Definitely an example of an instrument I’d love to have but would go to waste. I can barely squeeze out a scale or a chord on a regular keyboard.
Lack of easy programmability is a typical complaint attributed to the Osmose and other EaganMatrix-based instruments. The Osmose and other EM instruments are foremost designed as an expressive instrument and not as an easily accessible synth. Expressive means that you’ll be able to influence and explore the sonic boundaries of a preset in a deep manner while performing it with your fingers. The macros provide you with additional control akin to the knobs on a Rhodes piano.
In the beginning, the EM engine did not even exist, so the Continuum was mainly a control surface. This changed when Edmund Eagan joined Haken. Even then, in the early EM days, a number of presets were not editable in the EM because they were binary code from Kyma. Later on, those sounds got ported to the EM.
yeah, this was all a very long time ago now
… its easy to forget (or not know, if you are only hearing about this due to the Osmose) , but the Continuum and the Eagan Matrix have been around a very long time, and are. very mature.
there’s some good history, and discussions available via the Continuucon YouTube channel, I’d heartily recommend, as it shows how much thought has gone into all of this.
unfortunately, not quite as simple as this…
whilst some stuff is pretty high level (e.g. resonators), some of the EM is pretty low level stuff, and relies on things like understanding how to excite bi-quad filters, using oscilators as wave shapers (etc etc) and how this relates to your inputs etc.
dont get me wrong, its not ‘hard’, but if you play with the EM for a bit, you’ll realise the actual mechanics of patching is not difficult at all, but rather what you do with it afterwards, how you piece all the bits together.
its similar to modular… individual modules are simple, but the power is in the combination.
just like modular, recreating a simple subtractive voice is pretty trivial (or even a pluck) but doesn’t really get you much (other than a tutorial video ;). ) , but coming up with some new/original is much more difficult… but just needs practice, and investment of time.
some will say its the formula is the off putting part… but unfortunately , that misses the point too… as this is where (I believe) most of the power lies in the EM, and what makes it unique for expressive instruments/control… because you add expression at the ‘intersections’ , so they are core to sound design … not just simple modulation sources added on after, like most ‘mpe’ synths…
to represent this in a modular form (e.g. scale/offsets/attn) or mod matrix, whilst more familiar would create ridiculously large/complicated patches… as the current formulae method is so ‘efficient’ in space… and frankly, they aren’t much more complicated than some of the more complex mod matrices which has many ‘modifiers’ etc.
sure, perhaps EE or Haken will provide a simplified editor one day, but if they do, it will likely have much less functionality and so sound design possibilities.
so I for one, prefer to just dive into the EM, and just start building the experience required … its actually fun, once you get over the initial hurdles
Pretty much exactly what happened with me and my Osmose. Between preordering many moons ago and it arriving it became apparent that I had given up on really mastering instruments with music making firmly in the hobby camp - ie. doing wha tI’m comfortable and good enough at to have some fun. The Osmose, while brilliant, was just a bit too much for me and ended up feeling like a super expensive paper weight.
Folks who can play or are happy to experiment will have a great time with an Osmose though. Like, I know it’s a brilliant device but it just wasn’t for me.
As an ex continuum user who made a few patches from scratch, what is the most annoying to me is having a super expressive playfull surface and having to program it using one of the most rigid ui ever…
Entering manually .001 increments after .001 increments at the intersection of direct + and osc x F is not fun…Turning a cutoff button or at least dragging the mouse up and down feels better…
Actually, to be honest, it is as fun as excel spreadsheet.
MSP even at core level is more friendly to me…
But, I have no other choice so…
funny, my main dislike is that the (computer) keyboard interface is not complete
e.g. I wish you could move around the matrix with the cursor keys.
unfortunately this is a common issue in modern UIs, they dont have full keyboard interfaces, and rely on mouse point n’ click, which is extremely slow and bad ergonomics, since you now have to start messing with two devices … we end up with the worst of both worlds.
in a musical (instrument) context look at trackers, which are having a bit of a resurgence.
with these, the keyboard (later buttons) interface is at the heart of them… and makes them incredibly quick to use…
so personally , I think if they embraced the keyboard a bit more, add a few more shortcuts… you could ‘fly around’ the EM really quickly … just as we can with spreadsheets and trackers
but yeah, for sure, most experience sound design (on a computer) use point n’ click, so I can see many would prefer this.
Yeah I’m assuming it would only be the macro like training wheels interface and the underlying interface would always be the choice of the more advanced user.
yeah, I know how Id do it… (as a dev)
basically, you could construct some models/engines that basically are fixed, e.g. things like a subtractive synth, a blow pipe etc… which the user could choose from. then it’d expose parameters of these ‘fixed architecture’ models/engines.
basically it be similar to what we have now with presets and macros just with more parameters exposed… and perhaps some options on modulations… and not expose the user to the current EM engine.
this could kind of be built as a layer above what there is already… and have a more ‘familiar’ UI on the computer (and not require max), and would basically be building/modifying an EM matrix patch (hidden to the user)
however, there are certain limitations in the underlying EM engine, that would still be present (this is ‘just a new UI’) , which would mean these models/engines would also be still a little ‘different’ from what you’d normally expect.
e.g. its quite tricky to build an adsr envelope in the EM, esp multiple of them.
this is I suspect where there is a bit of a problem, if you try to ‘emulate’ some synth designs, often the EM would look pretty ‘limited’ , because its not designed to be used this way… e.g. you dont need as many LFOs and ENVs - so users may, incorrectly, believe the EM is not that powerful … rather than realise its a ‘mis-translation’… sound design is different on the EM.
also the issue with this approach is it would not help you learn about the EM, should you wish to ‘remove the training wheels’, as the whole approach is fundamentally different.
so tricky on the marketing side to get this message across… and not to have some ‘influencers’ end up getting things confused, and so implying the EM is not that powerful.
( of course, its possible the EM firmware could be radically changed, but given its maturity etc, I somehow doubt this will happen… or at least much less likely, that a new UI ‘skin’)
btw: I still think the first thing we ‘need’ is for expressive e, to update their app to allow preset management e.g. downloading/uploading new presets (from files) and also manage user presets (e.g. re-order them etc)
I said, this in January … when I got my Osmose, and still believe it to be the missing link.
this would enable 3rd party sound designers to create presets and share/sell them.
and users to then ‘install’ them, without ever needing to use the Eagan Matrix.
this kind of completes the vision, that the Osmose can be used by some musicians and never have to touch the eagan matrix and max.
(there could also be some other ‘configuration items’ added to this app)
Its actually pretty straightforward, and basically would ‘replace’ Expressive E’s existing ‘firmware updater’ app… since that function could easily be incorporated into this app.
I think for many, a vibrant 3rd party preset ‘market’ might be enough to mean they dont need to learn the EM… as they have an evolving library of new and interesting sounds for the Osmose.
… and in many ways, this has been the kind of vision of the Osmose/EM.
Interface wise, I think the main challenge resides in the matrix layout. While it has the benefit of being generic, it does not visualise the relationship between the various dimensions (parameters, X, Y, Z, W, macros, etc.) in the order that they will react.
It would already be feasible to design an ordered network kind of visualisation, if the preset’s data was available in more readable format. If I’m not mistaken, the currently available format is all sysex, isn’t it?
The other difficulty is to understand the parameters of each synthesis algo, each of which is shown along the axes of the matrix.
Sorry if it’s been posted recently but does anyone know the ETA for when a new batch will be available? Trying to figure out if I should get one used or hold out if it’s within a month or two.
expressive e says “batch 04 ships nov/dec 2023”, thomman says 15-19 weeks (so similar)
this assumes you reserve now, we don’t know, when you’ll be able to get ‘off the shelf’.
In the small chance that anyone else here preordered from Detroit Modular in the first batch, I just got confirmation that they are in the US and should be arriving to DM store early-mid next week.
I hope to have mine before next weekend.
F&$cking pumped!!!
My Osmose came in from BH today. Enjoying it quite a bit so far. Have only tried out a small fraction of the included patches.
Anyone know if it supports being controlled by MIDI yet? I know you can output MIDI, but after recording the MPE MIDI to my DAW, I can’t seem to get it to play back on the Osmose. One of the MIDI tabs, called “playing”, says “coming soon”, which makes me wonder if being sequenced is simply not supported yet.
You can definitely control it via MIDI, but MPE is a little more complex than standard MIDI. This guide (posted earlier in this thread) contains lots of useful details and instructions for several DAWs:
Right you are. I just had to set DIN mode to “2/5 dsp in”. The MIDI menu is a bit confusing, as is the fact that you can (it seems?) only send USB MIDI controls over port 2. There may be some conceptual structure here that I am not seeing described explicitly (like maybe port 1 is for the physical keys, port 2 for the DSP unit?)
It also looks like just playing back the recorded MPE MIDI from REAPER may not work 100% reliably. Some key presses are not creating a sound, despite the fact that I see them in the REAPER MIDI timeline. In any case, making progress.
I had a similar issue with Live. I resolved it by limiting the number of midi channels MPE used in MPE settings. I’m not familiar with Reaper, but I would expect there’s MPE settings for each channel somewhere?
Yes, it seems like the issue is that the Osmose firmware is happy to output MPE note messages on channels 2-16 (the full range), whereas the EaganMatrix reserves channels 15 and 16 for control messages.
I set “end ch.” under midi > adjust > mpe > mode to 14, and it seems to now only output notes on channels 2-14, which the DSP unit will also happily receive from my DAW.
A little weird that the Osmose defaults to sending MIDI data that can’t be interpreted by its own DSP (which I guess has its own MIDI stack). They should probably change that default and add a warning.
Also, AFAIK REAPER doesn’t have any special MPE support - I’m just playing back exactly what the Osmose sent to my DAW, without doing anything like re-allocating channels.
In bitwig, i use a generic drivenbymoss controller script, which records velocity over time, aftertouch over time, pitch over time and cc74 (or 73, the one allocated to the deep press) and it play backs pretty good. Recording midi is port 2 (with no alterarions to the settings) and playback of the recorder midi is sent via port 2 also.
For f$&@s sake, my preorder has been delayed again!
3 months in a row. What a pita!