Dawesome MYTH

[rj0]

Well there is the GRAINS module among the FX modules

Many thanks. I see it now. Under Delays. OK, missed it.

Thanks, again.

[BoogerSnot]

So, who else is left that could appreciate this?

People looking for a unique set of somewhat adventurous tools to work with, to hopefully arrive at an intriguing sound?

Synthesis isn't as hard as this. That is all. Again, give me all of the modifiers and processors, but let me play a goddamn sample without f**king it up. That's all. Or, BETTER YET, give me a wavetable. You're making my argument for me. (Salute! 🫡)

Yeah. I was absolutely agreeing with you. However... As I noted a few posts back, once I realized the Irises have next to nothing to do with sample playback and everything to do with creating an oscillator, it sort of clicked what this thing was. Page 15 of the manual gives a brief description of the resynthesis, and while it's still not especially clear what you're going to get from what you put in, at least you better understand what it is.

Bottom line, I think if you're expecting the Irises to deliver everything for your patch, I think you'll be let down. Really seems to me that this is very much a modular synth, dependent on different combinations of the available modules. That's not a statement for or against it, just an attempt to pinpoint what it is. As such, it is really opening up for me.

[foosnark]

Aside from the *core feature* of this synth, there are some positives, but they're also problematic. For example, a completely modular synth with no default mod routings is... intriguing... but then the user has to understand what the default architecture of a well-designed synth is useful for. If they don't understand that, they won't be able to create the useful mod routings. I'm not wrong on this.

I'm not sure what your point is here or why you think Myth's design (or modular in general) is "problematic." FWIW, Myth isn't fully modular, its modularity is quite restricted compared to a Eurorack or Serge-like system, or Bitwig Grid -- there is a fixed routing of "sections" and within each section, a mainly linear signal path; each module and each section has a single input and output. You can't, for instance, AM Osc1 from Osc 2; you can't just have an osc droning without MIDI input; you can't process audio with Math or slew modules, you can't feed a trigger or envelope generator directly to a Modal module, etc.

That said, to me the design limitations generally make good sense and mostly simplify the interface complexities of a fully modular system, and this *is* meant to be an instrument rather than a fully modular environment, so I find no fault with any of that, and there's some clever inclusions to make it quite flexible.

it kinda seems like the audience for this is the folks who (superheroically) did the factory patches and their immediate circle of associates. So, who else is left that could appreciate this?

I haven't even previewed more than about three of the patches, but I still feel like I'm 100% in the target audience for this synth; I find it very fun and easily capable of making sounds I'll use, granted mostly fairly busy and dirty, complex sounds. (I'm an experienced modular user who loves tinkering.)

To me the resynthesis makes more sense as I work with it, and it really does seem meant to complement/feed the transformers and additional modules. Treating it as if it's just a bad sample player or wavetable synth at its core is missing the point IMHO, and cranking the "purify" transformer up all the time is also missing the point. Embrace the dirt and weirdness, or modulate that parameter with an envelope or expression controller. I love that I can feed it a hi-hat or a recording of a crappy old toy keyboard and get some odd raspy, crawly, textured hybrid additive/FM sound that I can modify further very easily, filter it down and layer it and process that and run it through a resonator and so on. It's a killer drone generator, among other things...

Could it benefit from a simple sample player module? Maybe for some people. Sampling honestly isn't something I personally make much use of in modular, so I don't miss it at all here.

[Sampleconstruct]

At the end of the day it all boils down to sound (and GUI) aesthetics, some folks will like the krrsssfrmmppshhhhzzroarrshttzz delivered predominantly by the Iris oscillators (no matter what you feed them with), others will think -> WTF is this supposed to be, give me a sine wave instead. I for now have failed to join the MYTH cult and will move on.

[Sound Author]

Sometimes I wonder if MYTH would benefit from the removal of the resynthesis and its "iris" visualization altogether, and by default, the "untransformed" waveform is just a sinewave. The transformers are actually pretty cool, and in my less than humble opinion, if anything should be visualized, I think it should be the effect the transformers have on the waveform. The visualization of the resynthesis, in a way, makes the user feel kind of obligated to mess around with a part of the instrument that a lot of people seem to feel is their least favorite aspect of the synth.

As for using the blue transformer to remove harmonics to mitigate the harshness of the resynthesis, I do think this can be very useful, and you don't have to reduce something to a sine wave in order to do that. But I understand why that process might feel convoluted or unnecessary, which is why I've been wondering what our overall impression of this synth would have been if the main focus was on the transformers rather than the resynthesis.

[kcrosley]

Hey @SoundAuthor... just to be clear: If we transmogrify some waveform in spectral space, we are decomposing the waveform into the *SINE WAVES* that represent that waveform. Soooo... if we have some spectral representation of some waveform, we have a picture of the waveform (at some point in time) that is simply sine waves. That's just how sound works.

Now, one interesting thing that Myth might do is push a spectral representation of some waveform toward a sawtooth wave (a waveform that has ALL harmonics at a 1/n relationship -- I think this is the orange color?)... or it might push the spectral representation toward a square wave (a waveform that has only ODD harmonics again at a 1/n amplitude). It does those things. But this is just a modulatable version of things like what Korg modwave does with its built-in modifiers. It might also push the waveform toward a pure harmonic (the "blue" mod). Which again, it does, but literally what that modifer does is push the waveform toward a sine wave. Hooray, sine waves!

While this is interesting and (possibly?) useful, it doesn't make up for the fact that the core resynth algorithm is... SORRY TO SAY... basically crap. It doesn't matter if we can push the spectrum toward sines, saws, or squares, if the original spectral representation is NOT INTERESTING. (Also, if the representation ... as it is in Myth... is full of gritty discontinuities, it will be full of noise/clicks/bullshit, which is what we find in Myth.)

I cannot continue, it's just not interesting.

[El°HYM]

Some of the presets would really benefit from some sort of dynamic EQ-type effect to tame the high end when it gets excessively harsh... trying to figure out which LFO-ing effect is making a preset become too harsh / shrill at regular intervals can be tricky (and time consuming). Can put one after it of course, but it would be nice to have it right in the preset.

Sometimes I find myself thinking, "This is almost great, but I wonder if the preset designer can still hear those high frequencies that are getting too shrill..."

Do you mean something like a HF - Compressor or more like Nova's Smart Operations? I like the Idea for sure, as some Harshness can surely be introduced. Might not benefit the Cpu - usage though.

I think it would be sufficient to add a threshold control to the OTT downward compression (or at least for the high band) and have no makeup gain.

The Developer likes your proposal (add a switch where the OTT does not add makeup gain) and has put this on the list of feature proposals for a future update!



I my humble Opinion it would still be a good Idea to join the Discord.

[Sound Author]

Hey @SoundAuthor... just to be clear: If we transmogrify some waveform in spectral space, we are decomposing the waveform into the *SINE WAVES* that represent that waveform. Soooo... if we have some spectral representation of some waveform, we have a picture of the waveform (at some point in time) that is simply sine waves. That's just how sound works.

Now, one interesting thing that Myth might do is push a spectral representation of some waveform toward a sawtooth wave (a waveform that has ALL harmonics at a 1/n relationship -- I think this is the orange color?)... or it might push the spectral representation toward a square wave (a waveform that has only ODD harmonics again at a 1/n amplitude). It does those things. But this is just a modulatable version of things like what Korg modwave does with its built-in modifiers. It might also push the waveform toward a pure harmonic (the "blue" mod). Which again, it does, but literally what that modifer does is push the waveform toward a sine wave. Hooray, sine waves!

This is not my understanding of the "Pure" (blue) transformer, which removes harmonics until only the strongest harmonics remain. It does not reduce everything to a sine wave like a lot of people seem to think. Depending on the sample loaded into the resynthesis engine, "Pure" at its highest level might result in something like a sine wave, but you might also end up with something like a fundamental and some neighboring harmonics...or even a few harmonics a lot higher up the spectrum than you would expect. It depends on the sample you resynthesize. Only the strongest harmonics remain, which makes the "Pure" transformer a little more interesting than the equivalent of a spectral lowpass filter, because it means there's a middleground in between min/max values that have harmonics scooped out of the spectrum here and there rather than something like a lowpass filter sweep.

[Ou_Tis]

This synth is *REALLY* problematic. Everything on offer here has upsides and downsides. And what would seem to be the core feature (the "Iris" resynthesis) is just... I'm sorry to say it... not helpful. A multisample player and/or wavetable option as a substitute for the Iris would save the day. I get what this synth is trying to do, but it's just not doing it. Alternatively, an "FX" mode version of the plugin that lets you bypass the core sound generation/oscillator section might also save the day. But this is kind of a hot mess. It's OK not to apologize for it.

I know there are super fans here, but calling the resynthesis "fragile," "weird," or "experimental" isn't doing this synth any favors. The core feature just doesn't deliver. Saying that you should blue-mod the Iris down to the fundamental is ludicrous. (In that case, just load the sinewave Iris. But then you're just doing 2-Op FM on sinewaves... Hooray, I guess?)

Aside from the *core feature* of this synth, there are some positives, but they're also problematic. For example, a completely modular synth with no default mod routings is... intriguing... but then the user has to understand what the default architecture of a well-designed synth is useful for. If they don't understand that, they won't be able to create the useful mod routings. I'm not wrong on this.

I had really high hopes for this, but who, exactly, is this thing for? I'm an extremely advanced (in skills and in age) user of synthesizers who also has a gigantic library of samples, waveforms (which I computed myself), and wavetables (again, my own, you can have them too but I'm not posting this for SEO, so I will refrain from links) and I can't really make this work in the way the synth's own marketing and docs suggest. Sooo... what is the "average" user to do with this?

I think this might be useful as a physical modeling synth, but give me a sample player that can let me control the exciter, yeah?

Anyway, video coming soon with my thoughts on Myth, but it kinda seems like the audience for this is the folks who (superheroically) did the factory patches and their immediate circle of associates. So, who else is left that could appreciate this?

Maybe think of the iris as like noise, but with the timbral richness and meaningful time variation of a sample. The iris section is really more about the transformers. And they sound great. And I'm finding them easy to use.

But I agree that a sample/wavetable module would be great, especially for going into the modal filter and resonator or the Grains, etc.

In addition to the two FM operators in the Iris section you can add two FM modules and two Back to the FM modules to each oscillator section for a total of 2*(1+2+2) = 10---or 14 if you count the Orbit modules. No cross-modulation yet though.

[rj0]

Just throwing out a thought, something simple from a UI perspective, if not necessarily from a development perspective:

- Maybe a dial that biases the resynthesis from tonal to atonal?

[ChanceB]

This synth is *REALLY* problematic. Everything on offer here has upsides and downsides. And what would seem to be the core feature (the "Iris" resynthesis) is just... I'm sorry to say it... not helpful. A multisample player and/or wavetable option as a substitute for the Iris would save the day. I get what this synth is trying to do, but it's just not doing it. Alternatively, an "FX" mode version of the plugin that lets you bypass the core sound generation/oscillator section might also save the day. But this is kind of a hot mess. It's OK not to apologize for it.

I know there are super fans here, but calling the resynthesis "fragile," "weird," or "experimental" isn't doing this synth any favors. The core feature just doesn't deliver. Saying that you should blue-mod the Iris down to the fundamental is ludicrous. (In that case, just load the sinewave Iris. But then you're just doing 2-Op FM on sinewaves... Hooray, I guess?)

Aside from the *core feature* of this synth, there are some positives, but they're also problematic. For example, a completely modular synth with no default mod routings is... intriguing... but then the user has to understand what the default architecture of a well-designed synth is useful for. If they don't understand that, they won't be able to create the useful mod routings. I'm not wrong on this.

I had really high hopes for this, but who, exactly, is this thing for? I'm an extremely advanced (in skills and in age) user of synthesizers who also has a gigantic library of samples, waveforms (which I computed myself), and wavetables (again, my own, you can have them too but I'm not posting this for SEO, so I will refrain from links) and I can't really make this work in the way the synth's own marketing and docs suggest. Sooo... what is the "average" user to do with this?

I think this might be useful as a physical modeling synth, but give me a sample player that can let me control the exciter, yeah?

Anyway, video coming soon with my thoughts on Myth, but it kinda seems like the audience for this is the folks who (superheroically) did the factory patches and their immediate circle of associates. So, who else is left that could appreciate this?

Maybe think of the iris as like noise, but with the timbral richness and meaningful time variation of a sample. The iris section is really more about the transformers. And they sound great. And I'm finding them easy to use.

But I agree that a sample/wavetable module would be great, especially for going into the modal filter and resonator or the Grains, etc.

In addition to the two FM operators in the Iris section you can add two FM modules and two Back to the FM modules to each oscillator section for a total of 2*(1+2+2) = 10---or 14 if you count the Orbit modules. No cross-modulation yet though.

It’s hard to think of it like that for me because the irises are the focal point of the thing and they don’t really sound very good. It would all still work the same way if the resynth was more accurate. Even using the transformers it’s challenging to get a nice sound, without adding modules. I really like this synth otherwise but it’s feels broken to me,but I’m going to keep playing with it for the next few months anyway

[IvyBirds]

Experimented with "MYTH" over the weekend. It's core feature is Resynthesis, where you take a sample and the software will resynthesize it into something else and you can then take that and mangle and twist it into something else

I found the resynthesis part which is the core of the entire instrument to be very lackluster

When you compare it to the "Spectral Zone" engine inside of HALion7 which is also a Resynthesize engine the differences are night and day

Not only does the resulting Resynthesized sound, sound better but you have WAY more control over how it gets made. From there you can trust and mangle it however you wish

MYTH really needs to do a better job with what Resynthesis aspect as when your raw materials are better everything else is better

Anyway just my 2 cents

[gnu23]

At the end of the day it all boils down to sound (and GUI) aesthetics, some folks will like the krrsssfrmmppshhhhzzroarrshttzz delivered predominantly by the Iris oscillators (no matter what you feed them with), others will think -> WTF is this supposed to be, give me a sine wave instead. I for now have failed to join the MYTH cult and will move on.

krrsssfrmmppshhhhzzroarrshttzz wins the internets for today.

[elassi]

The Myth presets and sounds remind me of my failing efforts I made with additive synths.

Unless you know what you're doing you quickly end up with this a.m. krrsssfrmmppshhhhzzroarrshttzz especially when tweaking a bit more than just subtle.

[idoru97214]

MYTH and another recent product came in for a bit of a well-intentioned thumping by Skippy on his PlugInGuru YT channel this weekend. It's worth the watch - the KVR crew can draw its own conclusions.

2 New Plug-Ins but something is missing...

I feel like this guy is saying basically "Myth sucks because it doesn't try to do what every other synth does. If a synth doesn't play by boomer rules, it should be doomed to fail."