You need to use it with a great external reverb or delay. It warms it up. However, when putting it into a mix, you most often want to use it dry.
Yeah, I’ve actually used it in a song produced in the box where everything came together quite nice. It just doesn’t carry my songs when working DAWless the same way my Super 6 does.
I’ll try my Boss reverb pedal (the only thing I haven’t tried so far) with it to see if it actually makes me like it a bit more 
i agree to some extent. guess its less of a problem for pad / lead sounds but for bass definitely
Th eagan matrix has the ability to layer 3 patches…
Until Expressive-E gets their hands out of their pockets starts adressing their firmware, to use what’s possible under the hood, you are left with an engine tailored for flutes… And flutes aren’t particularly tailored for deep, punchy sounds cutting through the mix…
I’m a bit exagerating ? Probably ! 
But as an ex 2019 continuum user and despite all the annoucements, the matrix hasn’t changed for 4 years and still carry the same problems…Single patches are thin…
In a classical music oriented quartet you’ll get thru !
Add a Prophet or a Moog to the party and noone will hear you…( The osmose’s pseudo balanced outputs and the few db you lose are not adding any benefit to that )…
I don’t layer sounds internally with the osmose as I don’t want to have the editor running all the time but I almost always layer the output with some vst/synths and my H90 is locked behind the osmose to get a “better” sound…
3 Likes
Korg have announced Wavestate, Modwave, OpSix modules with Aftertouch & Midi 2.0, would that kind of spec make them playable via Osmose?
Sure, you can already play almost any synth with the Osmose. Even if non mpe.
With polyAT, I expect the wavestate to be able to have different bendings per layer which gonna be pure awesomeness !
I think much of the above is ‘inevitable’ after the honeymoon period, as we all become more familiar about that the Osmose IS… and isn’t.
the osmose/haken engine whilst ground breaking, is also a bit of a niche product.
the Eagan matrix specialises in physical modelling and having a tight interaction with its surface.
everything else for me was always going to be a nice bonus… and be a bit limited.
it was never going to be your ‘one’ synth… its too specialised.
its a similar story as an MPE controller, its great at what it does - great for key players.
but the limitation of the Y axis, means its never going to be the most versatile MPE controller.
can EE improve things?
sure… but you cannot change a leopards spots - many of the ‘issues’ are not faults they are part of the design.
all that said, I love the Osmose (and eagan matrix) it is ground breaking and I love playing it…
sure it excels when its played on its own (rather than mixed with others), but thats part of its charm
3 Likes
The Prophet is a very warm sounding synth and the EM’s colder and more percussive patches cut through the mix easily. It’s actually great using both together.
2 Likes
Yeah, badly worded and highly approximative but you get the idea…
Before Osmose I was using a 88Key main controller and two Seaboard blocks on top for specialty sounds, I can’t play everything in the Seaboard but was great for mostly mono sounds but in live situations sometimes difficult to control but for me was the Best keyboard I had in my life and I had a lot. When I read about the Osmose I knew I need it, now is my main controller and play everything there (normally I don’t play piano if so I need the 88) with the addition on top of a mini Keyboard to extend the range or have a second sound available. I’m not using the Local sounds for now but maybe can use some to layer what I already have. I don’t like the sound of Haken alone but can give a great edge to external sounds. I can control any plugin in Ableton Live and with some M4L devices can assign what I like. I hope Expressive fix a couple annoying things but anyway I dont see me going back to the regular 88 Keyboard.
2 Likes
do you mean on the FX or tone side of it? Because as soon as the Chroma Console arrives, Im going to hook them together
Most often its reverb, since I always found the matrix one to sound a bit metallic for my taste and wave shapping/overdrive to get more harmonics and a thicker sound…But that’s not set in stone.
1 Like
i don’t see the big issue with osmose internal sounds! i like them!
2 Likes
I don’t think there is an issue, no synth suits everyone tastes/style, esp. one that has ‘character’ as the Osmose does.
I guess, given everyone (k, social media) labelling as a game changer, it was in danger of being a bit overhyped - but in fairness, there were lots of demos before release, so everyone knew what was on offer.
thats why I said, the ‘honeymoon’ is over, I think, people are now discovering for themselves how its ‘character’ fits (or doesn’t) into what they make - natural progression really.
sure, there is a little room for improve from EE in the firmware for it to be used as external MPE controller… though even there, Id not expect anything too radical.
note: EE had a lot of time to refine 1.0 of firmware/presets due to covid and manufacturing delays, which meant it was in a very solid state at release.
1 Like
There are synths that cover a lot of ground, Osmose not, There are synths that easily can program sounds or adjust basic parameters, Osmose not, There are synths that don’t have bugs at 5 times less the price, But I’m regret that I buy one? Well not If I had the money Ill buy another.
1 Like
Random idea: What if someone managed to train an AI to create EaganMatrix patches based on natural language prompts?
6 Likes
Yeah, like a Synplant 2 with an expressive keyboard
1 Like
This is feasible in terms of data structure because the EM, or any synth engine for that matter, is basically a collection of links between nodes, each link being characterized by W, X, Y, and Z.
However, to train something meaningful (to a user) based on existing presets and ideally without destroying the user’s ears when applying the model, someone would have to enrich those links between nodes with relevant and relatively detailed metadata:
-
What does each link contribute in terms of broader sound aspects such as acoustic, percussive, metallic, pluck, pad, etc. ?
-
Which parameter range for any given link keeps the output musical (vs ear breaking noise)?
Reverse-engineering a preset like this takes a lot of time. Right now, some metadata are available at the preset level but practically none at the link level. So if you asked a model for an analog pad, the output data space would be quite vast and therefore susceptible to noise. Remember the strong recommendation to reduce volume when programming the EM.
1 Like
I think it’s definitely going to be the future of very complex physical modelling type synthesis, one way or another. Some kind of hybrid natural language/manual control interface would be great, so you could get the broad strokes of what you want quickly with a prompt then fine tune it.
EM is a modular synth. If you don’t want to give up on the modular aspect and its flexibility, large language models will only be able to act as a semantic layer / a translator and provide templates or constrained random presets, within an error margin.
Someone would still have to develop this layer. Currently, there are no signs of Haken or Expressive E jumping on this ship.
Such an approach will be relevant for every synth algorithm including FM, wavetable synthesis, etc. The question is one of accessibility and learning curve.
1 Like