Analogue-sequencer

Folks,

Thanks to a MIDI guitar (ab)using beta tester, we have a second build 
of the day...
Beta 26 is in files now, and doesn't leave record thru notes hanging 
if you manage to send the same note-on twice in a row without sending 
a note-off for the first. Something that's physically impossible to do 
on a keyboard, but dead easy for a guitarist.
Yet another reason why technology and plank-spankers shouldn't mix ;-)

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

> Beta 26 is in files now, and doesn't leave record thru notes hanging 
> if you manage to send the same note-on twice in a row without sending 
> a note-off for the first. Something that's physically impossible to 
>do on a keyboard, but dead easy for a guitarist.
> Yet another reason why technology and plank-spankers shouldn't mix ;-)

Thanks for sorting that one out Colin, It all works nice 'n sweetly 
now.  :-)

I could be churlish and complain that you've ruined a perfectly wicked 
feature - one that allows midi guitarists to hold ('hang') notes at 
will by playing the same note twice and then release ('unhang') the 
note by playing it a third time - but I won't as I can simply go back 
to doing it the proper way, with a footswitch assigned to hold.  ;-)

Regards, 
The Twisted Plank Spanker ;-)

hi Colin, 
i'm on beta 25 at the moment, and I've just loaded in all the material
for a liveset i'm doing tomorrow. I was trying some stuff out, and
there's something very wrong with the "randomize Aux[x]" event; I have
it set to a value of 1, but it always outputs a value of 127.
I have Aux C set up as "Randomize Aux D" and Aux D set up as CC#12,
which changes the waveform of a sound. Auxes A & B are set to do some
masking stuff that has no influence on Auxes C & D. I've just checked
the output with midiOX, and as long as I have no active steps in Aux C
or if they're active but set to 0, the CC value that is output is as I
programmed them on Aux D, but as soon as the value of Aux C is set to
anything higher than 0 I'm getting a value of 127 everytime, no matter
what it's set to. I've also checked the Randomize note & velocity
events, and they appear to be working fine; as far as I can tell it's
only the 3 Randomize Aux[X] events.
any chance of a quick fix?

grtz Chiel

Chiel, 

> i'm on beta 25 at the moment, and I've just loaded in all the material
> for a liveset i'm doing tomorrow. I was trying some stuff out, and
> there's something very wrong with the "randomize Aux[x]" event; I have
> it set to a value of 1, but it always outputs a value of 127.

Well spotted...
That one crept in on the 7th of last month, when I 'fixed' an existing bug
with the 'rndmz aux n' events overflowing.
It's definitely fixed in beta 27, which you'll find in Yahoo files now.

The previous fault with these events was that if the total result of the aux
value plus the random addition exceeded 127, the aux was deactivated. This
had been in there since those events first appeared... funny no-one noticed,
though I guess the deactivation would probably be quite intermittent, so
you'd need to pay close attention.

The fixed behaviour is that a result exceeding 127 is clipped to 127.
The bug you saw was because of an extra ampersand in the top bit test.

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

--- In analogue-sequencer@yahoogroups.com, "Colin f" <colin@c...> wrote:

> It's definitely fixed in beta 27, which you'll find in Yahoo files 
> now.
> The previous fault with these events was that if the total result of 
> the aux value plus the random addition exceeded 127, the aux was 
> deactivated. This had been in there since those events first 
> appeared... funny no-one noticed, though I guess the deactivation 
> would probably be quite intermittent, so you'd need to pay close 
> attention. 

cool! I just gave it a go and it works as expected again. And I can
imagine it'd be pretty difficult to find an error in a random
behaviour, as that's inherently pretty unpredictable ;-)
thanks again for the fastest support ever!!

grtz Chiel

On Friday 04 November 2005 10:43, Colin f wrote:

> The bug you saw was because of an extra ampersand in the
> top bit test.

Classic typo. :)

- Robert