tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 19:24:28 GMT
Xequence. And some M4A, the problem in the middle comes from the Arpeggio, i think. Attachments:
cubasis - tja00 1.m4a (443.89 KB)
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 20:26:26 GMT
Hi tja , OK, I could not resist and have reworked things. Attached is a Scene with v3 of tja autobass This one works correctly with manual playing, my test notes and I also piped in the ChordFlow events that you captured in GigRig. So far, so good. For me this is doing it well with no artefacts and no stray note offs. See how you go with this. Regards, Nic. Attachments:Scene-autobass.mfr (2.95 KB)
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 20:38:19 GMT
Last track by "autobass" from track 1 Attachments:
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 20:39:13 GMT
Oh, another update, did not notice - will check immediately :-) :-) :-)
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 20:46:39 GMT
Code for version 3 for those interested in my rework
# tja autobass v3 by audeonic IF LOAD # K0 = bass channel (0-F) # K1 = latency/min. note length ASS K0 = 3 1
MAT I0 = K0 + 1 SET LB0 I0 +D
# L80 currently sounding # bass note (80 == none) ASS L80 = 80 SET LB1 S--
# L0-7F table of all source notes # sounding (initialise to 0) ASS I0 = 0 IF I0 < 80 +L ASS LI0 = 0 MAT I0 = I0 + 1 END END
# handle source notes 9X XX 00 = 8X IF MT < A0 # set L entry, assume note off ASS LM1 = 0 IF MT == 90 ASS LM1 = M2 END
# wrap bass into sysex and inject+delay MAT I0 = MT + K0 SND F0 7D I0 M1 M2 F7 +I +DK1 END
# handle injected bass notes IF M0 == F0 7D # copy type into I0 for compare MAT I0 = M2 - K0
# is it a note on? IF I0 == 90 # ensure source is still sounding IF LM3 > 0 # is this note lower than current bass? IF M3 < L80
# if current bass is sounding, turn off IF L80 < 80 MAT I1 = 80 + K0 SND I1 L80 00 END
# sound new bass+mark SND M2 M3 M4 ASS L80 = M3 SET LB1 L80 +N END END END
# is it a note off? IF I0 == 80 # is this our current bass? IF M3 == L80 # turn off current bass+mark MAT I1 = 80 + K0 SND I1 L80 00 ASS L80 = 80 SET LB1 S—
# do we have any higher sounding notes? MAT I2 = M3 + 1 IF I2 < 80 +L IF LI2 > 0 # found higher sounding note, turn on # new bass note (via inject) MAT I1 = 90 + K0 SND I1 I2 LI2
# mark new bass note ASS L80 = I2 SET LB1 L80 +N ASS I2 = 80 # terminate loop early END MAT I2 = I2 + 1 END END
END
# block the sysex XX = XX +B END
# delay (most) source events to match # bass note IF ML == 3 SND M0 M1 M2 +DK1 XX = XX +B END
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 20:46:42 GMT
With Xequence, this looks good. With ChordFlow, I still get artifacts.
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 20:47:37 GMT
Any check with MIDI Monitor or GigRig?
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 20:49:44 GMT
Try changing the line near the top as an experiment:
ASS K0 = 3 1
to ASS K0 = 3 A
(this is 10ms minimum note length)
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 20:51:48 GMT
And if you could record that chordflow sequence into GigRig for me I will try that here
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 21:04:12 GMT
Going to try your change and record into GigRig. Meanwhile, I tested MidiSteps, which also produces artifacts: Attachments:
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 21:07:25 GMT
Just changing 3 1 to 3 A did not change nothing.
I used "Install rules" after that change
Now using GigRig
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 21:10:03 GMT
Oooh, I think I understand what is happening. Let me think about this. (again)
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 21:11:15 GMT
|
|
nic
Soapbox Supremo
Troublemaker
Press any key to continue
Posts: 2,011
|
Post by nic on Aug 16, 2018 21:49:02 GMT
When all else fails, try, try again! Here is v4 - another refinement Regards, Nic. Attachments:Scene-autobass.mfr (3.03 KB)
|
|
tja
Converser
Posts: 44
|
Post by tja on Aug 16, 2018 22:10:56 GMT
I am very sorry. First thing, I don't trust ChordFlow anymore, as I often even got problems recording the original chords from this App. I switched to MIDIsteps which works much better. But still:
|
|