[LAD] successive note on midi events

Ralf Mardorf ralf.mardorf at alice-dsl.net
Tue Apr 13 02:12:36 UTC 2010


Ralf Mardorf wrote:
> Jens M Andreasen wrote:
>> On Tue, 2010-04-13 at 00:55 +0100, James Morris wrote:
>>
>>  
>>> A string of note-ons following each other all for the same pitch n 
>>> without
>>> any intervening note-offs for pitch n, IS PERFECTLY ACCEPTABLE provided
>>> they are INTENTIONAL and NOT accidental.
>>>
>>>     
>>
>> Yes, except for that this is an absurdity that could only happen by
>> accidental programming.
>
> Isn't there the consensus that the c) thing will cause the same sound, 
> as the sound that will be produced, when the sustain pedal is used? 
> And that it's allowed by the MIDI specs to do this without using a 
> sustain pedal, but by using a sequencer were it's edited that the same 
> note will overlap? It might be useful to produce some random phasing 
> synth sound, but there's absolutely no natural instrument were this 
> could happen.
>
> By the way, I right now played a very good DX7 Mark III sound, hold 
> the sustain pedal and played the same notes again and again and the 
> sound became bad. I dunno, but I hope that for a Clavinova a Mark 
> sound even would be "reset" when the sustain pedal is hold and the 
> same notes will be played again and again.
>
> Anyway, "except for that this is an absurdity that could only happen 
> by accidental programming", it's not an absurdity, it's possible to do 
> this by using any sequencer, but most times it's unwanted, because it 
> seldom sound good and it never sounds natural. Some sequencers have 
> functions that allow to delete "same notes".

PS: Btw. I was mistaken because of the drum samples, even there a note 
off event won't cause the machine-gun effect, when the first note simply 
has it's release time instead of being interrupted, while the second 
note starts. I guess for good drum samples uncontrollable phasing is 
avoided by having layers, resp. fine random pitch shifting.



More information about the Linux-audio-dev mailing list