Practice Valves

Discussion in 'The Rehearsal Room' started by Queeg2000, Feb 20, 2018.

  1. Pauli Walnuts

    Pauli Walnuts Moderator Staff Member

    000 would also be a rest and without the pitch, how would you know for example that it's a bottom G and not a middle one? Without tempo you would need a set of messages for rest durations too. Then there is alternative fingering to consider.
    I suspect it is doable but the user interface would have to me more complex than the one we have today ie. 3 things to press (or not in the case of a rest!).
    I work for a major IOT/Data solutions company and this is beggining to sound like a great exercise for one their regular "hackathons" so will run it past them.
    One of the most logical approaches would need the music you are testing against being digitised and that points to something like an addon to software like Sibelius or one of the cheaper options. Our guys have been very clever in the past using things like Raspberry Pi but they have the advantage of some of the best data analytics tools sitting behind them and I suspect any commercial solution would be both pricey and a legal minefield if you accept that pitch and tempo are an integral part of validating the input.
    We have a number of software engineers/programmers on tMp and would be good to hear their thoughts.
    Also, maybe @philipsparke can share some views on copyright issues.
  2. Queeg2000

    Queeg2000 Member

    I've don a lot of programming, though if converting a document from another company is going to be required, then you are getting into a legal minefield as this would involve reverse engineering existing software to decrypt their scores. Musescore, being open source would probably be OK with this. Musescore can also convert a PDF file (though accuracy is hit and miss) I would be very sceptical about getting permission from Finale or Sibelius to reverse engineer their software for the purpose.

    As regards tempo and pitch, well my thinking was that this would be a silent exercise, so pitch wouldn't come into it for the purpose of this as it would be intended for exercising the fingers and promoting muscle memory. This however brings in the possibility that you could memorise the part incorrectly if you repeatedly practice the wrong fingering, with it being silent, then the inaccuracy would easily be missed.

    Tempo is not really an issue that I had given much thought. For terms of practicing the fingering, it's often preferable to ignore tempo and learn the part slowly in any case.

    There are several options for entering the notes. I would favour connecting to a PC for this where the notes could be entered (or possibly imported from an external file) with a GUI in musical notation. This could be uploaded to the device either as a string of 1s and 0s for each valve, although this doesn't allow for alternate fingerings (alternate fingerings could be added as a secondary signal or stored on the device for comparison) alternatively, the notes could be sent as A-G strings with a b or # following the letter. With respect to differing fingerings for example Eb in the E space or Eb on the E line, the top E could be represented as upper case and the lower E could be represented in lower case. The user wouldn't need to be aware of this as the software would take care of it. That said, the actual data instruction doesn't have to bear any semblance of logic at all. Provided the PC sends uses the same instruction set as the device, then it would work fine. If required, a rest could also be added in the same way. However, given that during a rest, any fingering could be held without being a mistake, this seems unnecessary.

Share This Page