MegaVoices, RX, MultiLayers, SAVoices, DNC e tc. Confusion

I wanna say few things here from my own experience, knowledge and opinion since many times
people get confused and also people CLAIM that a Style made on Yammy or Korg with RX or MV
wont sound any good on a DAW only due to the Multi Layers which is Completely Wrong.
First explanations for those who dont know:
RX-RealExperience - Korg's Technology for MultiLayered Voices.
MegaVoices - Yamaha's terminology of MultiLayered Voices
SA - Yamaha's SuperArticulation Voices which has nothing to do with layers but has to do with
commands and messages sent to the sound itself as Switches, Velocity, Portamento or Staccato
playing and many other things that can be implemented but at the end they all trigger just another
Layer or Oscilator of samples, nothing else.
DNC - Korg's Defined Nuance Control, same as the above Yamaha SA voices.
Korg PA2X/800 can have 6 layers for drums and 16 layers for sounds, DNC or RX, both ways.
About Yamaha i dont know exactly about sounds but i think Drums is only 7.
The Question is ''why a Style that uses MultiLayered Voices wont sound good on another Library
which actually also has MultiLayered Voices''...and my own answer is thats a wrong statement.
The Example i gave in the PA800/MS discussion is the same exact thing as this, the style uses
RX Voices on Korg, Multilayered, especially the Drums while at the same time the same Style
uses MultiLayered Drums in Goliath too but this time instead of 6 layers uses 127 layers...WOW!
Same goes for Sounds, why wouldnt a 16 Layers Guitar not sound good on a 127 Layers Guitar?
Actualy IN EVERY CASE it should sound better due to more correct layering and sounds.
Again this is my opinion and what i've learned, tested, compared and never found that it was
worst or impossible, in most cases was only better, livelier, fuller and warmer.
I wrote this only cuz few times people in here claimed Style Conversion due to MultiLayers wont
end up good which can be true but only Convrting a 6 Layered Drum to 1 Sinlge Layer Drum
on a PA500 not even there or bringing a T3 Style to a PSR1500 but not in the case of bringing
6 layers of Korg or Yamaha to 127 Layers of any of todays Libraries or even a Style from PA500
to a PA2X...the more the layers the more precise sound will come out, each hit is different.

To be honest, Nedim, you are making this sound a lot simpler than it is. Mega voices aren't JUST velocity switched layers. They are also non-pitched sounds, on single notes, which DON'T get transposed while the pitched ones do (that's quite a complex thing to get a single style Part to do on many arrangers), in addition to trigger notes that don't make any sound at all, but affect the notes that ARE playing (chord mutes, for instance).

Sure, there are similar things on GIGA layers and triggers, and many of the VSTi's out there, but the issue is, how easy (if it possible at all) is it to map the control triggers from one format to another..? They are CERTAINLY nothing like each other, one system may rely on CC#'s, another on specific notes that need to be held in combination with other notes being played, etc., etc..

Now, I don't know about you, but I don't know of ANY translation software that is complex enough to make rules for these types of gross differences so that the translation is automated (converting note#'s into CC# of a specific value, for instance, and another value for note off). And that leaves you basically hand converting each of the styles actual note data, part by part, division by division, chord type by chord type... For every style you want to do that on.

Once again, there's this aura of theoretical knowledge about this, rather than actual practice. Lord knows, I've tried my level best to get Yamaha Mega Voice Parts to translate well to Roland's... I'm not making this sh*t up! Perhaps AFTER you have actually TRIED to do this, and perhaps shown us the results, you can start to tell us all how easy it is?

BTW, Yamaha's SA2 technology can't be used in the style, but Korg's DNC voices CAN. Along with a VERY complex Guitar Mode that works in style mode, and gives very accurate guitar voicings and strumming/picking patterns that are interactive with the player. While, once again, there are VSTi's out there that do strumming and picking, it is usually in audio, for one thing, and uses a COMPLETELY different set of trigger notes and control codes. Which once again makes translation a 'by hand' note for note issue again.

I agree that in THEORY, a multi-sampled drumkit OUGHT to translate reasonably easily to a GIGA kit with more layers (though you may be forced to heavily edit the GIGA kit to reflect sample layers that aren't simply increased force ( (tom-toms that turn into tom-tom flams at the top of the velocity range, for instance), but admittedly, this is a 'once done, finished with' process. But some of the more complex things that Mega Voice technology, and DNC voices can do go FAR beyond this, I'm afraid...

I agree with you completely on this, DNC and SA is different thing but when MultiLayering
the closed Synths are not at that stage yet as Kontakt or Play or other Sampling Softwares do
for example using Keys/Notes as CCs to do many things such in Kontakt if you hold a particular
Key the whole Guitar will Strum thruought the Keyboard or Slide or whatever but its not Multi
Layering only, it also has to do with other functions on the Software itself such in this case as
Scripts in Kontakt. Basic Multilayering as in the case of bringing Korg's Style to Kontakt should
be very easy, as i already did with the PA800/MS comparison, almost all of the Voices are Multi
Layered, in both, PA800 and Goliath and i hear the Layers in Both if played track by track or all
at once but overall i agree with you but todays hardware synths are not up to that point yet.
The only thing that can get as close as that is M3 with KARMA which can do similar things as i
explained above doing things while holding a certain Key on the Keyboard. On the other hand
where you mentioned Guitar Mode on Korg, thats a completely different thing and has nothing
to do with layering, it can be brought to a MIDI, Library or a DAW but it is a lot of work doing
that, you have to Sample and tweak a lot of things, if you attach a Guitar to one of those big
Computer Libraries it might sound way ahead but its not worth doing it. It will be better using
just a Computer Guitar Emulator which i use Real Guitar and it works fine for me, better then
Guitar Mode on KorgPA since its library is really vast and relies on really big Samples.

[This message has been edited by Nedim (edited 04-30-2010).]

If all Mega Voices were was multi-layering, you'd have a point. But I pointed out the many things they do that are beyond that. Come on, Nedim. Enlighten us on how you WOULD translate those features?

Or even better, just DO IT, and show us the results.

You can't use the simplest thing that Mega Voices do as 'proof' that ALL of what they can do is possible to translate. You have to show how you would deal with the HARDEST thing they do, too...

As i said, i agree with you and that is correct but if MegaVoices work same as RX
which i believe they do then they dont do nothing besides Layering. RX is only
Layering of Voices on Korg while at the same time DNC kicks in and same goes
for Yamaha MV and then SA kicking in. No, i cant do nothing with DNC or SA.

Actually wrong statement (-: so you dont catch me on words...i can do a lot and way more
then Korg and Yamaha, i dont have to translate their DNC and SA but if i use a Style on
an open System then i wont use Yamaha's SA, i will use a SA from a 150GB Library...
Am i wrong? UPZ!
Cuz we said, why would we transfer them? If we go to an open System...well if we go there,
why would i wanna translate SA or DNC? I'll just use some thing WAY better then them.

...

[This message has been edited by Nedim (edited 04-30-2010).]

Well, I believe that you are wrong, then in that case. Mega Voices are NOT just simply velocity triggers, as I have tried my best to explain. Some of them are, some of them are NOT.

You are also forgetting about the issue of non-pitched sounds not transposing, while pitched ones do. If the MS transposes the entire Track, then those non-pitched sounds will change (into Lord knows what!) as the chords move. On a Yamaha, they don't.

If Mega Voices and RX were the exact same thing, Yamaha wouldn't have needed to invent a whole different term for them...

BTW, once again, my knowledge about this ISN'T mere conjecture. I made a determined effort to see if I could use a Motif Rack to play Translated Yamaha styles from my Roland, it having some Mega Voices on board. This is when I found out about the issue of Tracks having transposed AND non-transposed notes all on the same Track. I simply wish that you would take the time to actually TRY this stuff out before you go ahead and tell us about it.

Like I have, for instance...

Well then i am wrong, i own a Motif but i never tried running styles, as for Korg its
simple, all they do is have their own layering and nothing weird going there, Korg wins...LOLZ!

Try translating a Korg Guitar Mode track, if you think Korg is simple...

Hi Nedim
Mega voices are not like RX voices; in fact only one other manufacture outside Yamaha can play Mega voice styles accurately without conversion, and to achieve this they had to program and design a whole new sound bank for use with Yamaha styles. (This has the added advantage that owners can add them to their existing styles)
As to whether they sound better or worse, is down to whether you like how the programmer has voiced them, and whether you prefer the Yamaha or Wersi (Hypersonic) sound engine.

Quick question
I have read all the Korg manuals on DNC and heard live demos, however as far as I was aware, the DNC only worked on Manual voices not styles, could you just confirm that they are actually available for the style. (The Guitar mode is a separate entity to DNC)

Here are 5 styles in 2 parts, (There are a number of seconds gap between the recordings) the first part is an un-tweaked EMC conversion to Wersi, and the second is played directly in the OAA, and again no tweaking was involved. (The name of the style is at the end of the link)
http://www.4shared.com/audio/HHk3zkSK/Afro_Cuban.html
http://www.4shared.com/audio/2S_7Qx94/Big_Band_Samba.html
http://www.4shared.com/audio/rJbUg45_/Country_Brothers.html
http://www.4shared.com/audio/LV2Tx1ju/Easy_Listening.html
http://www.4shared.com/audio/4vjoa6rE/Italian_Muzurka.html

Bill

Hi Bill,
DNC on Korg arrangers as they claim and as i know works in styles too but i dont
think works in full potential as in Sounds becuse of the Switches which you use
when you play Solo. Commands that are triggered by Legato/Staccato playing,
Velocity, AfterTouch and other Factors would work. I have tried with certain DNC
Sounds to record a part only with one Sound and then listen to that part ad i can
hear the difference but when recording the Part trying to use the switches to be
recorded too it wont work. So in my knowledge in overall works but not full as one
would use the sound when playing it Solo in Style Play Mode.

To Diki:
In the begining of this MP3 http://www.basaristudios.com/sitebuildercontent/sitebuilderfiles/korgpa800.mp3
you can hear the MultiLayer on Korg's PA800, listen to it just like 4 seconds and stop it and then switch to
the following MP3: http://www.basaristudios.com/sitebuildercontent/sitebuilderfiles/goliath.mp3 and you will
hear that even Korg's own style on its own Layering machine doesnt sound as good as on other MultiLayer.

As I said, Nedim, if you only stick to the simplest examples, then sure, it's going to be easy to translate and make good. Trouble is, most modern TOTL arrangers have gone FAR beyond simply multi-layering samples, and use a variety of techniques to get better realism. And NONE of these translate well, if at all.

In Korg's Guitar Mode, the only thing in the style is the trigger notes for the strums and picking patterns, and a totally different engine picks the actual notes and performance noises, etc.. That data isn't in the style at all... Hence, impossible to translate.

I have been told several times that DNC works in style mode, Bill. Some of the rules are intervallic, and I would expect at least THEM to work in style mode. Thing is, Korg have yet to release dedicated samples for DNC, so it isn't quite the factor it could be, but Guitar Mode is another thing altogether... I just don't see how that can be translated at all.

The Guitar Mode translates actually easier then DNC or RX, if you get a Style
with one simple Guitar and record a short MIDI with it in Sequencer shows up
as a standard MIDI file, all of the triggers it diw th the GM Mode now are spread
into regular notes, at exact timing, length and place, now, if you run that MIDI
track thru the same Sound again it will Sound exactly the same as GM.
I have tried Guitar Mode with other Sounds from other synths and VSTi, and it
works almost perfectly if you find exact Sound which nowadays is very easy.
I dont mean the exact Sonic character but only the Noises, Picks, Licks and
stuff that are used in the track. Most new Workstations, Synths and VSTi already
have this as the MOTIF too, the Guitars for GM on PA have standard placing of
all these Guitar FX as on other synths and VSTis, i actually used it for producing
certain songs and albums while not using the Sound itself, only the MIDI info.

Yeah, but that's making an SMF of the output, NOT an actual style...

How do you deal with the notes, triggers and non-pitched sounds all being on one track? Would you separate those out onto different tracks, so some can be transposed according to the chord, and some not? And it still doesn't deal with the issue of the Guitar Mode providing accurate guitar voicings for every chord, including open and barré chording, whereas whatever engine plays it on the MS is going to impose some generic NTT that is going to destroy the voicing.

The devil is in the details, my friend...