Genos2 and MegaVoices

[rattley]

Hello Friends!

I mentioned in a different post I've become more than just curious about MegaVoices. I've read a lot of Yamaha documentation and installed Yamaha's MegaEnhancer. What started out as reverse engineering, dissecting a style, and examining all its elements showed me MegaVoices are in every factory style I have checked. Whenever I tried making a style I just mostly chose panel voices, never MegaVoices.

I watched a few videos on youtube last night. One showed how to demo every MegaVoice. Clearly these MegaVoices sounded much richer and detailed than regular voices. Of course if you try to play a MegaVoice with the right hand melody, they appear unusable!

I chose a factory style as my default..... EasyCountryBallad.T554. This style is only 2 bars and very simple. I use it a lot. It has been a favorite.

RHY2 Ambient JazzBrushExtended
BASS MegaVoice ActiveBassFngHmrOn
CHD1 MegaVoice SteelAcousticPick
CHD2 MegaVoice JazzGuitar
PAD MegaVoice KinoLarge
PHR CharacterGrand

I was so surprised at first to see so many MegaVoices. The CharacterGrand is only used in the Endings, so we have 4 MegaVoices doing all the accompaniment voicing. I thought I might swap out the MegaVoices with their regular voice equivalents. I didn't find exact voices but substituted whatever I thought was the best equivalent. Things sounded horrible................

So now I'm wondering how would you record those MegaVoice parts? Do you make the MegaVoices later? I've got so much material about MegaVoices. I'm not sure where to resume. I should also mention that my goal is to make styles from scratch using only Genos2. I don't want to have a computer involved at all except possibly making new MegaVoices with MegaEnhancer. Am I asking too much?? Thanks for reading. -charley

[ChrisHarding]

Hi Charley

I'm not really sure what you want to do? If you use MegaEnhancer and set it for Genos 2 (I obviously set mine for Tyros4) and bring in a midi it will replace any voices it can unless they are already Megavoices. I assume you have already done this so what is your next step, if I may ask. I also assume you have had a detailed look at the Megavoice Map in you Datalist Manual too. After that you still have a midi file and not a style yet so would you use the Style Creator ??

Chris

[rattley]

Hello!


I guess the short of it is I want to replace non-MegaVoices with MegaVoices in a STYLE, not a song. I'm also not sure how to record new style parts to be MegaVoices since they aren't recorded directly. Am I making sense? (he asked him knowingly!! nod nod wink wink) Thanks. -charley

[ChrisHarding]

That's what I thought Charley!

The MegaEnhancer does exactly that BUT only in a midi file so where it can it will replace the existing voice with a megavoice BUT it still saves it as a midi file so you would still need to convert the midi to a style. You already have styles that you just want to change the voices to Mega correct? So what you need is a style editing program that allows you to swop existing voices (like in pad or phrase1 ) and make them megavoices!! I looked at Jorgen's software Midi to Style but didn't see any Mega voices in the list for Genos 2 ..(that would help you anyway!)

DAW software has the same issue as well ..it works on midi !! let me know if you find something

Chris

[BogdanH]

..Whenever I tried making a style I just mostly chose panel voices, never MegaVoices.

MegaVoices (MV) do not sound better than regular voices and so that shouldn't be the reason to use MV instead of regular voices -MV should only be used when needed.
Besides normal primary sound, MV also contain certain articulations/expressions sounds that are characteristic for that particular instrument. What expressions particular MV contains, is described in Data List (MegaVoice Map section).
Here we can see, that the sound that we get from MV, depends on what velocity certain note is played.

For example, a MV acoustic guitar has the following velocity sections:
0-20: Soft string pluck
21-40: Medium pluck
41-60: Hard pluck
61-75: Dead pluck
76-90: Muted pluck
91- ... ...
As we can see, for velocities up to 60, MV behave quite similar to regular voice (except in case of MV, the differences in intensity are bigger).
When we use a regular voice and play a note at velocity 80, we will get a normal string pluck, which is (as expected) louder than at velocity of 60 (and usually also sounds more harsh/intense).
However if we use MV and play a note at velocity 80, we don't get louder pluck (than at 60): we will get muted pluck instead. And that's why we can not simply substitute regular voice with MV.

If we wish to create guitar accompaniment that sounds like this, then we simply need to use MV -because regular guitar voice doesn't contain those articulation (string mute) effects.
Technically speaking, we could achieve similar by using a regular voice and a secondary voice which would contain those effects, but we would then occupy two track for that.

So now I'm wondering how would you record those MegaVoice parts? Do you make the MegaVoices later?...

I record MV track the same way as if I would use regular voice. In case of guitar MV, I play notes relative gently at recording, because they sound normal only up to velocity 60. Later I adjust velocities according to my wish in StepEditor -that is, at what beat guitar strings should sound normal or muted.

I just wish to emphasize again: if you plan no expression effect for particular voice/track (because it's not needed), then just use regular voices.
I would say, it depends on for what music genre you're making accompaniment. I think that MV are often of benefit for acoustic accompaniment where only maybe 3-4 tracks are used in style.

I should also mention that my goal is to make styles from scratch using only Genos2. I don't want to have a computer involved at all except possibly making new MegaVoices with MegaEnhancer.

I think that's the right approach! .. and forget MegaEnhancer

Bogdan

[rattley]

Thanks Guys!!

Things have gotten clearer. I think.......... While listening to some styles I am able to change existing MegaVoices to other MegaVoices. In a video the gentleman demos the Megavoices on Genos2. It's hard to see in the video, but it looks like he goes to the MegaVoice folder and I see the megavoices. I think he clicks on the demo button and the selected MegaVoice plays. However, when I do this on my Genos2 the demo button looks disabled and will not work. I must be in the wrong mode or something? Am I seeing what I think I'm seeing in the video? It would be nice to hear those MegaVoices just as regular voices work. Thanks. -charley

[pjd]

Yep, you can replace a Megavoice with another Megavoice. However, your mileage may vary (YMMV).

If you browse through the Megavoice maps, you'll see that not all maps are identical. Compare the bass Megavoice maps (in order to keep things simple). Sometimes the maps align pretty well, sometime not. If maps do not align well, you may hear anomalous sounds.

Remember all of the posts on the old Forum, "Why am I hearing a scratching noise?" That's because you are (re)using the same MIDI note data and a few of those note events are triggering some other sound in the target Megavoice.

Mega Enhancer is another case of YMMV. Often, I don't hear much change at all.

Hope this helps -- pj

[pjd]

I suspect that Yamaha have tools to help them create Megavoice tracks, especially guitar tracks. The guitar parts may be played on a MIDI guitar controller, for example.

Unless someone wants to go crazy (!), they wouldn't try to create Megavoice notes by hand in a DAW. It is truly painful to do this.

That's why I suggest using the equivalent Super Articulation voice instead. You won't need to worry about all of those velocity levels. What you'll lose is precise control over certain sounds like slides or fret noise. In terms of basic detail, the SA voice will give 90% without all the manual labor.

An SA voice uses the same samples as the equivalent Megavoice. For example, the SA Seattle String voices use the same samples as the Seattle String Megavoice.

-- pj

[pjd]

A style file is a Standard MIDI File in disguise. A style file contains "chunks" of Yamaha proprietary data for OTS and CASM.

It's possible to read a style into a DAW and edit the MIDI notes, voices, etc. Unfortunately, DAWs throw away the Yamaha chunks.

Jorgen has the split/splice tool. The tool splits a style file into the MIDI part and the non-MIDI part (OTS, CASM, etc.) So, one can split the style file, edit the MIDI, and splice (merge) the non-MIDI part with the edited MIDI.

So, if you have the MIDI part, run Mega Enhancer on it! Then splice the two parts together again.

Good luck -- pj

[ChrisHarding]

Thanks PJ

Just out of interest (you mentioned replacing mega voices with SA ones) Once you split the style file ..is the midi part moveable? in other words could you load it into something like PSRUTI and change it and then save it back to Jorgen's program and sync it up to the non midi bit???

Chris

[pjd]

Just out of interest (you mentioned replacing mega voices with SA ones) Once you split the style file ..is the midi part moveable? in other words could you load it into something like PSRUTI and change it and then save it back to Jorgen's program and sync it up to the non midi bit???

Hi Chris --

This is exactly the usage scenario for Jørgen's split/splice tool. The split operation produces a Standard Midi File (Type 0, extension .mid) and a binary file with the non-MIDI data (extension .nmi). The splice operation does the reverse and combines a .mid with a .nmi to produce a style file (extension .sty).

There are a few limitations. Since the CASM is the same (CASM in, CASM out), the outgoing MIDI needs to adhere to the original CASM programming on a track-by-track basis (root note, chord, etc.)

Hope this info helps! It's worth experimenting. Keep a copy of the original style, just in case. It's only bits...

All the best -- pj

[ChrisHarding]

Thanks PJ

If you replace the megavoices wit SA voices surely then CASM numbers will change with the new voice.. Just asking!!

Chris

[BogdanH]

It makes no sense to directly replace MV with any regular (incl.S.Art) voices, because the result will be wrong. In regard of notes velocities (and that's the difference between MV and other voices), S.Art voices behave the same as regular voices and so we can replace regular voices with S.Art voices (and vice verse) -but that's not the case with MV.

Let's take regular (or S.Art) guitar voice for example.. If we play certain note at velocity 60, we get a normal guitar sound and if we then play the same note at velocity 80, we will get the same (or more precise: very similar) sound which will only be louder than at velocity 60. The same is the case if we now play that note at velocity 40: as expected, it will sound quieter (and probably softer).

Now let's play the same note at velocity 60 by using MV. We will get a normal guitar sound (string pluck), however it will louder than if regular voice would be used and it will also sound more harsh -it will actually sound similar as if note would be played at velocity above 100 by using regular voice.
If we now play the same (MV) note at velocity 80, we won't get normal guitar sound (as that was the case with regular voice): we will get muted guitar string sound -and that's the difference between MV and other voices.
If we now play that (MV) note at velocity 40, we will get normal guitar sound, but it won't necessary be quieter than at velocity of 60 -it will only sound softer.
To put it simplified: in case of (guitar) MV, the sound will have the same loudness at velocities 20, 30 and 60 -it will only sound differently. That's also the reason why it's impossible to use MV for normal playing as RH voice -and let's not forget, that if we play MV at velocities above 60, we will get some muted "thump" instead of guitar sound.

The only way to replace MV with any other voice (and opposite) would be, if we would change each note velocity in the track. But even then the result will sound wrong -because we can not simply replace muted sounds with normal sounds.

Bogdan

[pjd]

If you replace the megavoices wit SA voices surely then CASM numbers will change with the new voice.. Just asking!!

Hi Chris --

I agree with Bogdan's post. And for that reason, converting Megavoice (M) to a regular or SA voice is a difficult job.

As to CASM, it gets tricky. Yamaha invented SFF2 (also called SFF GE/ SFF Guitar Edition) which adds the means to deal with a Megavoice in a single Style part. The problem with MV in Styles: The regular notes need to be transposed while the FX notes should not be transposed. SFF2 allows for this. If the original style is a Yamaha factory style, it probably uses the full SFF2 (SFF GE) capability. That SFF2 stuff is encoded in the CASM.

Soooooooo, when I do a conversion and I need to replace a Megavoice (e.g., converting a G2 style to G1), I try to find a target MV with the same or similar note map as the source MV. I try to leave the original MIDI data untouched. Most times, you can get away with this kind of conversion! It's how I converted G2 R&B styles to G1.

Charley is probably going crazy with all of this nerdy detail. Unfortunately, wading around in MV means learning the technical details. No easy way around it, sad to say.

Hope this discussion helps. That's why it makes sense for MV discussion to have its own (this) thread.

Take care -- pj

[rattley]

Hello!!


The more nerdy details the better. I do retain some of this. Sooner or later (unfortunately more later) it sinks in and that bell goes ting-a-ling!

After hearing what was said here I now know I don't have to spend as much time with MegaVoices. Forgive me.... .... If SA voicing gives the same qualities as MegaVoices, why aren't they in the accompaniment as prevelent as the MegaVoices?? I'm sure Yamaha "makes" their styles with special propriatary software and doesn't use the Style Creator. Maybe I'm not so sure..........I'm making many assumptions here. It's all probably done entirely by computers, but what a computer that might be!!! -charley