesis, but it’s not exclusively a granular synth...

Myth takes inspiration from granular synthesis but expands beyond it. It’s a resynthesizer that combines granular techniques with machine learning and other innovative features...

In summary, while granular synthesis is a fundamental part of Myth, the synth goes far beyond it...

Let’s zoom in on the granular aspect of Dawesome Myth. While Myth is not exclusively a granular synth, it incorporates granular techniques in its resynthesis process. Here’s how:

Granular Resynthesis:

When you drop an audio sample into Myth’s IRIS (the visual representation of sound), it undergoes granular resynthesis.
The sample is dissected into tiny grains—short snippets of sound.
These grains are manipulated using the TRANSFORMER dials, allowing you to control parameters like grain size, pitch, and density.
Myth’s granular approach lets you create evolving textures, glitchy effects, and intricate sonic landscapes.

Grain Clouds and Textures:

Myth’s granular capabilities extend beyond traditional granular synths.
You can create lush grain clouds by layering grains, adjusting their pitch, and modulating their parameters...

Time-Stretching and Morphing:
Myth’s granular engine allows for time-stretching. Stretching or compressing grains in time alters the overall texture.
You can morph between different grains, smoothly transitioning from one sound to another.
This dynamic approach adds depth and movement to your patches.

Randomization with a Granular Twist:
Myth’s randomization system isn’t just about presets—it affects the granular aspects too.
Freeze sections of your sound, breed new presets from existing ones, and experiment with grain properties...

In summary, while Myth isn’t solely a granular synth, it embraces granular techniques alongside other innovative features. Whether you’re diving deep into grains or exploring the broader sonic palette, Myth invites you to create something truly unique!"

"What's the hook on this one? So far from what I see, it claims that you can drop an audio file in and have it become an 'iris', which is supposedly somehow different than just loading a sample in a sampler or granular synth, but I really haven't seen anything yet that explains HOW that's different, or what it's actually doing.

I am all fully for new and strange methods of synthesis, but I also want a plausible theory behind it. Tone2 did (does?) a lot of promo texts promising *the most unique form of synthesis EVAR* for their synths, but a lot of it sounded like a lot of woo, frankly. Can someone point me to a more clear description of what's supposed to be happening with Myth?" ~ platzangst, ReasonTalk

----

"The only description I could find is in the manual but I still do not understand it." ~ Popey, ReasonTalk (same thread at above link)

[Jac459]

idk what to tell you...

"Dawesome MYTH is a resynthesis synthesizer... Instead of a bunch of static waveforms that you then shape with filters, you can drag and drop samples into an Iris, of which there are two. MYTH then resynthesizes that audio into a complex waveform.

This process, we are told, uses machine learning to deliver this waveform. Quite what this does and how is not immediately clear, but it feels somewhat akin to what a granular synth does. ." ~ Gear News, https://www.gearnews.com/tracktion-dawe ... esis-wheel

Me: "Is myth synthesizer granular?"

MS Copilot/Bing:

"Indeed, Myth Synthesizer is closely related to granular synthesis, but it’s not exclusively a granular synth...

Myth takes inspiration from granular synthesis but expands beyond it. It’s a resynthesizer that combines granular techniques with machine learning and other innovative features...

In summary, while granular synthesis is a fundamental part of Myth, the synth goes far beyond it...

Let’s zoom in on the granular aspect of Dawesome Myth. While Myth is not exclusively a granular synth, it incorporates granular techniques in its resynthesis process. Here’s how:

Granular Resynthesis:

When you drop an audio sample into Myth’s IRIS (the visual representation of sound), it undergoes granular resynthesis.
The sample is dissected into tiny grains—short snippets of sound.
These grains are manipulated using the TRANSFORMER dials, allowing you to control parameters like grain size, pitch, and density.
Myth’s granular approach lets you create evolving textures, glitchy effects, and intricate sonic landscapes.

Grain Clouds and Textures:

Myth’s granular capabilities extend beyond traditional granular synths.
You can create lush grain clouds by layering grains, adjusting their pitch, and modulating their parameters...

Time-Stretching and Morphing:
Myth’s granular engine allows for time-stretching. Stretching or compressing grains in time alters the overall texture.
You can morph between different grains, smoothly transitioning from one sound to another.
This dynamic approach adds depth and movement to your patches.

Randomization with a Granular Twist:
Myth’s randomization system isn’t just about presets—it affects the granular aspects too.
Freeze sections of your sound, breed new presets from existing ones, and experiment with grain properties...

In summary, while Myth isn’t solely a granular synth, it embraces granular techniques alongside other innovative features. Whether you’re diving deep into grains or exploring the broader sonic palette, Myth invites you to create something truly unique!"

"What's the hook on this one? So far from what I see, it claims that you can drop an audio file in and have it become an 'iris', which is supposedly somehow different than just loading a sample in a sampler or granular synth, but I really haven't seen anything yet that explains HOW that's different, or what it's actually doing.

I am all fully for new and strange methods of synthesis, but I also want a plausible theory behind it. Tone2 did (does?) a lot of promo texts promising *the most unique form of synthesis EVAR* for their synths, but a lot of it sounded like a lot of woo, frankly. Can someone point me to a more clear description of what's supposed to be happening with Myth?" ~ platzangst, ReasonTalk

Resynthesis IS NOT equal to sampling my friend.

As you know, additive synthesis is adding simple sine waves in order to recreate more complex waveforms.
"Resynthesis" is about having a spectral analysis of a complex sound and breaking it down to sine waves (so it is the opposite path to additive synthesis).
Then the IRIS allows to read a sample at speed... 0 because it is breaking it down to granular items.

So now, it isn't a new process per say, Grain from Reason Studios does it as well, and Novum from the same maker also.
BUT the way it is done here is with a new workflow making it interesting.

[databroth]

"What's the hook on this one? So far from what I see, it claims that you can drop an audio file in and have it become an 'iris', which is supposedly somehow different than just loading a sample in a sampler or granular synth, but I really haven't seen anything yet that explains HOW that's different, or what it's actually doing.

I am all fully for new and strange methods of synthesis, but I also want a plausible theory behind it. Tone2 did (does?) a lot of promo texts promising *the most unique form of synthesis EVAR* for their synths, but a lot of it sounded like a lot of woo, frankly. Can someone point me to a more clear description of what's supposed to be happening with Myth?" ~ platzangst, ReasonTalk

Well, this is why I'm trying to explain it to you, and why I find it important that you understand it is not granular

I have actually explained what it does, so I'll try to simplify it
when you load the sample, Myth analyzes the sample, it does not keep the sample, it discards the audio file, when it analyzes the audio file, it is figuring out which oscillator frequencies in the bin array are best suited towards replicating each layer. The mseg, is a volume curve applied to each layer (not 500 separate envelopes, it's one continuous envelope with 500 "nodes" over time)

so the audio file is gone when it comes to playback and synthesis, Myth is using oscillators to generate the sound. (this is the bin array, nothing to do with granular)

I hope this helps (btw I'm not just speculating or guessing, I've spoken directly with Peter to learn this, what is more important is how it sounds and how much you enjoy using it)

As far as the articles you linked, these are entirely speculative, they are saying it "feels" like granular, this is an opinion, not a statement of fact. This is like saying a software synth "feels" analog, I actually personally disagree that they even "feel" granular. But what I am specifically clarifying is they are not granular.

EDIT: I'm sorry, one of those articles you linked is bing copilot? isn't that like a chat gpt thing, this(Myth) is software that has existed for only a month in the public eye, chat gpt has no place even being involved in the conversation, let alone as any remotely reliable source of reference. It's literally just making stuff up out of thin air.

I don't understand why you'd even ask chat gpt, I'm a human, who knows the answer and I'm explaining it to you. Do you think I'm just making stuff up or Lying?

[Borbolactic]

Resynthesis IS NOT equal to sampling my friend.

That's a red herring/straw man since I'm not claiming that.

[Jac459]

Resynthesis IS NOT equal to sampling my friend.

That's a red herring/straw man since I'm not claiming that.

Ok sure, I am not looking to argue. But if you understand that Myth is doing spectral and granular resynthesis, what is your point exactly ? I am lost.

[Borbolactic]

Thanks for the elaboration, guys.

[jobinho]

MSEGs aren't granular though, those are two different things, like adsr and FM

I realize that, but they are apparently used for granular.

It's like saying filters are used for granular so a completely irrelevant statement.