Higher DynTime values availability

[mj15003]

Hello Dmitri,

is there any reason why is the DynTime parameter limited only to 60ms? I think it would be useful to be able to set it far longer, perhaps even ten times (600ms). Then even real acustic triggers could be used with Megadrum without double triggering. I suppose there might be some value type limitation (int8_t 0-127), however I don't believe it is not posible to handle this on 32bit ARM MCU.
Anyway, I (maybe also someone else) would be very grateful for this feature implemented in some further firmware release.

MJ15003

[dmitri]

It can be made longer but even with 60ms DynTime fast rolls become problematic. What is the trigger you're having the double triggering problem with?

[mj15003]

I experimented with DDRum PRO Snare triggers (those red). With mylar heads they give nice exponentialy decaying response, however lasting more than 0.5s. I could also fix the trigger input impedance with a resistor which making response shorter but getting lower sensitivity. It produced not double but multiple notes with a single hit even with the longest DynTime and Retrigger.

In order to make those triggers usable with Megadrum I built my own circuit (see attached picture - one channel only) placed between the trigger and Megadrum providing the dynamic threshold function on pure analog basis. It worked very well with input signals of almost any decay time. (then I used Retrigger/DynLevel/DynTime - 8/0/0). It gave me always only a 1-2ms pulse with proper amplitude to be detected correctly by Megadrum. However, the circuit is not trivial and for six channels I needed, it requires 24 opamps, 6 comparators and many other diodes, resistors, trimmers and additional power drained from USB port (even with CMOS opamps).

I did it primary because electronics is my hobby, but otherwise it is a waste of time and money if the same can be done by a (probably simple?) software change. I agree that it might not work with some type of triggers. But according to my own research with osciloscope and experiences with tweaking my analog frontend I realized there is no better algorithm to make it trigger properly than dynamic threshold decaying with time (as is already implemented in Megadrum). Only THE time has to be longer for me (if I am right knowing how it really works in Megadrum).

[dmitri]

Can you make ~0.5s wav recording of (an oscilloscope screenshot) from a trigger:
1. of a single hit
2. of a fast double hit

and post it here.

[mj15003]

Here it is. Single hit and two hits with ~46ms delay between them.

[dmitri]

I will compare it with signals from Drumtek and Pintech snares. But as far as I can see they should work fine with the current range of DynLevel/DinTime.

[mj15003]

Hmmm, then the DynLevel/DinTime function in MD does not work as I supposed to. Maybe I should remove my analog frontend and play with settings again. However, as far as I remember the final impulse to start building the frontend was unsuccessful game with settings.

[mj15003]

Dmitri, please could you describe more technically what is the algorithm for DynLevel/DynTime function used in MD? I would like to find the right values for those parameters more exactly, not by trial and error. I am sorry, but I simply don't understand how you can eliminate retriggers having input signal 0.5s long with DynTime = 60ms. I presume after the DynTime (maybe + 50%) the actual threshold value would drop to the Threshold parameter which should be set quite low to preserve good dynamic range. Or what values would you suggest for those waveforms I posted from my osciloscope. Thanks.

[dmitri]

Sorry, I have not had time still to compare it to my Drumtec and Pintech. I'm not sure what I can add to the description besides what is already in http://www.megadrum.info/content/pads-settings . I think you understand it right. If your trigger produce 0.5s signal then it probably needs longer DynTime but I need to compare it first to my test pads.
Also, when you tried to configure it, did you make sure that the signal is not too "hot"? If it is too "hot", Dynamic false triggering suppression will not work correctly.

[dmitri]

Btw, is possible to adjust those triggers, i.e. tension/pressure? With higher tension/pressure signals should subside quicker.