On the subject of the internal sound. The 2014 paper on the Striso says:
The sound of the Striso is inspired by a string, where you
can both have natural decay, when plucking or hammering,
or a sustained note, with other actuation like bowing or
resonating with an amplifier
Does anyone know to what extent has this already been achieved? How feasible is it, given the current understanding of the problem, and hardware/software involved?
-Frank
@fkberthold Thanks for raising the question, it’s been a while ago that I wrote that paper. The basic idea is there, the current sound is a combination between decaying and constantly actuated, controlled by all the different parameters from the buttons and motion sensor. I’ve used physical modelling in the first version, but found that hard to control, so went for a filtered oscillator approach later.
There’s still many things I’d like to try and improve in the synthesizer, but as that’s more of a creative process it doesn’t combine well with the current phase of production and solving more practical issues.
There’s also interesting physical implementations to get a similar combination, like the magnetic resonator piano, or the ebow for guitars.
Did you end up sticking with Faust as your main tool for creating the synthesizer?
Yes, I did. I love the way Faust gives a same kind of signal flow feeling like Max/MSP, PureData or Axoloti, while at the same time making it easier to enter more complex functions.
I will post this here as this is relevant to the title of the thread.
After updating to 2.0.3 I can’t get any (1%) response on aftertouch control. I have tried Equator and Vital on desktop and SpaceCraft Granular and Continua on iPad. All of them with nice real time visual feedback.
The internal synth responds well as far as I can tell.
Velocity is also difficult. I have to hit really hard to get to a 90-95%. Most of the time however I’m between 20-45% with “normal” playing.
Edit: there is a solution on the way: calibration. Thinking about picking up my old flute 
@tiantong The calibration hypothesis was probably too quick, I guess it’s just an old fashioned bug. That happens when quickly modifying things without testing. I’ll look into it.
BTW You can always go back to previous firmware releases by copying them to the flash drive, v2.0.2 shouldn’t have this issue.
All right. I wasn’t cautious enough to make a backup when upgrading to v.2.0.3 so I’ve lost the previous version. I’ve sent you the current.uf2 file and will wait for you answer.
I got expression back with v2.0.2.
My Mac always ejects the drive right after copying the file with the subsequent warning message: “drive no ejected properly”. The file was copied though.
Sensitivity is still rough (I don’t know if others have experienced this): I really have to press hard to get a 127 velocity or aftertouch.
and yet, I love how quick it responds to a light touch. Also the Y-axis control is nice. At least with the board on the table there is a lot of control. The note layout is logical and easier to get than what I initially thought. I like it a lot.
Ok, I’ve posted a bugfix release which fixes the aftertouch.
Since the flash drive is not a normal drive but a fake drive this is nothing to worry about. However I’ll have a look if there’s a way to remove the warning.
Thanks for your notes about the velocity and aftertouch. I’ll have a look at it and hope to come with a solution soon. The response curve is very linear up to about 400 gram, after that it gets less sensitive, it would be good to correct for that.
Note that you can already choose between 3 levels of sensitivity via the settings button. See the settings card in the box.
Great to hear, those notes make me really happy!
Well, I’m glad you’re being so responsive 
Aftertouch is perfect for me now. I’m all over the place with the settings. I’ve got everything to high sensitivity (velocity, aftertouch and pitch bend). The last I still don’t get but I before I say something I want to spend more time with it.
With aftertouch I still have to squeeze to come to the last 5% but here we get into very subtile territory. I’m measuring both in the amount of strength I have to apply and how squeezed the buttons are.
Regarding velocity I have to say that I have managed to get to 100% but hitting is like a hammer, very violent and I have to be very precise, 90 degrees against the board so the buttons don’t tilt in any direction. Most of my tapping is between 10 and 50% (pinky finger is between 1 - 10%). I still love how well it responds to low velocity but somehow I would love to get max amount of expression by being able to get that 100% without using a hammer.
But I’d like to hear the experience of others too…
Equator is proving an excellent testing playground with all that visual feedback.
So a few months later and now that I’ve decided to sell my Linnstrument, the Striso could be my main controller but honestly the wobbly silicon pads are still my biggest issue with it.
I can clearly see the increase in sensitivity but still can’t manage to get to full velocity and so there is a lack of expressivity. I feel like all the energy I apply is lost and wonder if it wouldn’t be better with wooden buttons (like an accordion?).
So @pierstitus how much of a project would be to replace the keys? As a DIY?
Other than that I love the form factor and I’m still impressed on how fast learning the new layout has been. It feels really natural.
I don’t think that solution would be a solution at the core of the issue.
I amplify / rescale the velocity in my DAW. I like the native sounds as is, but I agree that the output midi needs higher velocity. I don’t think this would be hard to do at the firmware level!
Well, I recall @pierstitus having conflicts between rescaling and ghosts notes. My problem extends to the physical interaction with the keys. They remind me of candy: I want to bite them
Good that you raise this point again, the velocity range is indeed easy to fix. However I’m reworking the sensing logic, so didn’t spend time tweaking the sensitivity in the old logic. That of course takes longer than wanted and got pushed over the summer holiday, so for now I’ve released a firmware with higher midi velocities: Release v2.0.5-4 · striso/striso-control-firmware · GitHub. I hope that works for now, I hope to release the reworked sensing firmware soon too.
I don’t notice an increased velocity range, but I did notice something else: the accelerometer channels now send position instead of acceleration, which makes them massively more useful!
I’ve patched x-axis to a simple pitch shifter and y-axis to a bitcrusher to experiment with it, and it’s works great 
I do have noticed differences in the three velocity modes. I still have to decide which one is better for my needs (I guess it depends on the sound source) but for the moment (only one quick test) I stay in the middle.
I may have forgotten to set that to high after the firmware update 
Thanks for reminding me!