Technical Posting Guidelines


If you cannot post, please visit our Forum Talk forum for answers to more frequently asked questions.

Is your post missing? Please read forum descriptions & guidelines within before posting in the wrong forum.

See more
See less

Is positional sensing accurate?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Is positional sensing accurate?


    I am not a drummer, but am working on a project where I will write music for an electronic drummer, so I apologize for my lack of knowledge about this.

    I plan to write for electronic drums where the playing position from centre to rim changes the timbre of the sound, for instance controlling the filter frequency of an electronic drum sound. I tried to use positional sensing for this, but without luck. I used a Roland TD-50KVX including a PD-140DS snare, PD-128-BC pad, and a TD-50 Drum Module.

    I set up a software to track the positional sensing data and mapped this to the pitches of a synthesizer to measure how accurate it was.

    As far as I can understand, positional sensing was sometimes very inaccurate, both on the PD-140DS snare and PD-128-BC pad. Sometimes it seemed okay, but often it triggered errors, and it was pretty unpredictable, even reacting sometimes to how hard I hit the drum.

    Is it me who has done something wrong, or is positional sensing simply a bit inaccurate, and this is quite normal?

  • #2

    I posted this three weeks ago, but did not get any reply. Sorry for bumping this, but I’m really curious about this.

    It seems to me, at least from online videos, that the positional sensing from Sunhouse Sensory Percussion works much better than the one from Roland.

    Is positional sensing simply a bit inaccurate?


    • #3
      I use a td30 and a diamond snare. I would say it is not always accurate but I find it works well with SD3 which is what I use. I don’t know about the td50 but I would have thought it would have been at least as good if not better
      Roland TD30 module on TD20 kit SD3 with various kits. Pearl Masters Kit, Yamaha 9000RC original natural wood finish. Cymbals from Zildgian Pasite and Sabian. Loads of percussion bits. Cubase and Wavelab always current versions.


      • #4
        Megadrum and Diamond snare here; no I've never got it 100% throughout the range. It seems that Roland is no better than Megadrum at positional sensing which is surprising to me. I wrote about it once here:

        The working range is actually very small in terms of the calibration; meaning there's not enough usable CC16 data to make smooth sweeps, but there is enough to make a definitive cutoffs / zones whereby you can be sure at least 95% of the hits are either centre or edge with say 5% fuzz from mesh head chaos.

        There are also insurmountable flaws that will make it unreliable such as; hitting the hotspot directly, or hitting too hard causing distorted waveforms.
        ◾ Diamond Drums 4pc in Di-Noc carbon ◾ MegaDRUM
        ◾ Roland UA-1010 / cymbals / KT-10 (x2) ◾ Tama / Gibraltar hardware ◾ JBL LSR3 Series 2.1 Monitoring ◾ Pearl THMP-1
        PA Comparison Sheet


        • #5

          Thanks for the feedback! Good to know that I did not do anything wrong, and that positional sensing simply is a bit inaccurate. I guess it comes down to Rolands old waveform detection technique, which seems a bit too simple to work flawlessly.

          Have any of you looked into the positional sensing from Sunhouse Sensory Percussion?

          It seems to be based on machine learning and work much better than the one from Roland. There are some convincing videos on their page