Hi Yves, Thank you for the feedback and for your kind words. Storing Chords per Bank is coming with next firmware update, V9... then, each of the 16 banks will have it's own set of chords and the 12 patches on a bank will share chords. This is ideal for creating songs with 12 patch variations but all sharing scale. The Chord Chains take up considerable memory (that is limited by ARP's hardware) so it is just 12 chains for now. I originally developed Chord Chains to give my ears a break when editing parameters, but it evolved into a pretty powerful little sequencer. ARP was envisaged as a way to play and compose on the fly and not a device to create and store whole songs. For rapid composition, not a DAW in a box. It's so easy to change and customise chords that you can get back to a point pretty quickly. There are many more chords than the 42 in the list. Hold the Chord button and toggle the notes to create custom chords. I will take on board your ideas and suggestions and see what I can do. Keep the feedback coming, it is very much appreciated.

Tried again but still no joy. I've sent more information via email.

For a cadence (short 2 or 3 chord sequence), it's probably best to quick-load a chord chain. There's no way of automating the sequencing of chord chains currently. Of course you could use external gear to send chords to ARP. Keeping it internal... an alternative might be to change the length of a chord chain on the fly, but I don't think this will be easy as, when editing chord chains, only the selected chord chain is played. Perhaps this could be a macro target in the future? That could then be assigned to MIDI CC. Or perhaps a dedicated control for Chord Chain start position and length, that could be adjusted, selected or invoked, independently of the editor? Just speculating. :)

So, I got hold of the H4MIDI Hub again. So, I've tested the following Sketch: - #include MIDI_CREATE_INSTANCE(HardwareSerial, Serial1, MIDI1); void setup() { Serial.begin(115200); MIDI1.begin(MIDI_CHANNEL_OMNI); } void loop() { delay(2000); Serial.println("Send Note On"); usbMIDI.send(0x90, 66, 66, 1, 0); // USB = F#4 MIDI1.send(0x90, 48, 48, 1); // DIN = C3 delay(2000); Serial.println("Send Note Off"); usbMIDI.send(0x80, 66, 0, 1, 0); MIDI1.send(0x80, 48, 0, 1); } The script sends MIDI ON and OFF messages ever 2 seconds via USB and MIDI DIN When running this directly connected to the PC (MIDI DIN via a Focusrite 8i6), both USB and MIDI messages are received. TIMESTAMP IN PORT STATUS DATA1 DATA2 CHAN NOTE EVENT 3967 6 -- 144 66 66 1 F# 4 Note On 3968 1 -- 144 48 48 1 C 3 Note On 5967 6 -- 128 66 0 1 F# 4 Note Off 5968 1 -- 128 48 0 1 C 3 Note Off 7967 6 -- 144 66 66 1 F# 4 Note On 7968 1 -- 144 48 48 1 C 3 Note On 9967 6 -- 128 66 0 1 F# 4 Note Off 9968 1 -- 128 48 0 1 C 3 Note Off However, when connected via the CME Hub, only the MIDI DIN messages are received. TIMESTAMP IN PORT STATUS DATA1 DATA2 CHAN NOTE EVENT 57520 2 -- 144 48 48 1 C 3 Note On 59521 2 -- 128 48 0 1 C 3 Note Off 61521 2 -- 144 48 48 1 C 3 Note On 63521 2 -- 128 48 0 1 C 3 Note Off There's no name.c file. Any suggestions would be greatly appreciated. :)

When I get chance, I'll create a stripped down code package with just the necessary MIDI code. I'll send you that and a built version too, so that we can compare the Arduino built and the Visual Micro built versions. I'll try to get hold if the Hub again, so I don't leave it to you to do. :)

@Submo, Cheers for taking the time to look at this and for sending the video. I've compared the interface_3x3 code and ARP's firmware and here are the differences I can see... 1) ARP has a name.c file with the following data format #define MANUFACTURER_NAME_LEN 8 #define MIDI_NAME_LEN 16 #define SERIAL_NUM_LEN 12 #define MIDI_PORT1_NAME_LEN 12 2) The MIDI Thru logic isn't handled inline, within MIDI.read(), ARP uses the setHandleNoteOn() delegate. But this equates to the same thing internally, I believe. 3) ARP uses MIDI.sendNoteOn() etc. vs the sketch which uses MIDI.send(). But it calls send() under the hood anyway and I've compared the MIDI output, it's the same. 4) For USB Host, ARP uses USBHost_t36.h, plus MIDIDevice_BigBuffer hostMIDI(usbHost); but I don't think that's relevant because your example uses MIDI DIN and it's ARP's internal notes that aren't being detected by the CME Hub. 5) Rather than being handled in the main loop() method, ARP uses an interupt loop that takes priority over the main loop, but at a lower priority than the hardware interupts. This loop run at a really high rate and it is this that calls MIDI.read(); My thoughts... the only thing I can see being relevant is the name.c file and the CME HUB detecting the device at all. Could you try adding a name.c file to the sketch and rerunning your test? (I would but I only had the hub here for a couple of days.)

Hmm, that is interesting because when I did that the H4 didn't pick anything up. Odd. Can you share the script and was it loaded via Arduino IDE or direct using Teensy Loader?

Cheers for the info, you know way more about this topic than I do, I've relied on the amazing work done by PJRC to handle the USB/MIDI. Any info you can get from CME would be very useful.

So, I've spent some time with the H4MIDI WC and I cannot get it working. I tested everything I could find in the studio that outputted MIDI over USB. The Arturia MicroLab, KeyStep and Beatstep, Roland J6 and the Novation LaunchControl all worked ok for Notes and CC. However, the hub couldn't pick up any MIDI data from a Launchpad, LaunchKey, NDLR, BCF2000, MC-101, TR8 or SeqTrak. I tried @submo's suggestion of disabling Serial Output from ARP and that did nothing. Consecutively... MIDI Only build, stripping out all the Serial Outs, suppressing Serial.begin, even trying a simple Arduino UNO script with nothing but the MIDI Library added. No joy, I'm afraid. ARP's USB over MIDI works fine with other Hubs (such as Blokas or MRCC) and it connects to MPCs and other Arduino based gear via USB without any issues. I'm not sure what else can be done but I'll keep a look out for any info online. My suggestion would be to continue with MIDI DIN or upgrade the hub. Sorry it's not better news.

Performance Lock is coming with V9. You'll be able to prevent any of your parameter tweaks being saved and your patches will reset when you restart ARP. I can get you an early look if you want to join the beta test for V9? Well done on finding the scenario that causes the issue. I'd not been able to replicate but I can dig into it now and I'll let you know what I find. Thanks

Cheers for the extra info, The CME hub should arrive soon and I will dig into it.

Thanks for the info, I'll try to get hold of one of those hubs and see what's I can find. I have seen people using various MIDI with ARP but only with DIN. I'm wondering if its a power issue or if the USB messages aren't 100% compliant. I'll probably have to dig out an oscilloscope. :)

Can you also confirm which Hub device you are using? H2, H4 or H6? I don't think it matters but the H4 and H6 are USB-C ports.

If you connect ARP to a PC or Mac via the USB, I assume it works OK?

Check if the MIDI Output Channel is set to a value. Go to SET Mode > Menu > MIDI Output Channel (Note Sync) and set it to OFF. This setting is only useful if you are Daisy Chaining multiple ARP's or want to output the global chord, in order to harmonic-sync other devices. (My apologies for the delay in responding to this post, I don't know how I missed it, sorry}.

That is odd. I'll try to recreate the setup and see if I can replicate it.

I've done a little more work on the "Live Lock" idea and I think it's achievable. I'm working on adding it to the next update. Regarding MIDI Chain settings on Patch Reset... Resetting track or patches overwrites the track's MIDI Out Channel so that may need to be reconfigured. Unless you do a factory reset, all global settings are unaffected. Regarding the stutters with Chord Chain... * Is it just a stutter or does ARP go out of sync with other gear? * Is there a specific scenario that causes it? * Are any of the tracks set to Chord, Pad or Drone mode? These * odes retrigger on chord change, I wonder if it could be related. * Can you send me a video example of it happening? Cheers

Here's a great little Jam from @BayMud with the Fred's Lab Töörö. https://www.youtube.com/watch?v=VB-ByMLqmYE

Cheers for the suggestion. I can certainly look into it. Up/Down already have a hold function assigned, but perhaps this could be overridden for this specific case. Cheers

Thanks for taking the time to capture that info. I'll take a look at the data you sent and update you in due course.