StiX : MultiSynthesis Drum Machine (v1.6 Released + Flash sales)

[AnX]

@mcbpete
You need to change the update speed of the seq cursor. By default its way out and useless. Set it to high/fast and it works fine. Really dont know why there is an option. It should be accurate by default.

[mcbpete]

@mcbpete
You need to change the update speed of the seq cursor. By default its way out and useless. Set it to high/fast and it works fine. Really dont know why there is an option. It should be accurate by default.

Wait, let me get this right - there's a option for the rhythm generator to stay in or out of time, and the default is go out of time ?! I've got faith in Xils with their synths but ... why am I suddenly getting buyers remorse !

[PALU]

@mcbpete
You need to change the update speed of the seq cursor. By default its way out and useless. Set it to high/fast and it works fine. Really dont know why there is an option. It should be accurate by default.

Wait, let me get this right - there's a option for the rhythm generator to stay in or out of time, and the default is go out of time ?! I've got faith in Xils with their synths but ... why am I suddenly getting buyers remorse !

Not the engine - just the GUI. I found it annoying too

[mcbpete]

Ah gotcha thanks, so that solves the misaligned steps thing Do you know if it solves the tempo change issue too (as in Stix only matches the host DAW tempo at the start of each pattern cycle) or is there a different hidden option to do that ?

[Lotuzia]

I'm liking some of the sound demos but I'm finding the plugin itself, and the preset sounds, a bit harder to get on with.

I was surprised that there was no way to create a 'proper' clap, which seems to be a strange omission. It would also have been nice to see some tools for 808 style cymbals (ring mod?), or maybe some comb filtering.

In general this synth seems to have a kind of blurry quality, which might on some occasions be pleasingly 'analog' but I think may be symptomatic of a lack of synthesis tools to deal with sounds with a lot going on in mid to high frequencies (other than FM which, from the sound demos, does seem to come into its own when using the internal sequencer).

User error/unfamiliarity is of course likely but I got the same feeling from going through the presets. Not only claps were missing, but any variety of snares and cymbals (discounting samples). At the very least, would it be possible to put a tilt filter on the noise oscillator? The 'tune' knob could be used as it seems to have no function if noise is selected.

I recently bought RMV, finding that it was still on sale, which augments a general percussion synth module with specific modules for clap, snare etc. I appreciate that you are probably too far down the road to consider this approach (or already discounted it for whatever reason) but this really does seem to be the best way to deal with the variety of synthesis methods that seem to be necessary to make interesting electronic percussion.

Thanks for trying it.

There are 12 claps in the current Dtumpad beta factory library. There will be more when the 1.0 will be released, also some made with different techniques. ( to find them easily, sort the drumpads by 'subgenre', and they are all in the Snare sub category ) There are several full analog claps among these 12 ones. I'll post an audio clip later to illustrate this. They might not be your traditional 808 claps, but it's there. We'll however have the Claps covered in a minimal way with the demo kits from our partner Wave Alchemy, with all the susual suspects, and 1rst quality samples.

Same with the snares. There are a lot of snares, from all variety, but we'll add some more in the future.

Interesting is the discussion about *specialised synthesis models for each kind of sounds*.

But first I'll mention that the RMV added the 'generic drum synth module' years later after the initial release that RMV is now discontinued, when StiX is still in -advanced- beta stage. Then RMV has nosequencer, so you'll loose all the sequencer AND synthesis engines/sequencer interactions. I'll just add that RMV was an ambitious project. And finally that StiX doesnt sound at all like the RMV.

Let's go back to the discussion 'specialised drumsynth modules' vs 'Global model'. I choose the general model for sevral reasons, one beeing the interactions beetween the different sound engines. In StiX you can do virtual analog, FM, Samples, and cross synthesis. Meaning that all the module can modulate each other. This makes a difference. Also, the ineractions with the sequencer required a kind of 'unified' synthesis engine. If you look closely to 'different audio engines' vs 'Stix' you'll see that StiX has way more parameters, oscillators, envelopes, and actually everything else to allow you yo make a 'drum sound'.

This doesnt mean btw that StiX can do all that RMV can do ( no slices and loops for example), or specialised tr-xxx emulations, wich require a special topology for each drum modules, but in counter part have a limited set of parameters per sound. So to each his own. I said it before, StiX is not here to replace all drum machines, it's there to propose an 'all in one' solution, that can compmlete an existing arsenal, or become your main drum machine unit, with a lot of unique features as well, that you wont find on any other drum machine.

My primary goal was to achieve a virtual analog drum machine like in 'make all the drum sounds that an harware synthesizer can do' and not 'have 50 different drum synthesis modules to emulate 100% all the drum machines of the past'. Then I added FM, Sample and Cross Synthesis on board, all related and with interactions, because I wanted StiX to have a wide palette of sounds. And meanwhile I tied all this with the -unique and redefined- sequencer, the general macro concepts (like Stretch) to have a very fast worflow, and the unique functions, like polystep.

So that, finally, StiX have 'a sound' (and thanks to Xils oscillators, filters etc expertise maestria, a deep/warm/woody sound, that I could not find with RMV for example) and 'unique features'. I'm not expected it to please 100% of users. But well, it's there, and what you can get with it, you will not get for the most part from other DM.

Now, the Tilt filter you mention is a good idea, and it's true that the non active tune knob when noise is selected for the 3rd oscillator has tempted me fore a long time. We'll consider it. I also have other plans to increase StiX versatility in the future ( For example a special pulse oscillator, or a delay for the oscillators, or prefiltered noise sources could be some possibilities as well to obtain another kind of 'claps', or more generally still another synthesis tree to build different sounds) Time, testing and experience will decide wich solutions seem to be the *best', and compatible with StiX current engines and library legacy. Please remember that StiX in still in beta, and is just born.

[Lotuzia]

Ooh just thought of something while playing around with the full version (and going wild with the random button!). How about a probability slider in the step sequencer, something like this:



ps (and more importantly) I think I've spotted a possible bug (Running through Ableton 9.5, Windows 10 if that makes any difference): The BPM in StiX only seems to get updated after it's played through a complete pattern (or until you restart transport on the DAW).

As an example: say you start the pattern with your DAW set to 40 bpm and then quickly turn the DAW bpm up to 120 bpm .... StiX will stay at 40 bpm during the length of the whole pattern, and once it gets back to the first note in the pattern will only then jump up to 120bpm. This'd mean that any project with a variable BPM would be completely impossible for Stix to stay in sync. Big eek !

pps spotted another bug - I accidentally went extreme with the BPM (slipped on Ableton's BPM slider!) and put it up to around 600 BPM, when I put it back to a sensible speed again I noticed that some of the drum hits had moved out of time. Here the snare drum is one step behind and the closed hi hat one step ahead :



I could get it to do it every time by repeating the above (going to a crazy high BPM and then back down to a slower speed) and a random instrument or two on Stix would be ahead or behind by one step ....

Yes probablitily is tempting. We have thought of it for a long time. If we find a clever way to implement it, we might implement it the future. There's only room for one additional line in the Sequencer single line advanced view, so we'll think twice, or seven times, before deciding how it should be used. Probs could also be global, or per line, or even *intelligent* -but this would require the dev of a whole engine, so a very long time- .

I'll add that if you want to make more *controlled* variations to patterns, you should also consider the mute drumsounds possibility ( octave c2) and the possibility to simply add drum hits by playing them on the keyboard. When you use these two functions, you get 'controlled variations' ( wich, admitedly, is different than 'random variations' )

Yes Stix resyncs at the beginning of every bar. I'm not sure we can do something about that, and that it reflects in real time tempi changes. But we'll see. can't promise anything. As for some beats changing positions after drastic tempi changes, does it alter the true position of the hits (ie what you hear) , or is it just a graphic issue ? ( Saying this because I cant see how it could change the beats triggers )

[Lotuzia]

@mcbpete
You need to change the update speed of the seq cursor. By default its way out and useless. Set it to high/fast and it works fine. Really dont know why there is an option. It should be accurate by default.

Wait, let me get this right - there's a option for the rhythm generator to stay in or out of time, and the default is go out of time ?! I've got faith in Xils with their synths but ... why am I suddenly getting buyers remorse !

Not the engine - just the GUI. I found it annoying too

Yes GUi refreshing speed option is there to save CPU if you choose slow/medium.

So for example, you could work with StiX while creating rythms at high refreshing speed, then turn it to low when you work on other instruments, or dont need no more the high settings.

We could set 'high' as default, but it would add a penality for those who have machines with a moderate, or less cpu, power.

Otoh, as your posts demonstrate it, it could be seen as bugs by people who are not aware of this option.

The real definition of ' a dilemna'. When no absolute good decision is possible. I think that if the flaws of the slow refresh option are not so disapointing ( like changing the tempo from 600 to 100 can be very unusual) we might let the default as it is (ie favoriting arbtirary people with less powerfull machines ). What do you think ?

[Lotuzia]

And finally, though we're patiently reorganising every *block/section* of the GUI/UI, a more global rework, to give more punch, depth and clarity to the GUI will also be made/tempted. It should help to identify at a glance the different areas of the GUI, and break the global *monotonous* feel.

[mcbpete]

Yes Stix resyncs at the beginning of every bar. I'm not sure we can do something about that, and that it reflects in real time tempi changes. But we'll see. can't promise anything.

Oh my, that is a huge problem if that can't be achieved - My albums usually segue from one track to another sometimes using field recordings and sometimes through drum patterns that transition from one track BPM to another BPM. It's also a trick I use for live performances from going from track to track, not being able to gradually shift the tempo is very much a sticking point !

Yes GUi refreshing speed option is there to save CPU if you choose slow/medium.

So for example, you could work with StiX while creating rythms at high refreshing speed, then turn it to low when you work on other instruments, or dont need no more the high settings.

If that is indeed the case (the GUI being slightly behind/off rather than the sound engine) in order to save CPU then that's actually a good call and makes a lot of sense I'm guessing the reason the line can drift off time is because it sometimes needs to when playing back polyrhythms ?

[Lotuzia]

Yes Stix resyncs at the beginning of every bar. I'm not sure we can do something about that, and that it reflects in real time tempi changes. But we'll see. can't promise anything.

Oh my, that is a huge problem if that can't be achieved - My albums usually segue from one track to another sometimes using field recordings and sometimes through drum patterns that transition from one track BPM to another BPM. It's also a trick I use for live performances from going from track to track, not being able to gradually shift the tempo is very much a sticking point !

Yes GUi refreshing speed option is there to save CPU if you choose slow/medium.

So for example, you could work with StiX while creating rythms at high refreshing speed, then turn it to low when you work on other instruments, or dont need no more the high settings.

If that is indeed the case (the GUI being slightly behind/off rather than the sound engine) in order to save CPU then that's actually a good call and makes a lot of sense I'm guessing the reason the line can drift off time is because it sometimes needs to when playing back polyrhythms ?

Good news, oh well ... news at least : There's a chance that Stix can follow gradual temp changes sometimes in the future VS tempo changes beeing followed on the 1rst beat of next bar. Not 100% sure, and not for v1.0 but .... chances there are.

And yes it's mainly because of the polyrythms etc.

[mcbpete]

I'll cross my fingers - Thank you for finding out

[Lotuzia]

Claps : Here's a quick rough audio clips with claps. All belong to the current factory library (beta 0.96). The first three are only made with the analog engines. The 4rth is a sample, and the 5th is the beginning of a draft version of a track I made with StiX. All the drum sounds in this track are made with the analog engines.( but I'll change the HH in the def version)

Some StiX Factory Claps A

SOoo : StiX can do some *claps*, by synthesis, or samples, or FM, or else ( I'm currently investigating claps made with cross synthesis, and the results are very interesting _imho- Ymmv.

[jbuonacc]

(EDIT: i know all this sounds harsh, but i'm just picking apart a few points that were brought up. for the most part, i think the existing drum synthesis in StiX is fantastic. just trying to expand/build on that.)

Interesting is the discussion about *specialised synthesis models for each kind of sounds*.

what's "interesting" is that you didn't seem to take this into account from the start. i would think that studying the SOS 'Synth Secrets' drum tutorials (for example, among other things) should have been first priority. i'd also suggest that Xils Lab obtain a Nord Modular G2 as a "development tool", plenty of complex drum synth patches made by users over the years to use for reference. steal from them for ideas. seriously, at least get the G2 Demo software running and check out some patches from the user library. an Elektron Machinedrum might not be a bad idea either (at least read through the manual), each instrument type has a specific set of parameters.

please don't mention "samples" as part of this discussion at all, it's great that StiX has that option, but it's nothing but a workaround in this case. plenty of classic x0x (and others) drum sounds are pure synthesis, and there's no way that StiX can pull them off currently.

... I choose the general model for several reasons, one being the interactions beetween the different sound engines. In StiX you can do virtual analog, FM, Samples, and cross synthesis. Meaning that all the module can modulate each other.

wtf? as if you couldn't do this if the 'engines' were more specialized? come on, man. please.

If you look closely to 'different audio engines' vs 'Stix' you'll see that StiX has way more parameters, oscillators, envelopes, and actually everything else to allow you yo make a 'drum sound'.

when it comes down to it, this doesn't really mean a thing if it can't do the sounds that people want.

... Please remember that StiX in still in beta, and is just born.

i think you're off to a great start. i don't see any reason why this can't evolve to being one of the best drum synths out there. no reason why it couldn't eventually have "50 different drum synthesis modules to emulate 100% all the drum machines of the past". honestly, i think this should be one of the goals. why not? you've spent a great amount of time working on a very unique sequencing engine, why intentionally limit it in regards to synthesis?

[Lotuzia]

(EDIT: i know all this sounds harsh, but i'm just picking apart a few points that were brought up. for the most part, i think the existing drum synthesis in StiX is fantastic. just trying to expand/build on that.)

Interesting is the discussion about *specialised synthesis models for each kind of sounds*.

what's "interesting" is that you didn't seem to take this into account from the start. i would think that studying the SOS 'Synth Secrets' drum tutorials (for example, among other things) should have been first priority. i'd also suggest that Xils Lab obtain a Nord Modular G2 as a "development tool", plenty of complex drum synth patches made by users over the years to use for reference. seriously, at least get the G2 Demo software running and check out some patches from the user library. an Elektron Machinedrum might not be a bad idea either (at least read through the manual), each instrument type has a specific set of parameters.

please don't mention "samples" as part of this discussion at all, it's great that StiX has that option, but it's nothing but a workaround in this case. plenty of classic x0x (and others) drum sounds are pure synthesis, and there's no way that StiX can pull them off currently.

... I choose the general model for several reasons, one being the interactions beetween the different sound engines. In StiX you can do virtual analog, FM, Samples, and cross synthesis. Meaning that all the module can modulate each other.

wtf? as if you couldn't do this if the 'engines' weren't more specialized? come on, man. please.

If you look closely to 'different audio engines' vs 'Stix' you'll see that StiX has way more parameters, oscillators, envelopes, and actually everything else to allow you yo make a 'drum sound'.

when it comes down to it, this doesn't really mean a thing if it can't do the sounds that people want.

... Please remember that StiX in still in beta, and is just born.

i think you're off to a great start. i don't see any reason why this can't evolve to being one of the best drum synths out there. no reason why it couldn't eventually have "50 different drum synthesis modules to emulate 100% all the drum machines of the past". honestly, i think this should be one of the goals. why not? you've spent a great amount of time working on a very unique sequencing engine, why intentionally limit it in regards to synthesis?

No probs. Seems you describe your dream drum machine.

Just to adress a few points :

* SOS articles are on my HD for ages, and long before I could imagine that StiX could exist one day. I used the methods in a lot of my presets for different synths. It's precisely because of these articles, and later having studied more in detail certain circuits, that I decided a unified synthesis engine would be better than to handle 'dozens of different topologies with different parameters sets that can't interact'
* Can a Vermona make the same sounds than a tr-808 ? No. Reverse is true. Is Vermona better ? Worse ?
* Stix synthesis engines and sequencer interactions request an unified global synthesis engine. Not ' different modules with different topologies with different parameters* for reasons I wont develop here.
* We have ideas to add new possibilities to the global synthesis engine. And we welcome all suggestions ( see my post above about 'claps' )
* Can a tr-808 make the drum sounds I can do with My ARP Odyssey ? No. The reverse is true. My Odyssey can actually produce a wider palette of sounds. And StiX can produce a wider palette of sounds than my Odyssey and a 808 together.
* You might have missed a good part of the power of the implemetation of the sample engine in StiX. Its not only a sample player. You can do cross synthesis with it, ie modulate any synthesis parameter, including va oscillators for example, with the sample. This open new possibilities.
* But even if you would have only considered loading a tr-808 *ride cymbal* in it, you can also layer it with the analog and/FM engines, or make it come alive with all the goodness of Xils filters, oscillators, analog modelled enveloppes and various modulators. Very few sample players have this on board. Ok. Actually, none offers this precise combo.
* Should people want only a few dozens sounds of the past emulated in the most more/less accurate way, there are units available for that. Just buy them all, pile them, possibly deal with separate sequencers simulataneously -if they have one- to operate them, and you're done. This said Stix will approx more/less all these iconic sounds, one way or another. But it's much more than this. And I think that in 2016, some people might want more. At the very least, I do.

So, StiX is not *intentionnaly limited*. Has never been. Will never be.

It can produce a very wide palette of sounds. And these sounds are 'alive', 'organic'. And play them via an innovative and next level interactive sequencer. It can do things that no other DM can do, including sounds and sequences. It has a pristine VA engine, and many other multi and cross synthesis possibilities.

This said it will never be 'the drum machine that can do everything' and 'that kills all other drum machines' If you have read me sometimes, you know that I dont believe at all in this kind of things and commercial blurb. Otoh, I think its already an excellent drum machine, and already possibly 'one of the best out there'. Then, of course : Ymmv.

And it will evolve. Adding new possibilities, and new nodes.

[plugin'd out]

Thanks Lotuzia for taking the time to make such a detailed reply to my post. And thanks also to jbuonacc, you articulated a lot of what was going through my mind when I was writing it.

I was initially thinking in terms of feature requests but I could see that adding more stuff would clutter up something that is already crowded which is why I dragged modules into it. And of course throwing in every possible feature to try to keep everyone happy would just make a mess of a synth.

I'm interested by your comment about making a drum machine that does all the sounds that a hardware synthesiser can do. Some earlier posts make much of the quality of the resonant filters, it seems to me that this is far from being the most important thing when creating percussion sounds. Percussion sounds are very short but potentially with a lot of detail over the course of the sound, both in the attack and the decay, so things like fast envelopes/lfos (up to audio rate?) - with the ability to fine tune at higher than usual speeds - and adjustable envelope shapes (giving more options for sculpting kicks, for example) become more important. I just wonder if looking at this through the lens of hardware designed primarily for creating melodic/harmonic content is a bit limiting - was it ever that great at percussion sounds in the first place? Analog hardware designed for percussion (or what can be done with a full on modular) is of course another matter!

On the subject of the clap (yes I know there are claps, but these are effectively variations of white noise snares) I could not recreate the clap sound that most people would recognise, i.e. a short roll of noise bursts followed by a single more diffuse one. Perhaps a simple way around this without messing too much with the synth as it is would be to allow a second oscillator to double as a noise source, then one oscillator can have a repeating ramp down lfo for the attack while the other could be used with different treatments for the body. But another problem is that lfo currently only modulates filter or pitch - if it could modify amp as well it would be much more useful.

I don't want come across as a clap obsessive, that was just one of the first things that struck me when I tried the demo. I had the same feeling about other sounds with a noise component or in higher registers. I'm also not that interested in samples in drum machines with synthesis engines - not because I never use percussion samples but because the attraction of percussion synthesis is the possibility to shape the sound. Samples also tend to break the underlying coherence of the sound of the synthesis engine.

Sorry if this comes across as rather negative - unlike one or two of the posters here there are things I like about Stix and I would love to be able to like it better - if I wasn't interested I'd just move on.