Synapse Audio Minimoog emulation "The Legend" for VST/AU and RE released!

[EnGee]

So a month wasnt long enough to demo a synth???

It is more than enough. The thing is related to me personally, so I needed to demo it again. Usually I buy it in the first hours of demoing, but I had to make some choices especially I do have a limited source of income.

Anyway, I was really ready to buy The Legend, but I just wanted to be sure of two things I haven't tested yet. 1. How useful is the polyphony (because I never put attention to it, as I just demoed and compared with Monark).
2. The support. I had doubts about the support because in the case of Dune 2, 'Marcin' only sent email to the seller and never notified me of the license that was transferred into my account!

So, if Marcin answered me this time (even with "Sorry, we can't do anything about the demo"), it would be OK and I would buy it, but to ignore the email since few days? Not nice at all.

Anyway, if it is not important to answer one client, why do you care! The majority are happy and it is enough for them. For me, I still have a great synth (Dune 2) but that's it. I prefer to spend my money somewhere else

[Ingonator]

What I wanted to check is how useful would be the polyphony for such synth and how it compares to Retrologue or Diva demo or LuSH in pads. Anyway, I don't want anything now

The poly feature is a very nice addition that i do not want to miss in The Legend and the adjustable stereo spread could be quite useful too for polyphonic patches.

Of course no every monophonic patch is also suitable to be used with polyphony properly while the poly patches included in teh factory content mostly seemed to be descigned in a way to be used with polyphony.
Overall the factory content includes more than 400 presets and around 50 in the "Poly" category. AFAIk the demo incudes much less presets.
The Synth Brass patches i posted yesterday (mostly based on Memorymoog patch sheets) are intended to be used with polyphony too.

The additional CPU use for 4 voice polyphonic patches seems to be quite low. It is around the CPU use of one additional voice compared to the monophonic mode.

When all 4 voices are used voice stealing could happen also depending on the Release time and the playing style but with 3 note chords usually this should not happen even with longer release times.

[EnGee]

What I wanted to check is how useful would be the polyphony for such synth and how it compares to Retrologue or Diva demo or LuSH in pads. Anyway, I don't want anything now

The poly feature is a very nice addition that i do not want to miss in The Legend and the adjustable stereo spread could be quite useful too for polyphonic patches.

Of course no every monophonic patch is also suitable to be used with polyphony properly while the poly patches included in teh factory content mostly seemed to be descigned in a way to be used with polyphony.
Overall the factory content includes more than 400 presets and around 50 in the "Poly" category. AFAIk the demo incudes much less presets.

The additional CPU use for 4 voice polyphonic patches seems to be quite low. It is around the CPU use of one additional voice compared to the monophonic mode.

When all 4 voices are used voice stealing could happen also depending on the Release time and the playing style but with 3 note chords usually this should not happen even with longer release times.

Thanks Ingo for the info.

Yes I agree with you. It was my mistake to think about the already 'thick' monophonic sounds and didn't find the idea of polyphony exactly appealing! So, I rejected the idea of polyphony, but then (just lately really) when I thought of it, I realized that if the sounds are not very thick, I might be able to have nice pads with no more three notes chords. So, that I needed to demo it again to compare with Diva demo and my own (LuSH, Retrologue and somehow ACE among others).

[Ingonator]

Yes I agree with you. It was my mistake to think about the already 'thick' monophonic sounds and didn't find the idea of polyphony exactly appealing! So, I rejected the idea of polyphony, but then (just lately really) when I thought of it, I realized that if the sounds are not very thick, I might be able to have nice pads with no more three notes chords. So, that I needed to demo it again to compare with Diva demo and my own (LuSH, Retrologue and somehow ACE among others).

Well, given the fact that some people in other synths do use Unison or a "Multisaw" feature with up to 8 or more voices having up to 3 detuned Oscs in the poly mode of The Legend does not really seem to be too much.

In Unison mode where you play only a single not you could play up to 12 detued waveforms at once which could be really nice for certain patches while for others it is maybe too much.
As the detuning of the dedicated Oscs could sound a bit different to detune in the Unison it could also make sense to use a single Osc in combination with Unison instead of 3 detuned Oscs (and you coudl still use Osc 3 for modulation in this case). If you use 2 Oscs at different octaves you could have 4 voices playing for each of the octaves. You could also program a 3 notes chord with the 3 Oscs and add Unison on top of this. Opposing to using the Mono mode with 3 Oscs you could also add stereo spread with the Unison.

When using the poly mode in The Legend also the mono legato is off by default (there is no poly legato AFAIK) and also the volume of single notes is lower than with the dedicated mono mode (where legato is possible).

The Glide seems to work with both the monophonic and polyphonic mode.

Even if all 3 Oscs are alraedy used for otehr stuff and OSc 3 is not really suitable for modulation in a specific path (while you could still try...) you still have the option to use teh Noise source for modulation by setting the mod source knob to 100% Noise (where i prefer using the pink noise).


Last but not least also the FX version of The Legend could be quite useful and when triggered with an additional MIDI track you could use the built-in enveloeps and also the oscillators of the FX plugin (actually when triggered by the MIDi track the Fx version seems to work like a ful featured synth).

[fluffy_little_something]

When you play chords with longer release times, what happens when you play several chords in a row? Is the voice stealing - in view of the 4-voice limit there must be lots of that - smooth and quiet?

The good thing about chords on Legend is that the CPU load stays almost the same, no matter whether you play 1 or 4 notes at a time

[Gucky]

I am still hoping that the Legend gets a sequencer.
I have much fun with the sequencer of Repro-1. With that sequencer I make another kind of music as I would do only in my daw.
Here is an example:
https://soundcloud.com/musikus/bataranga
That track would never be made without the inbuild sequencer of Repro-1.
So I think that a sequencer in Legend would also be a great inspiration.
I don´t think that an external sequencer is the same. With an inbuild sequencer I can store the sequencethat belongs
to the sound and that is great for me!

[fluffy_little_something]

There are some cool external sequencer plugins. The DAW indeed feels rather different, despite being a sequencer as well.

[Ingonator]

When you play chords with longer release times, what happens when you play several chords in a row? Is the voice stealing - in view of the 4-voice limit there must be lots of that - smooth and quiet?

As i tried to mention earlier at this page when using all 4 voices and playing multiple chords in a row voice stealing could not always be prevented also depending in the attack and release times. With 3 notes this should not be a problem even with longer attack/release.

Some softsynths even when set to e.g. 4 notes polyphony could actually use/play more than 4 voices (especially if thy could use polyphonic Unison) which makes noticing voice stealing or held notes that stop playing very rare or it doesn't happen. With hardware analog polysynths (and also digital hardware polysynths) or proper emulations such additional voices that go beyond the original polyphony or amount of voices (in the case of analog polysynth those are decicated voice cards) are not possible and it is possible that certain hold notes "dispappear" if you reach the voice limit (with last not priority normally the first note that was pressed). For example if you hold a not in a lower octave while play 3 more in a higher octave it would be the low note that could "disappear" if you exceed the voice limit by overlapping notes (which mostly happens with long release times).

Please do not tell me that this kind of behavior is crap because it does not happen in your favorite softsynth...

The "vectorized core" implemented in The Legend currently allows creating 4 voices with only a small amount of additional CPU used compared to the mono mode but this only seems to work that way up to a maximum of 4 voices and not more.

This is a quote from the official website:

The Legend is capable of synthesizing four voices for the cost of little more than a single voice, making it a highly efficient analog-modeling synthesizer plug-in.

This was a further explanation by Richard psoted at another forum:

The key idea in The Legend is to process large parts of the voices, in particular the CPU-demanding VCF and VCA, entirely in parallel using SSE. Since SSE does 4 things in parallel, this means 4 voices poly or unison can be computed in parallel. Note that this is a different concept from developers using SSE to speed up this or that portion of the code. What we do is to run just one large circuit simulation, which then returns the outputs of 4 voices simultaneously. This all happens on a single CPU core, so you fully benefit from your other cores still (e.g. if you run many instances of The Legend).

We could double the voices using AVX to 8, but not everyone has a processor with AVX at this point, and it is not supported in RE just yet.

[Chris-S]

Hi,
are the new "Model-D re-issue" features implemented in The-Legend?

[EvilDragon]

Nope. Velocity/aftertouch support coming in an update.

[Ingonator]

Hi,
are the new "Model-D re-issue" features implemented in The-Legend?

Currently in v1.0.7 additional features in The Legend compared to a hardware Minimoog Model D are:

- patches could be loaded/saved and the Modwheel setting (= mod amount) could be saved with the patch (not included in the real Mini or the re-issue)
- 4 voice polyphony with adjustable stereo spread
- 4 voice Unison with adjustable stereo spraed
- Poly and Unison modes due to the "vectorized core" at maximum use the double CPU amount of the mono mode (usually less)
- filter switchable between 24dB/oct (default like in a real Minimoog) and 12dB/oct
- full ADSR envelopes (instead of ADS or ADSD with Release witched on)
- dedicated fine tune knobs at the Oscs (besides the SEMI knob that allows setting the fine tuning too)
- Delay and Reverb FXs included
- Drive and Feedback parameters (+ additional saturation amount at the back panel)
- switchable between an Early and Late (= default) Minimoog revision (switch at the back panel)
- note priority could be switched between last note and low note
- adjustable pitchbend range (up to 24 semitones, seperately for up/down)
- Futher adjustments to the OScs and filter with trimpots at the backpanel
- FX plugin included (envelopes and Oscs could be triggered by routing a MIDI track to the FX plugin)
- Midi Learn added
- A dedicated RE version for Propellerhead Reason is available besides the VST2 and AU plugins for Windows and Mac

Support for aftertouch and velocity seems to be planned for future updates.

A dedicated LFO module (besides using Osc 3 as a LFO like in the original Minimoog) does not seem to be planned yet but multiple hosts (e.g. Live 9 with Max 4 Live or Bitwig Studio) offer using one or multiple LFOs with automatable parameters. Besides an Arp FX included with Live 9 dedicated Max 4 Live modules (included with Live 9 Suite) also allow adding envelopes, a step sequencer, velocity and aftertouch.

[Mutant]

Someone said maybe in this thread, or maybe on U-He forum, that he managed to get Diva to sound almost exactly like The Legend.

Yes, we are currently at a stage of "almost exactly", but there is a big difference between "x sounds like y and x sounds almost exactly like y".
To my ears The Legend sounds way closer to a vintage Model D for a much wider set of sounds possible on a vintage D.
It is truly next gen.

[fluffy_little_something]

As i tried to mention earlier at this page when using all 4 voices and playing multiple chords in a row voice stealing could not always be prevented also depending in the attack and release times. With 3 notes this should not be a problem even with longer attack/release.

Some softsynths even when set to e.g. 4 notes polyphony could actually use/play more than 4 voices (especially if thy could use polyphonic Unison) which makes noticing voice stealing or held notes that stop playing very rare or it doesn't happen. With hardware analog polysynths (and also digital hardware polysynths) or proper emulations such additional voices that go beyond the original polyphony or amount of voices (in the case of analog polysynth those are decicated voice cards) are not possible and it is possible that certain hold notes "dispappear" if you reach the voice limit (with last not priority normally the first note that was pressed). For example if you hold a not in a lower octave while play 3 more in a higher octave it would be the low note that could "disappear" if you exceed the voice limit by overlapping notes (which mostly happens with long release times).

Please do not tell me that this kind of behavior is crap because it does not happen in your favorite softsynth...

The "vectorized core" implemented in The Legend currently allows creating 4 voices with only a small amount of additional CPU used compared to the mono mode but this only seems to work that way up to a maximum of 4 voices and not more.

This was a further explanation by Richard psoted at another forum:

We could double the voices using AVX to 8, but not everyone has a processor with AVX at this point, and it is not supported in RE just yet.

I don't think you are in a position to tell me what not to tell you
Apart from that, yes, some synths, even SynthEdit ones, have "spare voices" (that are not listed as official voices) in order to prevent problems with long release.


That's interesting, both that even 8 voices are possible that way, and that SA are kind enough to think of those people with more dated processors

[Chris-S]

Nope. Velocity/aftertouch support coming in an update.

Ok. Also the reissue has a dedicated LFO, key modes (hi, lo, last) and different legato trigger modes.

[Ingonator]

Someone said maybe in this thread, or maybe on U-He forum, that he managed to get Diva to sound almost exactly like The Legend.

Yes, we are currently at a stage of "almost exactly", but there is a big difference between "x sounds like y and x sounds almost exactly like y".
To my ears The Legend sounds way closer to a vintage Model D for a much wider set of sounds possible on a vintage D.
It is truly next gen.

Besides other possible differences in teh sound the parameter ranges i nDiva are totally off so re-programming patch sheets is a real PITA wit hDiva while in The Legend it is mostly 1:1.

I also know this because i had recreated some Minmoog patches during the first public beta of Diva and did the same with a lot of patches (100+) during the beta of The Legend.