What would be the correct procedure to replace a certain voice of a selected performance with a different voice without altering the performance parameters currently active for the particular part?

Skipping through different voices and then reverting back to the original voice, I figured out that it will always end up in a different sound experience than it was audible before any action was taken.

Even if the “Parameter With Voice” (P.WithVce) setting is switched off before making any voice change - and according to the users manual - there will always be some voice settings which will automatically be copied over to the related performance part.

However, without having tried yet (I’m currently not at my keyboard), does it eventually help to turn the original part “off” first, then make the desired voice change and then turn the particular part back on? At least there’s a note in the users manual that no voice parameters are copied if the voice is turned off. Or will all performance parameters for a certain part will be also lost/initialized when turning that part off first? Thanks.

Sorry, I just found out that I mixed up parameters only related to Song/Pattern mode with those related to Performance Mode. So please forget my statement related to the P.WithVce parameter.

However, my initial question stays the same, as it doesn’t seem to be possible to keep the perfotmance parameters unchanged while changing voices forth and back for one particular performance part.

If Parameter with Voice is turned off before selecting a new Voice for a Part (in a Performance or a Mixing), then what can change are Mono/Poly and three Portamento parameters. If those parameters do change, it should be audibly obvious - if you’re unsure, you could display those parameters and use the EDIT/COMPARE button to check.

Turning the Part Switch off for a Part doesn’t change the Part’s other parameters.

Thanks 5pinDIN for your recommendations. But what if I’d like to change many voices, one after another? Do I really have to review each performance after simply changing a particular voice preset (or just its bank) as part of a performance? Let me explain my problem using another example:

Let’s assume I want to load original Motif XF Performances into my MOXF and I’d like to adapt them in such a way that they will be used in the MOXF efficiently. Let’s further assume most of these performances also use some “factory” voices in their parts, normally located in Motif XF’s voice bank USR1 (which corresponds to MOXF’s voice bank PRE9 by default - containing exactly the same voices). It would now be a good idea to remap all performance parts currently pointing to a USR1 voice to the same voice located in PRE9 while not altering the initial setting of the whole performance in any way. After this general step the whole USR1 bank in MOXF could be used for other customized sounds while still retaining the full functionality of playing the Motif XF performances on the MOXF.

However, in that case the problem is that every attempt to change any USR1 voice bank to PRE9 will also reset some of the current part parameters in the Performance which is unappreciated. No matter if the “P.WithVce” setting is turned on or off (or the voice is generally turned off before the voice/bank change and then turned on again) - the new Performance sound will always differ from the original sound, even though the voices are exactly the same in USR1 and PRE9.

Do you see any possibility to do such a “glabal” change? Or is it really necessary to send all midi bulk data out, edit them manually and try to send them back to the MOXF? As far as I know not even Melas Tools can do this trick by a similar “edit” batch function, right?

I assume that you’re using an XF “all” file (.X3A) to load your MOXF, which is how the MOXF’s USR1 Voice Bank winds up with the XF’s USR1 Voices. Is that the case?

I’m asking because if on my XF I create a Performance with a Part using a Voice that is identical in both a Preset and a User location, changing between them in the Performance doesn’t seem to affect anything. That is, when either of two identical Voices are used in a Part, the result is the same, no matter what bank they are located in.

I’m not sure what you’re experiencing. An experiment might help to determine what’s happening. Could you store an unedited copy of one of your PRE9 Voices to a USR1 location, so that the two locations hold Voices known to be identical? Then try switching between the two in a Performance Part, and see if there’s any difference in the sound. If not, then it brings into question the XF Voice data that I presume has been loaded.

Could you store an unedited copy of one of your PRE9 Voices to a USR1 location, so that the two locations hold Voices known to be identical?

Yes, I tried that approach, too. I always made sure that both banks (PRE9 und USR1) hold exactly the same voice data.

Then try switching between the two in a Performance Part, and see if there’s any difference in the sound. If not, then it brings into question the XF Voice data that I presume has been loaded.

That’s exactly what I’ve been doing. Let’s take the Motif XF performance USR1/001 “Piano Electro” for example. The second part of this performance is mapped to USR1/043 (let’s ignore the third part for now which is also mapped to USR1). If you just change the voice of part two to PRE9 without changing the voice number, you will immediately notice that the bass will immediately change its sound (the “Analog Bullet Bass” in this case). Not even switching back afterwards to USR1 will bring the original sound back. And if you try the same change with the “P.WithVce” parameter turned on, the bass sound will disappear at all (but that’s why at least I’d like to get things working with this parameter turned off).

That’s what can be experienced here on a MOXF - I don’t know if the Motif XF will work differently and will keep its performance settings unchanged while exchanging voices. If yes, this seems to be a MOXF specific “feature” only…

I’m trying to better understand what you’re doing, and a reply to my previous question would help:
I assume that you’re using an XF “all” file (.X3A) to load your MOXF, which is how the MOXF’s USR1 Voice Bank winds up with the XF’s USR1 Voices. Is that the case?

 

B.Minor - 29 July 2014 02:06 PM

Could you store an unedited copy of one of your PRE9 Voices to a USR1 location, so that the two locations hold Voices known to be identical?

Yes, I tried that approach, too. I always made sure that both banks (PRE9 und USR1) hold exactly the same voice data.

Just to make sure…
You selected a PRE9 Voice, and stored it in a USR1 location (for example, stored PRE9/043 Analog Bullet Bass in USR1/043)? Then if you switch back and forth between those two Voices in a Performance Part, the sound changes?

If you are experiencing differences with identical Voices (other than their being in different PRE/USR locations), then please use the COMPARE function, determine what the differences are, and let us know.

Under normal circumstances, there should not be a problem. See:
http://www.motifator.com/index.php/forum/viewthread/470068/

I assume that you’re using an XF “all” file (.X3A) to load your MOXF, which is how the MOXF’s USR1 Voice Bank winds up with the XF’s USR1 Voices. Is that the case?

Yes sir. I used Yamaha’s “CP1 Piano.X3A” library which also includes the “factory” Motif XF voices which will load by default into MOXF’s user voice bank #1 (these are the same voices already included in MOXF’s preset voice bank #9).
In addition this library also includes the “factory” performance banks for the Motif XF.

Just to make sure…
You selected a PRE9 Voice, and stored it in a USR1 location (for example, stored PRE9/043 Analog Bullet Bass in USR1/043)? Then if you switch back and forth between those two Voices in a Performance Part, the sound changes?

Exactly. After experimenting with the sounds directly loaded from the “CP Piano.X3A” library into USR1 (see my first statement), I repeated the same procedure, but this time by copying over the MOXF voice presets manually from PRE9 to USR1, just to make sure the sounds are really the same in both banks. As you mentioned, I did this in particular for the sound I took for my explained example: voice PRE9/043 -> USR1/043 “Analog Bullet Bass” in conjunction with the USR/001 performance “Piano Electro” to be tested. All tests ended up with the same result: From that moment on when the bank is changed from USR1 to PRE9, the bass sound will change in the playing performance.

In fact, some Portamento parameters will change automatically, as you predicted:
Switch: off -> on
Time: 64 -> 2
The rest seems to stay the same. However, these changes will cause the different bass sound which can be easily noticed.

I understand from the users manual that the procedure of changing voices of any performance parts will always change some parameters, even if the “P.WithVce” parameter is turned off. But that’s exavctly why I was asking if there’s a known procedure to overcome this “automatic” behavior. However, what I don’t understand is your comment, that the Motif XF will not cause such changes in the performance sound which is hard to believe for me. If this is really the case, why there are differences here between Motif XF and MOXF?

Under normal circumstances, there should not be a problem. See:
http://www.motifator.com/index.php/forum/viewthread/470068/

Btw. the article you pointed me to describes a different situation. In this case the Motif XF performance will stay untouched and just the correct voice must be copied to the proper user position inside the MOXF in order to make the performance work correctly.

In my case it’s the opposite approach. I want to get rid of all these user locations (in USR1) which are referred to by the performance and I just want to exchange these locations with the corresponding PRE9 voice locations (which are carrying the same voices anyway). This can only be done by updating the performance itself by pointing to the new preset voice in the affected part instead of pointing to the user voice. And that’s exactly the explained problem which drops in here…

B.Minor - 29 July 2014 03:44 PM

[...]In fact, some Portamento parameters will change automatically, as you predicted:
Switch: off -> on
Time: 64 -> 2
The rest seems to stay the same. However, these changes will cause the different bass sound which can be easily noticed.

I understand from the users manual that the procedure of changing voices of any performance parts will always change some parameters, even if the “P.WithVce” parameter is turned off. But that’s exavctly why I was asking if there’s a known procedure to overcome this “automatic” behavior. However, what I don’t understand is your comment, that the Motif XF will not cause such changes in the performance sound which is hard to believe for me. If this is really the case, why there are differences here between Motif XF and MOXF?

For clarity, I should have specified that the Parameters all remain the same, except for (as I previously mentioned) the Mono/Poly and Portamento parameters. Again, these are very easily detected via EDIT/COMPARE.

The XF and MOXF behave similarly. As far as I know, there is no way to avoid this.

B.Minor - 29 July 2014 04:18 PM

Under normal circumstances, there should not be a problem. See:
http://www.motifator.com/index.php/forum/viewthread/470068/

Btw. the article you pointed me to describes a different situation. In this case the Motif XF performance will stay untouched and just the correct voice must be copied to the proper user position inside the MOXF in order to make the performance work correctly.

In my case it’s the opposite approach. I want to get rid of all these user locations (in USR1) which are referred to by the performance and I just want to exchange these locations with the corresponding PRE9 voice locations (which are carrying the same voices anyway). This can only be done by updating the performance itself by pointing to the new preset voice in the affected part instead of pointing to the user voice. And that’s exactly the explained problem which drops in here…

The Mono/Poly and Portamento settings will only remain unchanged in the Performance if they are identical in a new Voice selected for a Part, regardless of where that Voice is located. The Part “inherits” those settings from the Voice.

The only way to determine if those settings are identical in the Performance’s Part and the Voice it refers to are to compare them - in your situation, after changing the Part so it points to its PRE9 location is probably easiest.

The Mono/Poly and Portamento settings will only remain unchanged in the Performance if they are identical in a new Voice selected for a Part, regardless of where that Voice is located.

Okay, understood. Unfortunately that will not be the case most of the time.

The Part “inherits” those settings from the Voice.

Yep, that seems to be the general reason why the audible sound of the whole performance will be altered every time when changing any voice number (or voice bank) inside a certain performance part, even if the “old” and the “new” voices are identical.

But anyway, thanks 5pinDIN for your time and your explanations. I simply have to accept that there’s no other way than to compare also the mono/poly and portamento settings after each manual voice bank change inside a performance part.

For changing USR1 to PRE9 for each affected part in all of the 256 performances inside the MOXF, it would just take about one hour, provided that each performance takes about 14 seconds to be edited and stored back (which is doable in that time). If the performance parameters also have to be compared and adjusted each time, it will take at least twice as long, I guess (or even longer if I edit the third Motif XF performance bank, too). But that’s life. Let’s see if I will do this job one day. Once done, I will have all the Motif XF performances “MOXF-ready” whenever I decide to load them in. Thanks again.

A last time I’d like to drop in here at this point, as there might be an important information for other musicians who are also using the Melas Tools for MOXF (recent version is V2.3.1).

The Total Librarian application provides a very smart way of “source and destination bank mapping” which might be well-known in general, but there’s also a kind of hidden functionality which was not explicitely described in the manual. It exactly allows the automated mapping I wanted to achieve and I’ve explained before, namely re-directing any XF USR1 voice bank pointers (especially referring to XF factory voices) to the equivalent MOXF PRE9 voice bank locations (holding identical built-in MOXF voices) in order to play and to use original XF performances in the MOXF very efficiently (without wasting voice memory space). And here’s how it works:

After creating an empty Total Library instance in the MOXF Total Libraian application and importing the native “X3A” file holding such factory XF voices in USR1 bank and its related XF factory performances in banks USR1 to USR4, a special “bank mapping” dialog will appear. Accepting this dialog by keeping the presented default settings would normally work out in most other cases. However, the good thing now is that the automatic mapping from XF USR1 voices to the corresponding MOXF PRE9 voices can be achieved by entering the following settings:

MOTIF XF Banks --> MOXF Banks

Voices:

USER1 (unchecked)
USER2 (checked) --> USER1 (checked)
USER3 (checked) --> USER2 (checked)
USER4 (checked) --> USER3 (checked)
DR:USR (checked) --> DR:USR (checked)

Performances:

PERF1 (checked) --> PERF1 (checked)
PERF2 (checked) --> PERF2 (checked)
PERF3 (unchecked)
PERF4 (unchecked)

If imported that way, the two generated performance banks 1 and 2 can be immediately transmitted to the MOXF. They will now contain the special mapping where all references to previous USR1 voices have been replaced by references pointing to the identical PRE9 voices, even though “PRESET9” was not an explicit MOXF voice mapping option in the mapping dialog! It’s seems to be just enough to uncheck the USER1 voice bank on the XF side and to apply the other options as shown above.

A later transmission of the imported USER1 voice bank (as well as the imported DR:USR voice bank) to the MOXF is not necessary anymore, as the created XF performances will only use the built-in MOXF PRE9 voices and the UDR voices instead. Yes, even the user drum voices assigned in the factory XF performances are still mapped to the same voices and preset locations inside the MOXF, therefore no additional changes or transmissions are necessary, provided that nothing has been altered in the MOXF before by the user.

After saving the result for later use (from the Total Librarian application as an “allmoxf” file and from the MOXF to USB as an “X6A/All” file) also the two other XF performance banks 3 and 4 can be imported to the Total Librarian application the same way in order to be used in the MOXF. To do so, the whole procedure has to be repeated again, but this time applying slightly different settings while being in the “bank mapping” dialog:

MOTIF XF Banks --> MOXF Banks

Voices:

USER1 (unchecked)
USER2 (checked) --> USER1 (checked)
USER3 (checked) --> USER2 (checked)
USER4 (checked) --> USER3 (checked)
DR:USR (checked) --> DR:USR (checked)

Performances:

PERF1 (unchecked)
PERF2 (unchecked)
PERF3 (checked) --> PERF1 (checked)
PERF4 (checked) --> PERF2 (checked)

Don’t forget to save again (using a different file name of course) after transmitting the performance banks to the MOXF. Thanks to the dependency tracking functionality of the MOXF Total Librarian application, performances created that way will always point to the proper voice locations. As already mentioned, this is nothing special in general. The special thing about it is the fact that XF voice bank USR1 will always be mapped by default to MOXF voice bank PRE9 if it is unchecked on the XF side in the mapping dialog, provided all other settings will be also applied as shown above. A really cool feature, as it allows to do the desired trick without any manual editing!

Btw., personally I find the XF performances much better than those which come with the MOXF (I prefer getting any new ideas while jammin’ to old-school styles instead of suffering from annoying trance/techno beats)!