Analogue-sequencer

Jesper, 

> Phew, I finally got them in sync, thanks to all for helping.
> Colin, I would still love the midi trigger thing, but it is 
> not essential 
> anymore ;)

That's a relief - it wasn't something that would have been high on the list
of priorities for the remaining space.
Changes to the code that handle the clock doubling might have been more
likely, but if you start 'smoothing' the incoming clock, it can make the
timing get a bit loose.
P3 will only start losing time if it receives three clock pulses in a 5
millisecond window - otherwise it is able to catch up.
Three ticks in 5 milliseconds is somewhere around 1000 bpm, so you can see
that Cubase's timing is more than a little erratic...
 
Cheers,
Colin f

Phew, I finally got them in sync, thanks to all for helping.
Colin, I would still love the midi trigger thing, but it is not essential 
anymore ;)

If anyone cares this is what I learned.
I tried filtering out the "F8-Midi timing clock" signals using Midi-OX, but 
neither the P3 nor the Electribe were happy with it. The P3 didn't start 
and the Electribe didn't sync properly.
Using Midi Machine Control was a great idea, but apparently it only works 
if the controlled devices send back a sysex string telling "I'm running 
now, you can start too". I tried to send the sysex string from Midi-OX but 
I never got Cubase started.

So what I ended up with is this:
 From Cubase I send MTC and Midi Clock to Midi Yoke (a virtual midi port).
In Midi-OX I filter out all "F8-Timing Clock", but let Midi start and Midi 
stop pass thru.
To send midi clock to the P3 I use Midi-Ox's MIDI sync transport. In the 
MIDI sync transport I select both "Send clock" and "Sync and send MTC 
concurrently".

The MTC part isn't necessay, but I felt the sync was a bit tighter when I 
used it.
So basically Cubase is sending the start and stop signals, and Midi-OX is 
sending all the timing signals.

I'm sure this doesn't work tomorrow,
Jesper.

On Sun, 28 Nov 2004 14:31:18 +0100, Jesper <jj@opq.dk> wrote:

>Using Midi Machine Control was a great idea, but apparently it only works 
>if the controlled devices send back a sysex string telling "I'm running 
>now, you can start too". I tried to send the sysex string from Midi-OX but 
>I never got Cubase started.

MMC is typically good for remote control of tape machines, HD
recorders and so on. For example, I have a Cubase mixermap that sends
transport controls to my VS-880 and it works fine - one way is all
that's necessary as it merely duplicates the "run" hardware button
etc.

>The MTC part isn't necessay, but I felt the sync was a bit tighter when I 
>used it.

Pretty unlikely - MIDI Time Code is "SMPTE for MIDI" and an absolute
timeframe that step sequencers don't listen to. Or at least I don't
know of any that do... 

>I'm sure this doesn't work tomorrow,

hehe, I know where you're coming from <g>

Paul
---
Paul Nagle / Soft Room Music
Email: paul@softroom.co.uk www.softroom.co.uk
                           www.BogusFocus.com