Dawesome MYTH

[cb8rwh]

Hopefully there will be a lock on the polyphony. It has been brought up a couple of times.

Yeah the FX section is monophonic.

[cb8rwh]

Peter will be live, 6PM CET, talking about Myth.

https://youtube.com/live/UtuV_07iSnM

[foosnark]

Peter will be live, 6PM CET, talking about Myth.

https://youtube.com/live/UtuV_07iSnM

He didn't get deep into technical details, but he did confirm my impression of the resynthesis, describing it as "temporal evolution detached from the timbre" and "you can use that temporal evolution and imprint it on something different." And indeed, the example patches he created took something that sounded like one thing (a vocoder saying "Plugin Boutique" or a drum loop), and used it to create something else altogether.

"This is not a sample player" he said several times.

[Ou_Tis]

Hopefully there will be a lock on the polyphony. It has been brought up a couple of times.

Yeah the FX section is monophonic.

In the livestream video he said there will be a lock on the polyphony "in an update which is going to come very soon".

[Sound Author]

I don't know if this makes things better or worse but I tried to be helpful. I ramble a bit, but these are my personal thoughts and feelings about Myth as a whole, and some programming habits I picked up along the way that you might find useful.

https://youtu.be/eLCMKJB694k?si=IP-2cU86LCLAK6-t

[briefcasemanx]

With the Pure Transformer can you tilt or bias it's detection to include more or less high frequencies? Watching the Sound Author video above it seems like it would be awesome (dawesome, even?) to change the balance of frequencies attenuated, maybe even modulating this balance over time.

[kcrosley]

No, but it would seem to interact with the “spectral balance” control just to the left of it (the half-moon thing). So modulations of both of those might essentially be the same thing. FWIW, I think Sound Author is sort of splitting hairs here WRT the “pure” transformer. It removes higher harmonics and, indeed, in a crappy and crackly Iris, can remove the snap-crackle-pop at the expense of also losing desirable high frequencies that cannot be replaced.

[briefcasemanx]

I dunno, I think (if he is correct and I'm understanding him correctly) it's good to split hairs here otherwise it could lead to confusion. Most sounds it should roughly mimic a low pass filter because of how harmonics generally get lower in amplitude, but when you can load whatever sample you want and it can be spectrally balanced changed before going into the pure filter, anyone thinking it's a simple low pass will get weird non-intuitive behavior in some minority of sounds.

[kraster]

I dunno, I think (if he is correct and I'm understanding him correctly) it's good to split hairs here otherwise it could lead to confusion. Most sounds it should roughly mimic a low pass filter because of how harmonics generally get lower in amplitude, but when you can load whatever sample you want and it can be spectrally balanced changed before going into the pure filter, anyone thinking it's a simple low pass will get weird non-intuitive behavior in some minority of sounds.

It seems to be similar to the temporal decomposition in Novum. The half-moon dial seems to switch between analysis of spectral content of the sample.

In Novum you have discrete temporal splits that you can select whereas in Myth it's continuous.

Reminds me a little of the amplitude part of factor synth.

I think that the "resynthesis" has more of an effect on the texture (temporal aspect) than the tone, hence all the modifiers that bring out the tonal/timbral aspect.

I find that Myth works really well with random noisy sources in the Irises.

[Ou_Tis]

I dunno, I think (if he is correct and I'm understanding him correctly) it's good to split hairs here otherwise it could lead to confusion. Most sounds it should roughly mimic a low pass filter because of how harmonics generally get lower in amplitude, but when you can load whatever sample you want and it can be spectrally balanced changed before going into the pure filter, anyone thinking it's a simple low pass will get weird non-intuitive behavior in some minority of sounds.

It seems to be similar to the temporal decomposition in Novum. The half-moon dial seems to switch between analysis of spectral content of the sample.

In Novum you have discrete temporal splits that you can select whereas in Myth it's continuous.

Reminds me a little of the amplitude part of factor synth.

I think that the "resynthesis" has more of an effect on the texture (temporal aspect) than the tone, hence all the modifiers that bring out the tonal/timbral aspect.

I find that Myth works really well with random noisy sources in the Irises.

If you're referring to the different layers in Novum---that's timbral decomposition, not temporal decomposition, isn't it? Of course grain position is temporal.

[WasteLand]

https://youtu.be/p4AmnmtLoqo?si=3yZsy2a8775ltdYA

[foosnark]

If you load various things into the Iris, latch a MIDI note, and compare the simple lowpass filter (without resonance) to the Pure control in a spectrum analyzer, you can see what's going on.

The filter does exactly like you'd expect, attenuating higher frequencies first and smoothly progressing to lower ones.

The Pure control attenuates clusters of partials. Each (roughly) 15% of the knob affects a single cluster of partials at a time; when that cluster is at zero, it starts attenuating the next group.

From what I can tell by trying it on different samples, it seems to be grouping these partials based on their amplitude, not their frequency. With many sounds there is a GENERAL tendency for the quieter partials to be higher frequency and the louder ones to be lower. But that is less true with noisier and/or more inharmonic sounds. You'll see it pull out weaker subharmonics and interspersed partials as it goes... but always in groups.

[jobinho]

I've only had a short time playing with this but it's the first Dawesome synth I've really clicked with.

The two osc's are somewhat unwieldy/unusual which I like, and I'm still navigating them but really enjoying the dissonant sounds they produce. I'm not too bothered about the re-synthesis engine but will try that at some point -- its great to have the multistage envelope / LFO.

[kraster]

I dunno, I think (if he is correct and I'm understanding him correctly) it's good to split hairs here otherwise it could lead to confusion. Most sounds it should roughly mimic a low pass filter because of how harmonics generally get lower in amplitude, but when you can load whatever sample you want and it can be spectrally balanced changed before going into the pure filter, anyone thinking it's a simple low pass will get weird non-intuitive behavior in some minority of sounds.

It seems to be similar to the temporal decomposition in Novum. The half-moon dial seems to switch between analysis of spectral content of the sample.

In Novum you have discrete temporal splits that you can select whereas in Myth it's continuous.

Reminds me a little of the amplitude part of factor synth.

I think that the "resynthesis" has more of an effect on the texture (temporal aspect) than the tone, hence all the modifiers that bring out the tonal/timbral aspect.

I find that Myth works really well with random noisy sources in the Irises.

If you're referring to the different layers in Novum---that's timbral decomposition, not temporal decomposition, isn't it? Of course grain position is temporal.

No it's both in Novum.

It splits files into Timbral and temporal.

"At the heart of Novum lies the possibility to "decompose" a single sample into 6 independent layers. This gives you disentangled access to the timbral characteristic on the one hand and the temporal dynamics on the other hand."

Underneath each layer are the envelopes for each split.

One of the cool things about Novum is the ability to apply the temporal information of one sample to the timbral information of another.

[kraster]

Peter will be live, 6PM CET, talking about Myth.

https://youtube.com/live/UtuV_07iSnM

He didn't get deep into technical details, but he did confirm my impression of the resynthesis, describing it as "temporal evolution detached from the timbre" and "you can use that temporal evolution and imprint it on something different." And indeed, the example patches he created took something that sounded like one thing (a vocoder saying "Plugin Boutique" or a drum loop), and used it to create something else altogether.

"This is not a sample player" he said several times.

I've had a good play around with it and to my mind it's more that the "resynthesis" for the Irises in this case is more about extracting the amplitude variation over time. ie. Temporal elements.

In contrast to other resynthesis there doesn't seem to be much spectral/timbral content extracted from the sample.

This seems to be then overlaid with a basic oscillator that you can FM, push it more towards Saws, square.

It reminds me a bit of Anemond's factor synth which factorizes a signal into spectral and temporal components but Myth is infinitely more flexible.

It's also, as mentioned above, similar to what we saw in Novum.

I think the real power as someone mentioned above is the use of the resonators after the irises. So you can have very complex exciters in the shape of temporal data extracted from the sample running into modal and comb filters and resonators

You also have a pretty extensive set of traditional synthesis tools like regular oscillators and filters and all kinds of additional crazy stuff and some absolutely stellar FX to seal the deal.

I think this is why Myth is tricky to categorise. There's resynthesis, physical modelling, FM, subtractive all there.

It's not like anything else. There are similarities in some areas to things like Factorsynth or touches of Madrona's Kaivo but it's very much its own thing.

It's an absolutely wonderful synth.