Multiple zones with just two piezos

[Sylv1co]

DDriver, my approach should be to use a softpot as long as the pipe and plug it into the hihat pedal input of megadrum. The piezo should be pluged into the hihat cymbal input. It should work just like an hihat, MD send different notes depending on the resistance of hihat pedal input. You can even add two membran switches to have more notes.
Dmitri, is another hihat input on MD is possible in another MD firmware?

[ddriver]

I haven't calculated anything, I have read a lot of articles about MIDI devices through USB connectivity and latency seems to be pretty much a mandatory issue there - for one, MIDI over USB seems to be limited to 1 midi event per msec, and since it is a serial bus, transferring 5 note polyphony will require 5 msec, this is aside from the actual driver implementation that receives data from usb and interfaces it to midi.


Sylv1co - I get the idea, but as you said, it will require more than one "mod" channel, are softpots sensitive to the actual pressure or only to the location of pressure?

I have been presently surprised with the megadrum editor software, it is more than any of the commercial products I have dealt with have to offer, however one thing I didn't liked was the fixed or static approach. It would be best to have custom modular functionality, mono, stereo and mod modules, of which you add as many as you like and just assign them to whatever channels you want, this would be pretty easy to implement on the PC, but with the Atmega flexibility and modularity might prove to be an issue, I guess Dmitri knows best.

[dmitri]

Dmitri, is another hihat input on MD is possible in another MD firmware?

Possible but it is not on my priorities list.

[dmitri]

I haven't calculated anything, I have read a lot of articles about MIDI devices through USB connectivity and latency seems to be pretty much a mandatory issue there - for one, MIDI over USB seems to be limited to 1 midi event per msec, and since it is a serial bus, transferring 5 note polyphony will require 5 msec, this is aside from the actual driver implementation that receives data from usb and interfaces it to midi.

Again, what "actual driver implementation that receives data from usb and interfaces it to midi" are you talking about? What are the evidences that it is a bottlneck?
Standard MIDI is also a serial bas and has the same limitation of one MIDI message per msec.
We are dealing with a drum trigger here. Usually a drummer has only 2 hands and 2 legs which pretty much limits how many MIDI events a drummer can generate at any given moment. With this in mind I'm struggling to imagine a drummer who consistently generates 5 MIDI events at any given moment simultaneously.

I have been presently surprised with the megadrum editor software, it is more than any of the commercial products I have dealt with have to offer, however one thing I didn't liked was the fixed or static approach. It would be best to have custom modular functionality, mono, stereo and mod modules, of which you add as many as you like and just assign them to whatever channels you want, this would be pretty easy to implement on the PC, but with the Atmega flexibility and modularity might prove to be an issue, I guess Dmitri knows best.

I don't get you. In MDM you can configure every input of 32/56 inputs as individual (mono) input or join them into dual (stereo) inputs. What other modules are missing?

[ddriver]

Modular approach would allow exactly what Sylv1co asked for - multiple hihat type of controls instead of just one. Overall nothing is missing, besides "customizability" of the megadrum interface - the ability to instance modules according to specific needs rather than relying on a static interface.

I cannot shake the feeling of miscommunication, I have great respect for your work and I don't really enjoy the type of responses I provoke in you, so I'd better go quiet until it is time to order, sorry if my posts irritated you in some way.

[dmitri]

Modular approach would allow exactly what Sylv1co asked for - multiple hihat type of controls instead of just one. Overall nothing is missing, besides "customizability" of the megadrum interface - the ability to instance modules according to specific needs rather than relying on a static interface.

The is no point in supporting something in MDM which is not available in MegaDrum itself.

I cannot shake the feeling of miscommunication, I have great respect for your work and I don't really enjoy the type of responses I provoke in you, so I'd better go quiet until it is time to order, sorry if my posts irritated you in some way.

No irritation at all, I'm just trying to understand the source of your concerns, how you did the calculations and how you singled out "bottlenecks". I may be missing something.

[Sylv1co]

Sylv1co - I get the idea, but as you said, it will require more than one "mod" channel, are softpots sensitive to the actual pressure or only to the location of pressure?

Softpot are only sensible to location but you can use it in conjunction with a long FSR to get pressure and location detection. see here for long FSR : http://www.robotshop.com/eu/interlink-2 ... fsr-3.html

[ddriver]

dmitri - As for the usb->midi latency issues, I think it is pointless to drag ourselves into a discussion involving third party drivers and such, we both know executing code takes time, that different drivers work in different modes and priorities and stuff like that. To me it is irrelevant since another user has already reported there are no latency issues when using midi and a hardware sampler, a good way to avoid all driver and other software related latencies.


I think it was implied that this type of functionality will have to be present in the megadrum, since the manager is just a front end to program it. The idea is to allow the actual MD to be more flexible and configurable, however as I expressed my concerns a few times, that I don't know whether the Atmega chip allows for that kind of flexibility. It would be great to be able to add multiple mod/hihat type of controls and to generally do more with sensor information before it leaves the MD in the form of midi, like for example that first thing I asked about - calculating relative position of impact using two or three piezos (maybe the right term for it is POSITION SENSING as I've already seen your post on the matter), and to map all the available modules to any channels at will, however it is your project and no one but you is in the position to decide such things.

And don't get me wrong, your product already offers excellent functionality compared to commercial products, not to mention unbeatable price/performance/features ratio, it is me, I am the type of person who always seeks room for improvement, it is good that your product is that good, but it will be even better if it is better And since extra functionality only requires extra code, writing it once benefits everyone, which is the great thing about software.

[dmitri]

it is me, I am the type of person who always seeks room for improvement

Look at the changes log in Latest MegaDrum firmware version. Maybe I'm also " the type of person who always seeks room for improvement"?

[ddriver]

I sure hope so, in fact I strongly believe this quality is necessary for each and every one of us, if the world is to get any better

Even without looking at the frequent firmware updates, I have examined all DIY project I could find on the internet, and the reason I joined here is because I found yours to be the most advanced by a quite a large margin. I strongly approve what you are doing and will be glad to contribute in any possible way.