I have some questions and observations for you:
With the Pyramid connected to the Triton with MIDI out to the MIDI input of the Triton
The MIDI output setting on the Pyramid set to transmit clock and midi thru on BUS A
With a blank project (nothing loaded upon startup) are you able to select track 1, and then in live mode play the sound one Midi CH 1 of the Triton?
Is this also true for Track 2, 3, 4 etc?
Of course you can choose whichever channels you like, and put them on any tracks you like, but for this experiment don’t mess around with a complicated setup, keep it simple.
If you are able to select each track (and subsequent default MIDI channel i.e. 1,2,3,4) and play the sound loaded up on each channel then the question is what happens when you press play on the Pyramid?
Does the sequence in the Triton begin to play?
Are the Pyramid and the Triton in in sync?
Take a look at the Triton’s settings.
Is clock input enabled?
Is MIDI reception enabled?
What you want to accomplish here for testing purposes, you want to make sure that the Pyramid is set up to control the Triton, and that the Triton is set up to receive the control input.
Assuming that this basic arrangement can be established then there is hope.
I would not connect the in and the out between the Triton and the Pyramid.
In order to work this way I would disable the clock within the Triton and perhaps turn local off, and do all of your sequencing with the Pyramid.
To use both the Pyramid and the Triton as sequencers together then I would work with the Trition, and then sync it with the Pyramid and thus only be sending a clock to start and stop and sync the sequencing within the Korg.
It is, I assume, possible to sequence something on the Pyramid (tracks 1, and 3) leaving the tracks on the Triton blank, and then have something on the Triton also playing and in sync by sequencing this on the Triton, and clocking that sequence with the Pyramid.
But to connect them together will create a MIDI loop which will definitely double notes, crash the sequencer, and make all kinds of trouble for you.
This is frustrating because it’s nice to use the Triton as the input keyboard for Pyramid.
You can do this, but in order to do it without hassles it’s a matter of connecting the Triton’s MIDI out to the MIDI in of the Pyramid and do not connect the Pyramid’s MIDI out to the input of the Triton.
This is fine for recording and establishing sequences, but it won’t work for a live setup.
How to get these working together is a complex mystery of the MIDI ages… I want to do something that will double clock my rig, but I don’t want to double clock my rig, and I still want to do the thing!
I have worked around this by using a Keystep or the pads on the Pyramid for input, and slaved my remote devices and sequenced on them independently. The one exception is the Octatrack. I need to access it in many ways:
- Pyramid sends clock to the Octatrack
- Pyramid sends midi note data to the Octatrack
- Pyramid sends midi CC data to the Octatrack
- A Korg nanoKontrol is merged with the Pyramid midi output to also send midi CC data to the Octatrack
- Midi output of the Octatrack is used to clock other devices
- Midi note data comes out of the Octatrack and is sent to other devices.
Keeping it all clear, and unlooped like this makes for a lot of gear, but it always works and never crashes.
If I need to make a recording into the Pyramid form another device, I set it up specially (avoiding any loops) and make the recording, then stop and reconnect in playback mode to test the recording and to play it back.
It’s a lot of plugging and unplugging sometimes, it’s true, but there can only be one master, and one central brain in a midi rig under most circumstances. It IS possible with switchers and filters to get things working differently, but I recommend getting a workflow going that makes you happy, and allows you to make the best use of both the Pyramid and the Triton.