Using OT as a midi controller for some drum patches on Blofeld and Pulse2 modules. The drum patches really only require a pulse/trigger for the envelope generators. I’ve found that when the midi note LEN is zero, it is ‘hit and miss’ (pardon the pun): sometimes it will trigger the synths properly, and sometimes it seems to ‘sort of’, the sound is notably different sometimes. This is while a pattern is playing, all the trigs on the OT have identical values, but the results seem random. Increasing the LEN fixes it, but I just was curious what causes this inconsistency?
Did you try with other gear?
Just the 2 modules. It does the same behaviour on both.
Same brand.
I’d check with a midi monitor if OT sends notes off (Note On with Vel=0) with LEN=0, and if its consistant.
Why would you need LEN=0?
1 Like
Thanks Sezare. It’s not that I really need the LEN to be zero, I was just curious about why it would be inconsistent.
1 Like
Have you checked with a midi monitor, yet?
IIRC I had some weird behaviours with note length 0 as well…I’ve never really looked into it, though. Gotta run some tests tomorrows.
1 Like
I haven’t yet- I’ll need to set it up first (I don’t have a computer near my music stuff). If I do get around to it I will share the results. MIDI-OX will do right?
Any midi monitor theoretically.
LEN 0 makes chords with arp on, I really appreciate that kind of inconsistency with a random lfo on LEN. 
1 Like
Hmm, strange. Midi monitor shows nothing weird (it’s just a note on with velocity - in this case velocity 100 - followed by velocity 0/note off, just like other note length appear on the midi monitor) the recorded midi note length just gets shorter as I decrease note length, but my Shruthi clearly reacts in a different way to note length 0.
I’ve actually used this behaviour in some of my tracks as a kind of performance feature lol^^
Not sure whats going on there though…
3 Likes
I guess the culprit is a timing issue. Actually length 0 doesn’t really make sense with a serial protocol like MIDI, because there is always a time interval between the corresponding ON and OFF messages.
Nevertheless: With note length 0 the ON/OFF events should logically be sent right behind each other with no delay or other MIDI messages in between to get as near as possible to zero length.
So either the sender doesn’t sent them always in exactly this way (no delay between ON and OFF) or the receiver doesn’t process them always exactly at the moment when they arrive (like when there is some kind of processing loop going on which also has other duties to fulfill than only MIDI input).
Normally very small deviations in timing which are intrinsic to the MIDI protocol doesn’t matter, but when approaching length 0 they will do, because there the deviations make up the whole event. So what you experience here is - simply put - MIDI jitter in combination with processing jitter.
7 Likes
Interesting thoughts and kinda fits what I hear coming from the Shruthi.
It sounds like it’s playing many short tied notes…
I didn’t think about it in terms of zero length, which ofc can’t happen and I think the notes actually didn’t get shorter after note length = 1, have to double check that laters again.
Also gotta check how other synths behave.
1 Like
I think @tnussb nailed it. So if LEN=1 does the trick, use that instead of 0?
Interesting to see the limitations of Midi nonetheless tho 
Thank you- Great explanation. I tend to forget that MIDI messages still ‘exist’ in time and are not abstract perfections. The reasons you’ve put forward make perfect sense. A trigger has a duration and when it is extremely short, the chances of it being both sent and received/processed perfectly every time in the real world are very small. And it also makes sense that different synths will have different variations in how they receive and process such short messages.
This is going to turn me into a complete audio nerd if I’m not careful.
I’m thinking now it’s more to do with the synth not the MIDI message itself.
After playing around with this more (OT as MIDI sequencer on Pulse 2 and Blofeld) I’m hearing slight variations regardless of MIDI note length (maybe it’s just more noticeable at zero, and maybe that’s just the context I happened to notice it).
The recording is of the Pulse 2 playing a patch I’m making following the SOS recipe for a 909 style drum (OK so I’m learning). Although absolutely no parameters are changing on the Pulse or the OT, and the trigs are identical, you can clearly hear the difference in the starting ‘click’ on different hits. For the last few hits I reduced LEN to 0 but was about 1/16 for the rest. Changing the LEN didn’t seem to change anything.
1 Like
With our beloved EPPlus, I’d try to filter Notes off (Notes on Vel 0 on OT) , then map them to regular Notes Off to check if notes off handling are the culprit.
Off topic, still with EPPlus, once I mapped midi notes to Prog Changes, to play a Blofeld drum patches bank. 128 sounds with 1 midi track!
To make it more punchy add a pitch sweep please. 
2 Likes
I’m trying! Can’t find the right balance between ‘punchy’ and ‘cheesy’ and ‘dull’. But you are of course right, I need to crank that pitch modulation amount right up.
It would make sense, that some synths behave differently, because incoming midi notes have to be processed by the synths cpu, no?
1 Like
As I said, I’d check removing “Notes OFF”, because OT sends Notes On with velocity 0.
May cause the click.
I’d check after with Notes On Vel 0 mapped to regular Notes OFF.
1 Like
I now think this is to do with the phase of the oscillators in the synths. As far as I know when you trigger these synths, you can’t set the starting point in the phase of the oscillator wave. So whenever you trigger it, it could be starting at any phase point. This occurred to me watching Robert Henke’s tutorial on kick drums (just after 3 minutes he discuses this point): https://www.youtube.com/watch?v=rfeY0_k1ctk
The envelopes you can specify to start from zero when you trigger, but not the oscillators. I think this is the case for both the Blofeld and the Pulse2.
So if this is right, my initial guess was completely wrong, it’s nothing to do with the midi.
3 Likes
If you use samples on the Blofeld, they can always start at the beginning.
One of the reasons why many of us sampled / sample from old analogue synths - to improve consistency in subsequent triggers.
5 Likes