Prototype Wednesday: MIDIKalimba update

[@musicalmiscellany]

Great updates Shiela! Now that I see the 9-pad version, I really like it. You bring up an excellent point with the tritones in the new arrangement. That could be really useful. I downloaded the firmware update and you are correct. The new menu system is a lot more intuitive, even with the 4 buttons. I need to mess around with the velocity setting more. I'm so anxious to see how the product version materializes.

[@musicalmiscellany]

I played around with the new firmware over the weekend. The velocity really is well implemented. I think you did a great job and I didn't expect it to bet that easy. Ironically, I find the velocity to be less useful on the hammered sounds (I know I was the one to request it LOL) but extremely useful on the synth & string sounds. It's so nice with pads. This thing is awesome.

[@SianaGearz]

I think velocity estimation where you figure out how fast a pad was hit is the "correct" way followed by most instruments as opposed to pressure sensitivity.

[@midi_in]

I'm sure you're right, and I guess with pressure we get into aftertouch.

[@WhatHoSnorkers]

That makes a huge difference!

[@midi_in]

it certainly does!

[@marcushlm]

Very cool project! Would totally buy this, maybe in a DIY kit or something like that. I suggest you add some sort of mod wheel on the side, it would add a lot of performance options. Keep up that good work!

[@midi_in]

A modwheel is a very interesting idea! Thanks for your interest.

[@thefishermenandthepriestess]

Super cool!

[@tangentspace]

very cool and inspiring!

[@arthurcrime]

Sounds lovely, I enjoy hearing your thoughts, this device has been carefully thought through, velocity sensitivity is a game changer. Regrettably I am in the doldrums financially. I do so enjoy your content.

[@midi_in]

I always appreciate you watching and commenting. Thank you.

[@jamesgazin9447]

I really like this and will link to your page so I can see when it is available. I've been pecking on a Hugh Tracy kalimba for for a few decades and am used to a lot more keys. I'd like to see one change on your prototype: There is room for two more pads perpendicular to the lower left and right corners. I'd like to suggest that you use the space to add "shift" keys. The one on the Left could shift the output DOWN one octave, and the one on the right could shift UP. That way, you could have a full range Kalimba that only exposes one octave at a time. I think it would be very playable. Since everything is digital, you could have modes that do other things, such as add tremolo or reverb. Thanks for the demo! Edit: You could have a switch on the side, along with LED indicators to make the Shift pads "Sticky" so the player could quickly switch between modes and stay in each mode for awhile.

[@midi_in]

Thank you for your interest and your suggestion. That makes a lot of sense.

[@billcosgrave6232]

I'll be buying one!

[@midi_in]

.. is the correct response! lol. Thanks Bill, watch this space.

[@mmpp7547]

that is so cool

[@Miester7]

This device is amazing. Can you use it as a digital drum kit?

[@midi_in]

I've been meaning to try that. Yes, I just need to invent a new 'scale' where each pad maps to the midi note of the drums that we want and fix it to channel 10.

[@Miester7]

@@midi_in Thats so cool. I will buy one as soon as available 😁

[@EvilVitJoker]

If it's not a secret, how did you implement the velocity sensitivity? A few years ago I made some prototype electronic flutes similar to your kalimba. But I can't figure out how velocity or pressure is measured in this case

[@midi_in]

It's not a secret but it makes a world of difference to this device, which looks like it's going in the direction of a commercial product (alongside my e-ther and MIDISID) so forgive me if I'm not too specific. But the particular capacitive touch chip I'm using produces a lot of data over and above just the touch / release, and does a certain amount of filtering / processing on-chip. I've been able to get a really good reading for velocity, even with the charge time kept to a minimum to reduce any lag down to negligable levels. I did think about other ways to do it, such as fastening some kind of sensor under the top board, but this way works so well, it hasn't been necessary to explore other routes.

[@EvilVitJoker]

@@midi_in Thanks. You pointed me in a new direction. Somehow I haven't thought about capacitive sensors.

[@IamSamsilver]

How do I get it

[@midi_in]

This is still at prototype stage, but I am expecting a small run of boards early next week, and if they're good then there may be a small batch for sale soon. I'll probably post another video to show those off.

[@IamSamsilver]

Okay but I live in Nigeria I can help market it here

[@Juksemakeren]

how is the velocity sensitivity implemented?

[@midi_in]

The capacitive touch chip that I'm using produces a lot of data and can do a certain amount of processing/filtering of that on-chip. You wouldn't think that you could use that to infer how heavily the user hits the pads, but it's really effective, even with the charge time set really low to keep the latency negligable.