Flam with simultaneous pad hits

[jarosz]

While playing MegaDrum I've been upset about unexpected flams (over 20ms reported by MDM FX) with simultaneous hits on two pads. Especially with snare, when I felt pad vibration on the stick and heared the sound with a very noticable latency.

Forum search revealed similar problems reported earlier, but fixed with firmware updates, so I thought maybe it's the device's limitation to sample accurately only one pad at a time in around 20ms time frame. But then I discovered that if I hit precisely more than two pads at the same time, only the second stroke is flammed, and MDM FX reports 1-3 ms interval between suceeding hits. I wish I had 1-3 ms interval on the second hit as well, but I don't know how to debug the problem.

It's not easy to hit 5 pads at the same time, but I dit it for comparison

I'm using 32 input MegaDrum with PS board FW version 20180121.

Alt Sampling Alg is on. When it's off I sometimes manage to get simultaneous hits flammed only with 13 ms interval, but it's still noticable and overall triggering performance is worse.

MIDI
36 is Kick
48,28 - high tom
47,27 - mid tom
43,23 - low tom
41,21 - floor tom

[dmitri]

I guess it is cross talk suppression is doing it's job. Either reduce XTalk Level, or place the pads into different XTalk Groups (if they are isolated mechanically).

[jarosz]

I've made a video in which I tried to enclose my problem.

https://www.youtube.com/watch?v=qa9u3NVnJFo

The video contains me trying to play simultaneously on 2, 3 and 4 pads. You can watch and hear four hits with pad noise, and then the same four hits with VST. Maybe not all hits are perfect, but I can hear the difference and see it on the MIDI Log. Sometimes hits are very low velocity, because you have to hit tom pads exactly in the center to get accurate response.

I've set XTalk Level to 0 and put all pads in different XTalk Groups. Cymbals are isolated on separate stands. Toms are on a drum rack, but even though they are close to each other, they are attached to different posts of the rack.

I've recorded it with Alt Sampling Alg turned on (first part) and off (second part). In the middle you can watch and hear MegaDrum auto-triggered by noise and notice big differences in intervals between noise-generated notes (I would expect them to be random, but with values close to each other).

This is my configuration for first:

http://textuploader.com/du4v4

and second part:

http://textuploader.com/du4vq

I think something is wrong, but have no idea for further search.

[dmitri]

I watched the video and I'm either don't understand what the problem is or I just don't see any problem. Are you saying that you hit two pads simultaneously (apparently) but the hits in the MIDI Log are slightly apart?

[airflamesred]

You'll need to post a screenshot of your settings ( from MDM) which you can attach directly here. I, too, am not sure what you are asking.

[jarosz]

I watched the video and I'm either don't understand what the problem is or I just don't see any problem. Are you saying that you hit two pads simultaneously (apparently) but the hits in the MIDI Log are slightly apart?

Yes, IMO time interval between hits in the MIDI Log is bigger than in reality. That's what I tried to prove with the video. Furthermore, it only applies to the second stroke - third and fourth are only few miliseconds apart. And it happens even with Xtalk set to 0 with pads on separate stands (cymbals). That's why I included the recording of pads auto-triggering - once in a while the interval is significantly greater then few miliseconds - I thought it might be a clue.

You'll need to post a screenshot of your settings ( from MDM) which you can attach directly here. I, too, am not sure what you are asking.

My settings (in text format) are linked at the end of my previous post, above the picture.

[dmitri]

Are you absolutely sure you can consistently hit two pads simultaneously within less then 10ms?

[jarosz]

That's a good question which for now I can answer only: not absolutely, because I have no exact measurement. Room reverb in the video interferes with pad hits and makes it harder to distinguish by ear. In the afternoon I will set up a mic in front of each pad, measure the difference on the recorded waveforms and compare to the interval between generated MIDI notes.

[dmitri]

Also, can you scroll in the LCD menu to some Curves/Pedal/Pads settings so that the live hit bars are not shown and test it again?

[dmitri]

A few things to note.
1. I have not optimized MDM FX for performance to display MIDI Log live data so it will struggle (and stumble/choke) with a flood of MIDI messages trying to update MIDI Log. You can see it in a Task Manager - when I send a flood of MIDI messages to MDM FX, Task Manager shows it's CPU usage goes above 100%.
2. Java is not really a real time system so you should not expect it so calculate time between events consistently and precisely.
3. I've also tested sending MIDI flood (set Retrigger to 1, DynTime/DynLevel to 0, Gain to 8 and Threshold to 10) from MegaDrum to MIDI Monitor (native Mac application) and there were no stumbling/choking receiving MIDI data from MegaDrum with all events evenly and consistently spread within 1-2ms although it also used a lot of CPU in this case and it is also possibly not a real time application.