Problems with recording a poly sequence

[Razmo]

I've run into a problem...

I have my REV2 connected to my computer via MIDI I/O, and my sequencer is set to echo the REV2 MIDI data back to it... a very standard way of using MIDI gear with a sequencer... I also have the REV2's "Local Off" set to "off", so that I do not get doublenotes etc.

But with this configuration, I cannot record poly sequences... as soon as I press the "RECORD" button on the REV2, the REV2 sends out an NRPN command (that I do not know what is by the way). this is of course echoed by my sequencer, and when the REV2 recieves it again, it sends it out again, resulting en and endless MIDI loop.

So basically, I'm unable to record poly sequences in this setup... I've sent DSI a bug report...

[Razmo]

An additional note:

After I played with this a couple of times, my REV2 started to behave very weird... keys not responding to what I played on them... I then took out the MIDI cables, switched Local to On, but it still behaved weird... I reset the globals... no use... then I initialized a program with transpose/hold buttons... no use... then I powered down the REV2 and then on again... now I got the Basic Patch sounding, but with a weird decay on volume (it usually is a steady tone)... I then calibrated the oscillators and filters, and then things was ok again...

I suppose that the problem with the Poly sequencer going crazy in record mode with the MIDI loop must have damaged the calibration data somehow...

Honestly... I'm beginning to be scared to hook this thing up to MIDI... if it screws up the calibration data, what else will it screw up next time? ... I don't hope I'll end up with a "brick" at some point :/

[Robot Heart]

Sorry you're having trouble with your Rev2. Sounds like you've already contacted support so you'll get some help deciphering what's going on.

On this machine, it's impossible to end up with a brick. We have a bootloader recovery mode built into the OS, so even if something like a failed OS write procedure occurred you are always able to reload the OS via MIDI DIN by holding Glide while powering up to enter recovery mode.

[Razmo]

Sorry you're having trouble with your Rev2. Sounds like you've already contacted support so you'll get some help deciphering what's going on.

On this machine, it's impossible to end up with a brick. We have a bootloader recovery mode built into the OS, so even if something like a failed OS write procedure occurred you are always able to reload the OS via MIDI DIN by holding Glide while powering up to enter recovery mode.

OK Thanx... I know about the bootloader, I was just nervous that this could accidentally get overwritten somehow...

So maybe I should send the OS once again...

[Razmo]

I investigated a bit more into the MIDI loop I'm experiencing... The NRPN number that the REV2 is sending out and which is looped endlessly is the NRPN 4192... according to the latest updated list of NRPNs I've got from DSI, this NRPN does not exist... the last one in the Global NRPNs is number 4191.

On top of this, every time I change a program from the front panel of the REV2, the REV2 also send out an NRPN... that NRPN is number 4190 (Layer A/B Switch)... I have no idear why that NRPN is sent every time a program is changed, or what that NRPN does for that matter.

I've send another support inquiry about this...

[Razmo]

A bit more...

I've now disconnected the REV2 MIDI In cable, and just hooked it up to a MIDI monitor on my computer to see what it actually send, and when:

1. If local is set to "off" ... then the REV2 does NOT send any NRPN 4190 out with the bank and program change... if local is set to "on" it does send this NRPN 4190 (which is Layer A/B switch) ... it's always sent with a data value of "0".

2. no mater what Local is set to, the REV2 send out NRPN 4192 when you press the RECORD button for the poly sequencer... when the RECORD knob is switching on, the NRPN data is "1"... when it is switched off it is "0"... I thereby assume that NRPN 4192 is an undocumented NRPN switch for the RECORD button... the problem is that this NRPN is somehow send out again when it is received, creating that MIDI endless loop.