Poly chaining ???

[LA Keys]

I'm considering adding a prophet 6 module to increase the number of voices of my P6 keyboard and read the manual but still wondering about a few things.

I basically want to be able to select any patch on the master and simply have 12 voices instead of 6. Is there anything else that need to done to achieve this beside setting the master to poly mode?

I realize that I will now have 2 right and 2 left channel outputs but this is not a problem as I will send all four to my mixer.

With the P6 keyboard as the master:
- If I select a patch on the 'master' will ALL the parameters for that patch be tranfered to the 'slave'?
- If so, is it then possible to use the controls on the slave to make it sound different?
- If not, do I need to program the same patch at the same location for both unit?

LA

[Robot Heart]

I basically want to be able to select any patch on the master and simply have 12 voices instead of 6. Is there anything else that need to done to achieve this beside setting the master to poly mode?

Nope, that's it. Set the master to Poly mode, connect the two instruments via MIDI cable and you're done.

With the P6 keyboard as the master:
- If I select a patch on the 'master' will ALL the parameters for that patch be tranfered to the 'slave'?

Yes.

- If so, is it then possible to use the controls on the slave to make it sound different?

Yes.

- If not, do I need to program the same patch at the same location for both unit?

You need the same programs in the same locations on both units for poly chain to work correctly. Because voice assignment is round robin style, you will get a somewhat random blend of the 12 voices. If you want a "stacked" sound with 6 voices always allocated to one program and 6 voices always allocated to another program, don't use poly chain. Instead, send the MIDI Out of the keyboard to the MIDI In of the module and you'll be playing the two synths as a stack..

[LA Keys]

Many thanks for your answer Robot Heart

Here is what is truly confusing me. You wrote that all the parameters are transfered to the slave (well you actually just replied 'Yes' to that) but also that I need the same program in the same location on both synths.

I'm thinking that since all the parameters are sent to the slave this should be enough to reproduce the same sound (patch) on the slave, without having to first save the same patch (program) on the slave???

To me it looks like only the note information (on/off, velocity) and probably the program number itself are transfered since you need first to have the same program on both synths, but I certainly could be wrong?

LA

[Robot Heart]

Correct. You need to perform a program dump from one synth to the other for poly chain to work correctly. I guess I misunderstood what you were asking, my apologies. To clarify, poly chain transmits the voice allocation information along with all note and parameter change data. It does not automatically perform program dumps from one unit to the other, which would likely cause a lot of grief if we were overwriting people's programs without warning 

[dslsynth]

Poly chaining can be made much easier to use by having the master instrument in the poly chain sending edit buffer dumps in response to program changes causing all instruments to play the master instruments program and hence not requiring any modifications to poly chained instruments program memory thereby easing their life as voice cattle.

Only caveats: (a) edit buffer dumps have to work flawlessly, (b) MIDI bus traffic scaling and (c) slightly more involved implementation.

[LA Keys]

...To clarify, poly chain transmits the voice allocation information along with all note and parameter change data. It does not automatically perform program dumps from one unit to the other...

Great, this is more in line of what I was thinking. I'm fine with that but somehow agree that this would be easier the way  dslsynth described. After all since the whole point of poly chaining is about doubling the amount of voices I would prefer that both synths would 'automatically' have the exact same parameters all the time.

Also, there's no risk of overwriting user programs with a program dump on the temporary buffer, as long as the user himself doesn't actually save the program. Although I understand that since only few peoples will actually buy 2x P6 synth this is most probably not a top priority for your development team.

One last thing that I want to make sure: Since you wrote that the poly chain transmit all the parameters data change does that mean that ALL the changes made with any knobs or otherwise on the master synth will be reflected to the poly chained synth?? Or simply said, is there anything that will not be transmitted?

LA

   

[Robot Heart]

Agreed. dslsynth's method is a good approach, but it is unlikely we'll be making changes to poly chain handling anytime soon.

Yes, all knob movements, button presses, and keys played on the master are reflected on the slave. Once you setup poly chain and have matching programs on both units, the P6 Module essentially becomes a voice expander. It mirrors your actions on the master keyboard and you never have to touch it.

[mmpro1]

It does seem a bit of a waste to have to have all the programs duplicated on both instruments, when an edit buffer transfer could easily be done from the master to the slave. The other present shortcoming is that poly chain doesn't work from a remote midi keyboard triggering the system from the other side of the room.

[dslsynth]

Where the real fun happens is that the edit buffer dump have to be transmitted incrementally and restarted if a program change happens while the edit buffer is transmitted, parameter changes have to be folded in where possible, edit buffer messages have to stop the incremental edit buffer transmission and reset queued parameter change messages, MIDI timing messages have to be passed on correctly and other MIDI events have to be queued for later transmission. But its surely a very fun little data structure exercise!

[Robot Heart]

Why is it a waste if content has to be duplicated on both synths? Most people that buy a module when they already own a keyboard are doing so for the single purpose of poly chaining.

It's easy to do a single program dump, a 10s dump, or a single bank dump if you don't want to duplicate every program. Or, initiate a Basic Program from poly chain and design your own sound in the edit buffer and save the program if you like it. While I agree "it would be nice if poly chain operated based on the edit buffer only", it is by no means a deal-breaker or a clunky implementation.

Regarding your other comment on poly chain not working via MIDI, you are misinformed. You can absolutely trigger a poly chained system via external MIDI.

[dslsynth]

In the ideal world it would be great to have both the current and the edit buffer based poly chaining as these approaches have different timing and convenience properties. However this is not the ideal world. But who knows what the future could bring of improvements on the longer term?

. o O ( seed planting )

[LA Keys]

Well, it may not be perfect but personally I'm fine with the way poly chaining is working as I will use this for only few patches.

Yes dslsynth suggested way is probably 'better' but I'm happy that DSI gave us things like adding FM for the P12 (that I also own) instead on this.

LA

[mmpro1]

Because there is no need to duplicate anything. My Chroma & expander work this way. When chained, the program from the master gets sent to the edit buffer of the slave whenever you select a program. On other days, I may choose to layer the two units, and seeming the duplication is covered through the chained mode, I have twice as many memory locations for storing unique programs. This is why I think it is a waste.

Poly chaining is not working for me when connecting the MIDI out of my Kronos in the mix position to the P6 pair.  I get six voices from the first P6 in the chain & the MIDI program change is not passed to the slaved synth either. Sure would like to hear how you have it working.

Why is it a waste if content has to be duplicated on both synths? Most people that buy a module when they already own a keyboard are doing so for the single purpose of poly chaining.

It's easy to do a single program dump, a 10s dump, or a single bank dump if you don't want to duplicate every program. Or, initiate a Basic Program from poly chain and design your own sound in the edit buffer and save the program if you like it. While I agree "it would be nice if poly chain operated based on the edit buffer only", it is by no means a deal-breaker or a clunky implementation.

Regarding your other comment on poly chain not working via MIDI, you are misinformed. You can absolutely trigger a poly chained system via external MIDI.

[Robot Heart]

Poly chaining is not working for me when connecting the MIDI out of my Kronos in the mix position to the P6 pair.  I get six voices from the first P6 in the chain & the MIDI program change is not passed to the slaved synth either.

This is a bug. We've tested it here and have confirmed it is not working properly. This will be fixed in the next OS update. Thanks for pointing that out!

[mmpro1]

Thanks for looking at it. I am looking forward to the fix.

[Mystic38]

I feel you may have a narrow view of how folk use their synths..

I do not believe I am an outlier to say that I suspect the primary reason why people with a keyboard buy a module would actually be to have two synths, not simply 6 extra voices.. In my particular case, having the keyboard and module versions of the same synth would allow for flexible studio desk configurations, or different sound stations in the studio, or even usage in different room.

A second OB-6 synth and poly-chain are of equal merit to me, and would easily justify to me the cost of a module to complement my OB-6. However, with poly-chain requiring two synchronized synths, (a situation that I neither have the inclination, due to my usage patterns, or even the desire or discipline to comply with).. then to be blunt, that purchase will just not occur.. as poly chain is effectively ruled out to me as a real feature.
 
Hopefully I can encourage some re-think on the topic...

thx!

In my case i have

Why is it a waste if content has to be duplicated on both synths? Most people that buy a module when they already own a keyboard are doing so for the single purpose of poly chaining.

It's easy to do a single program dump, a 10s dump, or a single bank dump if you don't want to duplicate every program. Or, initiate a Basic Program from poly chain and design your own sound in the edit buffer and save the program if you like it. While I agree "it would be nice if poly chain operated based on the edit buffer only", it is by no means a deal-breaker or a clunky implementation.

Regarding your other comment on poly chain not working via MIDI, you are misinformed. You can absolutely trigger a poly chained system via external MIDI.

[hoodoo_ray]

I'm not sure.. I don't know, but I would imagine that there are very few people who have the luxury of both the module and keyboard. Even so, of you do have both and utilise them in the way you describe you use the module and keyboard as two completely different instruments sometimes in two separate rooms etc. Therefore it would strike me that of you needed to polychain you would do this in specific circumstances and to fulfil certain scenarios.

I guess it would mean backing up the module, loading the sound / bank / banks from the keyboard to the module, then using the 12 voices for whatever piece or project requires it, then when done restoring the backed up module and putting it back in the other room.

Your requirement seems to be that when this scenario crops up you want some kind of 'switch' for wont of a better word where you can plug in the module to the keyboard and its additional voices get used by the keyboard without the need to backup / transfer / restore.

If this is the case, that probably would be a bit on an outlier case - I totally get your idea, it would be better if you could simply plug and play the module, but the cost benefit probably does not stack up. But who knows, I could be wrong, maybe lots of people want this!

[eisblau]

Well, IMHO if you have the luxury to own two OB6 then the polychain feature should be on the same level. Meaning direct realtime program buffer dump to the slave so it does not need to have the same program written on the same location in advance. A true voice module-expansion should work that way. It‘s not very handy to match all sound parameters by hand on the slave. So I hope this poly chain feature will be updated still, then I might consider a second OB6.

Another question - can I poly chain two OB6 desktop as well or are there any differences?

[eisblau]

PS.: Seems I dived straight into a DSI Prophet thread after googling DSI OB6 poly chaining, so feel free to move my post under the OB-6 tree.