Future of this project, suggestions and general discussion

[privatex]

What do you think does megadrum can ever reach Roland Super Natural and Behavior Modeling technology? I'm very impressed with modules that have this tech. Precise rimshots with ps, cymbal crescendos...all that tiny differences this modules can differentiate and considering to the playing situation initiate appropriate algorithm. Essentially difference between Roland best modules and MD device is that extra processing power that can do this things (ok Roland have default sounds in modules with extra editing and tuning solutions-honestly I don't need this with VST).
My first thought was software processing solution. We have trigger to midi device, VST an software would be logical step.
Reading old posts I have clear picture of MD device from beginning and present MD. Differences are huge and megadrum have constant improvements thanks to our host Dmitry.
What could be the next giant leap?

[airflamesred]

I'm quite happy with my MD it does what it's supposed to. All that stuff with Roland is just smoke and mirrors to cover up for their genetically modified sounds.
I would swap to FSR though if it were viable.

[privatex]

Dont get me wrong I love my megadrum too, but I cant agreed with you that it's just marketing. I was in situation to play and feel on module with SN,BM tech and it is a game changer in some part IMO.
Yes some good quality FSR heads will solve many problems that cone pads have. My snare never trigger accurately like this one.
https://www.youtube.com/watch?v=pkNbnkTH4g8
No matter what cone I use there's always hot spot problem with imprecise (non perfect) signals. I know that huge part goes to the construction of pads, at the end it will be nice to have one of those modules but... MD do the job good enough.

[ignotus]

If you go through the Vdrums forum, lots of people complain about having a hotspot too with their td20's, 30's and whatnot. A good few factors come in to play there but it's something that can affect even the most expensive kits. On the other hand, as far as I can see, that Super Natural and Behavior Modeling technology is something that applies to the sound processing itself, not the MIDI data, which is what pertains to Megadrum. In the video you posted, the guy is playing a td30 and it sounds incredible, but he's triggering a VST - so in this scenario all the Super Natural and Behavior Modeling technology is moot, it's not being used. If you listen to a td30 playing its own sounds, for all the behaviour this and modelling that, it doesn't come close to a VST; the machine gunning is awful and the sounds just don't cut it. Cymbal swells are something that it does do well but it can't really be dealt with by MIDI - in our case the VST would have to take care of that, which by the way, I think is well overdue. MD can already do positional sensing so I reckon shallow rimshots (Addictive Drums has them, I assume others do too) may be a possibility. In any case, given that Megadrum is purely a trigger-to-MIDI device, in order to measure it against a high end Roland module, we would need to look at how they both process signals coming from the same pad and using the same VST. I think that's the only area where say, a td30 might do marginally better than a MD, but I must add that I don't even know if that's the case for sure - it might even be worse for a given pad. As for the next giant leap, with the new ARM chips I suppose Dmitri would have to say whether he has something in mind there, but like airflamesred said, my MD does everything I want and more. Any improvements are more than welcome, but I think stuff like cymbal swells, more natural roll sounds etc, have to come from the VST's we use. Any modern PC has infinitely more processing power than a Roland module so I think that with a TMI it makes sense to take advantage of that because the PC has to be in the equation regardless - so we might as well use it.

[airflamesred]

Yeah, MD measures the same things as other modules (voltage and PS) and I don't see as there is anything else to measure. As good as it appears, that vid is all about the VST.

[privatex]

have to quote myself: ''all that tiny differences this modules can differentiate and considering to the playing situation initiate appropriate algorithm'' so it is not only matter of sound,those are words of Roland's guy from one video.
I'll just ask you one question: Can you setup your MD and produce this kind of precise triggering,smoothness and feel?
You dont need special test for comparison, just look tons of videos of both modules and you will get a pretty precise picture.
Well next step has to be some kind of software processing in world of TMI, PCs and TMI have huge potential. TMI are reach that max line, theres nothing to do any more besides lowering latencies, all further big improvements will be in softwares.
For hotspot TD30 problem I wish I have that problem, TD30 machinegun rolls...hmmmm..
It was a mistake when I opened this topic, I knew that it will go on this side of river. I hope you undersund me here, if dont well, nevermind.
Again tune your MD and give me this triggering quality ignotus. If you reach it tell me a secret pls..

[privatex]

Yeah, MD measures the same things as other modules (voltage and PS) and I don't see as there is anything else to measure. As good as it appears, that vid is all about the VST.

It measure voltage it is ok but it know what type of playing situation is and it start special algorithm for that situation, it is advanced and it is not VST or what. That doing a TD30 module not superior drummer.
For eg. when you play regular single strokes on snare module do that with algorithm for singles, if you change to double strokes it has little changes in algorithm but if you start to move near to rim it start ps with maybe third algorithm that have best results for that playing situation. This is just eg in terms of explanation, but it have something very similar. It's logical and it can be done for shure, combination of time, position, velocity and articulation inserted in different algorithms for different playing situations, connected with sound engine produce this level from video.
I think that if all of those modules at market (that have TMI and sound engine as a part of them) work similar with VSTs according to their slightly different algorithms, than you are right but why I can reach that level of triggering for almost year with MD and all kind of pads, materials and tuning technics? Because they dont work only that way.

[dmitri]

privatex,

1. Disregarding the sounds from the TD30 video, which mostly depends on VSTs used, I honestly don't see what is it in this video which makes it so special in terms of pure triggering and how it is better than some videos made by MegaDrum users.
2. I had a TD20 for testing awhile back and I was comparing it against MegaDrum with Pintech 10" mesh snare, Roland PD-125 and Drumtek 14" among others pads and cymbals and they performed pretty much identically in terms of produced MIDI data, i.e sensitivity, velocity precision, rolls, false double triggering (or lack of it), hot spotting and etc.
3. I understand you have a TD30 to compare against MegaDrum with the same pad, right? Can you please make a video of both showing the difference between TD30 and MegaDrum. If it is due to problems with MegaDrum settings I will try to help to tune them. If it is due to problems with MegaDrum algorithms I will try my best to improve them. Usually I do.

And of course, any suggestions about MegaDrum improvements or new features are always welcome!

[ignotus]

It measure voltage it is ok but it know what type of playing situation is and it start special algorithm for that situation, it is advanced and it is not VST or what. That doing a TD30 module not superior drummer

Sorry but that is simply not true.

For eg. when you play regular single strokes on snare module do that with algorithm for singles, if you change to double strokes it has little changes in algorithm but if you start to move near to rim it start ps with maybe third algorithm that have best results for that playing situation

We're talking about midi here. How on earth do you suggest that these special algorithms are encoded into midi messages to be sent to the VST? It doesn't happen. You're basing your argument on a false premise: that Roland applies algorithms to improve performance in certain playing situations, and somehow gets the VST to understand and interpret these algorithms. But when dealing with midi, which is the "language" that VST's understand, this is neither realistic nor possible. All those artifacts to better simulate realistic playing take place at a sound processing level, not midi. That video you linked to shows a very skilled drummer. No, I can't replicate that playing on my snare but it's mainly because of my limited abilities. Can you do that on an acoustic snare?

[airflamesred]

There are plenty of alogarithms in Roland modules that are there to attempt to replicate the parameters that change as a drum is struck with differing velocities and in different areas.
For example, a drums pitch lowers the harder you hit it and there is a slight pitch change over the duration of the sound. I'm sure there are many other nuances as well. Someone, sometime will have investigated and written a paper on this and some boffin over at Roland will program an alogarithms to try and replicate these change to whatever base sound they use. You could do most of that yourself back in the 80s with a Simmons SDS7.
Personally, when I hit a drum I want to hear the wonderful rich sound of that sample. There is no substitute for quality.