Future of this project, suggestions and general discussion

[ignotus]

Interesting. Would you say that those piezos are less, more or as sensitive as regular disc piezos? Do they react to vibrations along their entire length? I'm tempted to buy some just to test a couple of ideas, they're dirt cheap on aliexpress.

[privatex]

They are slightly less sensitive than discs but nothing to worry about. Yes whole surface is sensitive and you can cut them to measure which you need. In fact 'in head aquarian system' use similar one for rim sensors (they are wider and commonly used for contrabass,cello...).

[privatex]

Well guys I re-built my snare pad with new mechanical parts, new piezos with patiently MD tuning... and I got way better results. I tried with different approach - pad are reconstructed in relation of MD needs. Much better isolation between head and rim senor, changed rim piezo disc for stick piezo, placed new foam cone allow MD to catch lower retrigger mask value which bring new dimension of pad sensitivity, clear flams, rimshots and pad triggering generally. I had to increase MinScan, DynLevel and DynTime but benefit of lowering retriger are huge. It is a very, very important variable and I finaly found a way to catch minimal possible value by changing mechanical characteristics of pad.
The only thing left to do is to set spongy bar for dampening as 2box pads have at the edges. Expected results: possible reducing of minscan,dynlevel and dyntime, without any possibilities for false or double triggering-perfect!
Only side effect is increased machine gun rolls effect. Do you have any clue how to reduce this but not at the expense of accomplished sensitivity? I suppose that I play there on edge, balancing between sharp, clean and fast signal and machine gun effect, so...
P.S. I'm not one of those guys that wont write this post after he asked few other questions earlier in this topic. Learning is a procces and sometimes we must go to secondary roads just to get one fragment of solution. Thank you for help and support guys, appreciate it.

[airflamesred]

As I understand it, The machine gun effect is to do with sample repetition. So this is delt with by round robins or some sort of sample massage alogarithm, depending on your sampler. I'm not quite sure if dyn level/time can help with this.

Good news with the piezo probe then? Can you post a pic showing it mounted?

[privatex]

Dys's can help but not if retrigger value is too low. However I'm satisfied with snare pad now, it works good but not as... :P
Seriously differences are huge than my previous setup.
If you low retrigger value you will have shorter time between two samples (+ note off lowered to the minimal value) which bring flams and rolls sampled better, but machinegun effect will be more expressive, so at the end you must find sweet spot between that two characteristic (flams are very good way to test this balance, if swiss triplet flam or just single flam are good, everithing else will be just all right beside muchinguning). Dampening will make life a litlle easier but it will not resolve this problem. Maybe some changes in algorithm can make improvement I'm not sure, it's not my shere, I only try to share my observations in this proces.
I can say now that I'm finaly broke the way of fine tuning this device, it have some limitation but hey this is electronic drummig machine and dmitry is open for suggestions. Lowering note off value resolved my positional sensing problem... From our discussion he can get an ideas for imrovements.
Piezo in stick is sealed with epoxy glue on part of hoop below contructive aluminium hoop of mesh head. I will post a pic this days, I have problems with photographic device (mobile camera):

Freehand:

[ignotus]

You're getting there

The only thing left to do is to set spongy bar for dampening as 2box pads have at the edges. Expected results: possible reducing of minscan,dynlevel and dyntime, without any possibilities for false or double triggering-perfect!

I have also considered doing this - I have some foam that would do the job but I need to think of a way to keep it in place, though I might just glue it/stick it with double-sided tape around the shell with it raised a bit over the edge.

Only side effect is increased machine gun rolls effect. Do you have any clue how to reduce this but not at the expense of accomplished sensitivity? I suppose that I play there on edge, balancing between sharp, clean and fast signal and machine gun effect, so...

As airflamesred said, I think this has more to do with the VST used. Machine-gunning is caused by the same sample being played repetitively - if MD is outputting the right velocity according to how you do your rolls, then that's its job done, the rest should be taken care of by the VST. The only way MD could counter that effect would be by "humanising" its response, i.e., varying midi velocity even when hits are exactly the same. I don't think this is desirable: we want it to send the precise velocity we hit it with.

Anyway, congratulations and post some pics when you get a chance!

[airflamesred]

Humanising. I agree Ignotus, that this should be dealt with by the VST. MD should be signal processing and routing.

Privatex, What are you using for VST?

Just a thought on PS technology.
Essentially it is measuring a physical distance between strike and piezo - yes?
Therefor can that info not be used as a crosstalk indicator? So the further you get from piezo (as in another drum away) no signal sent. Just speculating.

[privatex]

I agree, VST have power over samples, that's why we have need for assimilating different strokes and playing situations for different sounds (supernatural story...). By decreasing retrigger value, VST change samples more often so rolls must be better on output but... after increasing dyn's it comes all good. VST's samples layering is main causer (if layers are poor than strokes will be poor too especially rolls and fast triggering situations).
What bothers me is that I cant play snare headd and snare rim simultaneously even with no crosstalk rejection on snare pad. Power over this have MD and I suppose that have some level of crosstalk as default.
There are two infos that MD understand and processed: velocity and position I think. Main value for crosstalk suppression is taken from velocity information not positional. It's speculation too, I dont know for shure but it's logical.

[dmitri]

What bothers me is that I cant play snare headd and snare rim simultaneously even with no crosstalk rejection on snare pad. Power over this have MD and I suppose that have some level of crosstalk as default.

With a dual piezo/piezo pad you can play head and rim simultaneously. Check BNote, DualMidPoint and DualMidWidth in http://www.megadrum.info/content/pads-settings

[privatex]

You should make a note what are represent Bnote, Dual MidPoint and DualMidWidth in megadrum manager. All that parametres we have in MDM in menu for third zone and they cover rimshots in this case for snare. I dont think about third zone rimshots I mean playing on rim and head like on acoustic snare (I dont know how to explain). I think I know what are you suggest me - to set third rimshot zone with certain width and when I want to play rim and head (together, alternate between those two...) to aim outside that zone. That doesn't work. Rimshots are all right but head and rim can't be played together in this way.
Today I found that tom1 and tom3 have different response with same pad and same parametres in MD (similar as my ride's inputs but much worse behaving). Results are big false triggering. It's funny I plug pad in input for tom 1 everything is great,when I change and put in tom3 it start to give me 5-6 strokes per one real.