Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
MIDI woes (part 1)
#1
I am having persistent problems getting my NDLR to work consistently at a basic level. I am using it with an iPad and I have successfully routed the four modules to various synth apps. I have done this using port 1 and channels 1 to 4, although I have also tried using individual ports (1-4) with all channels (omni) selected, or even port 1/channel1, port2/channel2, etc. I don't really understand what the ports are, but in any case it makes no obvious difference to the following behaviours.

I am persistently getting a lot of stuck notes. Now I don't yet know whether this is down to the NDLR or something on the iPad setup, but in exploring it I have zeroed in on another, perhaps related issue I have noticed from day1. That is that I sometimes press a chord change button but even though the NDLR screen changes to show the new chord, the chord doesn't change. In fact it's worse than this, because sometimes the chord changes for one part (e.g. motif1) but not another or others. I have captured this with some MIDI monitor screen shots.

Here's what happens when I press the Pad button to start. The screen shows the chord as Ic in white and it plays a C major chord. So far , so good. But now I press the IV chord button. The NDLR screen changes to show the chord as 'IV F' in white, but the sound doesn't change. Here's the monitor - the Pad part is the leftmost column and the MIDI messages are shown with the most recent at the top:

   

Now I press Pad again to stop, but it looks like only one OFF action is sent. Here's the monitor:

   

This leaves some notes stuck on. I can clear them with shift+panic. Now I press the I chord button and Pad again, but I get an unexpected sequence of OFF/ON actions (I have cleared the log screen so only the new messages show):

   

And finally, here is the aftermath of the same problem but with the Motif1 part also running (column 2). You can see that Motif1 has been transposed to play an F chord, but the Pad has not:

   

Does anyone else see this?
MarkH

[Image: gGzCVFp.jpg]
Reply
#2
I've been having this problem for a while too. Just to check I captured the output of the NDLR (below) using MIDI-OX with the latest firmware (1.1.062). The NDLR was set to C Major, with I C selected. The pad was set to output to ALL ports on MIDI channel 14.

I pressed PLAY for the pad, then changed from I C to IV F, then stopped the pad. You can see that, for the change, the NDLR didn't output any new notes. However, when I stopped the pad it sent MIDI note off messages for the notes in the IV F chord.

There's definitely something amiss...

Ian

Code:
TIMESTAMP IN PORT STATUS DATA1 DATA2 CHAN NOTE EVENT             
00006C51  18  --    FA    --    --  --  ---  Start               
00006C51  18  --    9D    24    7F  14  C  2 Note On             
00006C51  18  --    9D    28    7F  14  E  2 Note On             
00006C51  18  --    9D    2B    7F  14  G  2 Note On             
00006C51  18  --    9D    30    7F  14  C  3 Note On             
00006C51  18  --    9D    34    7F  14  E  3 Note On             
00006C51  18  --    9D    37    7F  14  G  3 Note On             
00009FA1  18  --    FC    --    --  --  ---  Stop                 
00009FB0  18  --    8D    24    00  14  C  2 Note Off             
00009FB0  18  --    8D    29    00  14  F  2 Note Off             
00009FB0  18  --    8D    2D    00  14  A  2 Note Off             
00009FB0  18  --    8D    30    00  14  C  3 Note Off             
00009FB0  18  --    8D    35    00  14  F  3 Note Off             
00009FB0  18  --    8D    39    00  14  A  3 Note Off
Reply
#3
(12-07-2019, 04:37 AM)drGrov Wrote: I've been having this problem for a while too.

Thanks Ian, good to know it's not just me and, if I read you correctly, not the iPad.
MarkH

[Image: gGzCVFp.jpg]
Reply
#4
(12-07-2019, 07:16 AM)Almost Aviation Wrote:
(12-07-2019, 04:37 AM)drGrov Wrote: I've been having this problem for a while too.

Thanks Ian, good to know it's not just me and, if I read you correctly, not the iPad.

Yes - sorry, I guess I should have said Smile. The NDLR was connected to my Windows PC via USB for the above. However, I get the same problems with external synths connected via DIN MIDI...

Ian
Reply
#5
(12-07-2019, 09:55 AM)drGrov Wrote: Yes - sorry, I guess I should have said Smile. The NDLR was connected to my Windows PC via USB for the above. However, I get the same problems with external synths connected via DIN MIDI...

Yes, I guessed after I googled MIDI-OX Smile  And I guess this rules out USB as the culprit too!
MarkH

[Image: gGzCVFp.jpg]
Reply
#6
Okay, I am going to add a bit more to this. I have noticed something specific that helps (which I'll get to below), so maybe someone can understand what's going on.

Although I was initially assuming the NDLR is doing something wrong and then getting out of sync, I have noticed that the problem persists even after rebooting it (including removing the USB cable for half a minute). When I reboot it, it still seems to out of sync and sends baffling combinations of MIDI ON/OFF messages when I start/stop or change chords. I am interpreting this to mean there is something downstream of the NDLR that's actually going wrong, which in my case means something in my iPad setup. Now I do have quite a complicated setup hosted in Audiobus, a part of which is shown here:

[Image: wtqnLKh.jpg]

You will see that this includes connections for each NDLR channel to a synth app, but also in each case to another parallel destination, which is actually another iPad connected over Bluetooth - I have highlighted this for channel 4, which is the NDLR Pad. (This is so I can also drive other synths on the remote iPad.) To cut a long story short, if I remove the Bluetooth connection for the Pad channel, or if I create a simpler setup without any Bluetooth channels, it seems to work properly.

I don't know why this is, but I guess there must be a technical reason that someone can make sense of. I am a beginner at MIDI but I have understood it is a non-handshaking (fire and forget) protocol, so I am guessing this is actually going to be something about some other protocol that's involved with how the iPads implement MIDI over Bluetooth.
I'm not sure if this explains why you are seeing it Ian, but perhaps this will ring some bells for you.
MarkH

[Image: gGzCVFp.jpg]
Reply
#7
Interesting. I must admit I've never really used Audiobus on the iPad so can't really help with your particular problem.

My normal MIDI setup isn't particularly complex. The NDLR is connected via DIN MIDI to a MIO 10, which then forwards the relevant messages back out to the hardware synths  connected on other DIN/USB ports. I'm fairly confident there's nothing odd going on with the setup - I've not noticed any MIDI strangeness with any of the other devices I have connected.

The test I outlined above was a direct connection between the NDLR and MIDI-OX - no other synths or apps in the way. To double (triple?) check I just downloaded MIDI Scope for the iPad and ran the same test (again, the NDLR is only talking to MIDI Scope - nothing else in the loop). I got the same results so I reckon there's a bug with the NDLR pad output.

Ian
Reply
#8
(12-08-2019, 11:20 AM)drGrov Wrote: I got the same results so I reckon there's a bug with the NDLR pad output.

You may be right. I have continued to play and it is notable that it is only the pad that ever fails. I'm now not sure I was connecting with Bluetooth. Although I have been playing with BT connections it appears those destinations in my screenshot were connecting over the iPad's 'Network Session 1', which is WiFi. I suspected Audiobus so I'm now trying AUM, but I am getting similar results from time to time, again just with the Pad.

I will keep fiddling and share any updates if anything becomes clearer. Hopefully someone at CL will join in at some point.
MarkH

[Image: gGzCVFp.jpg]
Reply
#9
Thanks for the report, we'll take a look at it this week.
Reply
#10
I can confirm that I have observed the problem, whereby pressing a chord change button sometimes doesn't seem to be reflected in the MIDI data sent to the Deluge. I'm also persistently getting a lot of stuck notes. So don't think this issue is confined to the IPad.
Reply


Forum Jump:


Users browsing this thread:
3 Guest(s)