Do I need to purchase the flash mmory for virtual piano collection voice liabraries for xf?

my other question is that I need to decide whether I should buy the xs6 and purchase voice libraries from the motifator shop or should I rather buy the xf6 instead?

Does the full concert grand on the xf sound the same as the one on the xs?

I like to make a distinction between VOICE LIBRARIES and SAMPLE LIBRARIES… Yes they both contain VOICES, but the SAMPLE Library requires that you load audio data to SDRAM or to FLASH in order for the Library to have an osciallator (sound source).

The VOICE Library takes advantage of the PRESET waveforms (Oscillators) that are built in to the instrument and remain as a permanent part of the unit.

The VOICE parameters are every parameter you see when you press [EDIT] on the Motif XF while in VOICE mode. They include all the COMMON EDIT parameters and all but WAVE parameters.

The WAVE parameters are found in EDIT > Press [1]-[8] to view an Element’s parameters > press [F1] OSCILLATOR
WAVE BANK
WAVE MAIN CATEGORY
WAVE SUB CATEGORY
WAVE NUMBER

The WAVE BANK can target a PRE, USR, FL1 or FL2 BANK
PRE = 3,977 WAVEFORMS that Yamaha provides as your permanent sound sources
USR = 128 WAVEFORMS that you can sample to volatile SDRAM
FL1 = 2,048 WAVEFORMS that you can burn to FLASH board #1
FL2 = 2,048 WAVEFORMS that you can burn to FLASH board #2

VOICE parameters will always remain between power down and power up, what disappears is the data that is WAVE BANK = USR… Because this is volatile, when you power down, a VOICE that points to a USR waveform will be blank when you power back up… It will still point to a USR WAVE location - but the SDRAM will be empty. You must reload.

A FLASH Board (by contrast) does not store VOICE Banks, it stores WAVEFORMS and Samples. Those samples and Waveforms are used in the VOICES that you can store in the non-volatile USER 1, USER 2, USER 3, USER 4 and USER Drum VOICE locations.

aaronnkosi - 26 September 2010 07:25 AM

Do I need to purchase the flash mmory for virtual piano collection voice liabraries for xf?

The current Motif XS “Virtual Pianos Collection” is 193MB in size. Therefore, if you do not have a flashboard installed on your XF, you cannot load it.

We are working to convert DCP libraries that are larger than 128MB, so
that they will be divided up into two or more smaller (under 128MB) packages, and as a full-size library meant for the flashboard.

I hope I’m not hijacking the thread but as I’m understanding now, you have an empty 128mb SDRAM on the XF. Yea?

And, if I were to get use the small bank (uses 128mb) from the Signature Piano (KSounds) and load them in, I would not be able to load any other samples into the XF UNLESS I get the Flash?

So it’s basically a RAM but it’s not like a computer ram where it stores stuff to be accessed by the CPU but it acts more as a flash ‘drive’ to store samples in. Please correct me if I am wrong! :)

I was just confused as I always think RAM as a quick temporary storage and not as a permanent storage where it acts like a hard drive (in this case, like a thumb flash drive)

Hope I made sense.. ;)

ps. IF I don’t like certain voices and I know I am never going to use it.. can I load a custom voice or samples in spot of that voice? (So I can conserve some space?) <- This really didn’t make sense.. hahaha

Thank you!

dkijc - 30 September 2010 06:04 PM

And, if I were to get use the small bank (uses 128mb) from the Signature Piano (KSounds) and load them in, I would not be able to load any other samples into the XF UNLESS I get the Flash?

Yes that’s correct.

So it’s basically a RAM but it’s not like a computer ram where it stores stuff to be accessed by the CPU but it acts more as a flash ‘drive’ to store samples in. Please correct me if I am wrong! :)

Also correct.

The user RAM and the flash RAM are for storing samples ONLY. You cannot store voices or performances or songs or anything else to the flashboard or the user RAM. Just the samples.

So yes, if you don’t like a voice you can delete it from its location in any of the four user banks on the XS. You can then overwrite the sample or samples that voice used in the flashboard.

DavePolich - 30 September 2010 08:01 PM

dkijc - 30 September 2010 06:04 PM

And, if I were to get use the small bank (uses 128mb) from the Signature Piano (KSounds) and load them in, I would not be able to load any other samples into the XF UNLESS I get the Flash?

Yes that’s correct.

So it’s basically a RAM but it’s not like a computer ram where it stores stuff to be accessed by the CPU but it acts more as a flash ‘drive’ to store samples in. Please correct me if I am wrong! :)

Also correct.

The user RAM and the flash RAM are for storing samples ONLY. You cannot store voices or performances or songs or anything else to the flashboard or the user RAM. Just the samples.

So yes, if you don’t like a voice you can delete it from its location in any of the four user banks on the XS. You can then overwrite the sample or samples that voice used in the flashboard.

...But I can save Voices, Banks and samples to a USB Drive? Right?…

macalister - 30 September 2010 11:12 PM

DavePolich - 30 September 2010 08:01 PM

dkijc - 30 September 2010 06:04 PM

And, if I were to get use the small bank (uses 128mb) from the Signature Piano (KSounds) and load them in, I would not be able to load any other samples into the XF UNLESS I get the Flash?

Yes that’s correct.

So it’s basically a RAM but it’s not like a computer ram where it stores stuff to be accessed by the CPU but it acts more as a flash ‘drive’ to store samples in. Please correct me if I am wrong! :)

Also correct.

The user RAM and the flash RAM are for storing samples ONLY. You cannot store voices or performances or songs or anything else to the flashboard or the user RAM. Just the samples.

So yes, if you don’t like a voice you can delete it from its location in any of the four user banks on the XS. You can then overwrite the sample or samples that voice used in the flashboard.

...But I can save Voices, Banks and samples to a USB Drive? Right?…

..But I can save Voices, Banks and samples to a USB Drive? Right?…

Yes, you can save voices and samples, songs, patterns, anything to a USB
drive.

When you save voices that depend on samples in Flash, you have the option
of saving just the voices without the attached samples. After all, the samples in Flash don’t go away when you power down, they’re always there
until you overwrite them.

Bad Mister has posted lengthy replies about saving and loading to flash -
worth it to read them again, your questions are also answered there.

Thank you very much DavePolich.

From what I read from Bad_Mister, Flash saves a waveform and samples where the Voice can look up to use the custom ‘sound’. Is this correct? (Please bare with me, as this is all new to me! :) )

Then, what is the difference between a WAVEFORM and SAMPLES? Does a Voice look for Waveform OR Sample?

In my first question about overriding the default ‘Voice’ was, if I can actually override the default waveform/sample (this two concept is still confusing me, hence the question above) so instead of buying the extra Flash, I can just override the original waveform/sample.

I’m asking not because I don’t plan on buying the Flash but just curious as to how this all works :)

Thanks so much!

what is the difference between a WAVEFORM and SAMPLES? Does a Voice look for Waveform OR Sample?

Excellent question - to ask the question is begin to understand. This is what many people who use the technology don’t know - but it can only help you to know if you really want to understand how today’s sample playback based synthesizers work.

Let’s take the popular S700 piano VOICE (from the LoyaltyS700 giveaway) as an example. If you were to look at the data that makes this VOICE happen - from the top down - when the S700 is loaded to RAM of the Motif, it loads 3 new WAVEFORMS… (it is made from 3 S700 Waveforms plus a Preset KEY-OFF Sound waveform)

VOICE - these are the Motif parameters that allow you to play the sound. They include all the parameters you see when you press [EDIT] and [COMMON EDIT]… Name, Volume, Pan, Mono/Poly mode, Note shift, PB Range, Portamento, Micro tuning, Controllers, Arpeggios, Common LFO, Effects, the 8 Elements with each of their Pitch/Filter/Amplitude generator parameters, etc., each Element has an LFO, each has its own EQ, etc., etc., etc These parameters are non-volatile and remain in memory until you initialize them, or write over them. These parameters are used to ‘dress’ up the WAVEFORM

On the VOICE > EDIT > ELEMENT 1-8 > [F1] OSCILLATOR page you can see where each of the Elements of this VOICE point to a specific WAVEFORM BANK and NUMBER. It is on this screen that a VOICE references it sound source. The Sound Source can be a PRESET Waveform (Yamaha provides 3,977 of them in permanent memory), it can be a USER Waveform (you can load 128 of them into volatile SDRAM), it can be a Waveform that you’ve “burned” to one of the optional FLASH 1 or FLASH 2 boards (each board can hold a maximum of 2,048 Waveforms).

The S700 piano Voice data is prinicipally made up of 3 new Waveforms that get loaded to RAM: An S700 Hard Strike, S700 Medium Strike and the S700 Soft strike.

Just 3 new stereo WAVEFORMS!

Each of the three Waveform is made from 90 individual samples!

so a WAVEFORM is defined as follows:
A Waveform can have as few as one but as many as 128 samples. The samples are mapped to KEYBANKS (the word itself should help you understand what this is). A Keybank is a map of how the individual sample is going to be triggered: horizontally across the keyboard, and/or vertically by velocity.

KEYBANK = a sample that has been given KEY RANGE and VELOCITY RANGE. (there are a few other parameters: play mode (direction: one shot, reverse, loop), pan (in the case of a stereo pair), level.

As many as two samples can occupy the same KEYBANK (this accommodates stereo samples). Stereo is a mono sample hard left and a related mono sample hard right.

An example of a 1 KEY KEYBANK would be a drum or percussion sound. Typically it is mapped to a single key… You could however, stack multiple versions to make velocity swap… The Closed Hihat in the Power Standard Kit 1 - is a four way velocity swap. 4 different closed hihat strikes mapped vertically so each sounds at a different region. So the Closed Hihat in that kit is 8 samples (it is recorded in stereo) mapped to F#1. They do not all trigger at the same time, each has a unique velocity range.

Examples of horizontal key range mapping abound in musical instrument sounds. Musical sounds can have both key and velocity range mapping simultaneously.

So the S700 piano is 45 left mono samples and 45 right mono samples basically every other key is sampled in stereo for the hard strike.

45 left mono samples and 45 right mono samples (every other key is sampled in stereo for the medium strike).

45 left mono samples and 45 right mono samples (every other key is sampled in stereo for the soft strike).

3 WAVEFORMS made up of 270 samples = the S700 piano data that gets loaded into RAM.

Back to the playable entity: the VOICE
Each Motif XF Voice can be made up of as many as 8 Elements.
An Element contains a WAVEFORM.
A WAVEFORM contains one or more KEYBANKS
A KEYBANK contains as many as two samples.

So when it comes to understanding what goes on the FLASH boards. It is not just the sample (original audio data), you want to also have how that data is configured… that’s the WAVEFORM.

The WAVEFORM data organizes the Samples into something that can be used. Having 90 separate samples for the hard strike S700 unorganized would be a nightmare. Where do they go… when are they supposed to sound. Now multiple that by 3.

The sample’s data size uses the larger portion of the memory… rule of thumb: at 44.1kHz sample rate, 10MB is used for every stereo minute of audio. So audio samples are what are big in size and take a long time to load. The parameters that tell them where to go and when to play are, in comparison, a much smaller packet of data. But WAVEFORMS and SAMPLES.

The ALL WAVEFORM file will handle both the pointers and the audio.
But as you start to understand all of this, it will begin to become clear why you would want to be able to load (at times) “without samples”. Because, and this is a very important thing to understand: A WAVEFORM can be used in many different VOICES - you can change the parameters that “dress” it up and have a completely different result.

I like the analogy (and those old enough to remember actual Analog Synthesizer will appreciate this): In an analog synth you were give a handful of WAVEFORMS: Sawtooth UP, Sawtooth Down, Pulse, and Square and from those handful of waveforms you created hundreds of VOICES… it all depends on how you “dress” them up!

Now this is a simplified version of what is going on - it is actually a bit more complex but this should give you a really good idea of what is happening.

Bad_Mister - 01 October 2010 01:02 PM

I like the analogy (and those old enough to remember actual Analog Synthesizer will appreciate this): In an analog synth you were give a handful of WAVEFORMS: Sawtooth UP, Sawtooth Down, Pulse, and Square and from those handful of waveforms you created hundreds of VOICES… it all depends on how you “dress” them up!

Now this is a simplified version of what is going on - it is actually a bit more complex but this should give you a really good idea of what is happening.

Thanks so much, Bad_Mister! I think I have a ‘grasp’ of what’s going on and you summed it up nicely with the waveforms! Hahaha why couldn’t I make the connection between Waveforms and Waves.. :)

In short, Samples are what makes up the basic foundation of the ‘sound’ and the Waveforms are that collects all of that and determines which samples need to be played depending on what the user requested (eg. soft, med, hard strike). And these waveforms are stored as Element and a Voice can have 8 elements.

Is this, in general, correct? (most likely I’m totally off.. haha)

I didn’t quite get the Keybanks part. So in the example of the S700, how are the samples distributed to Keybanks?

Also, what are Elements, then? Can only ONE waveform be stored as an Element or all the waveform the voice uses are stored as one element? If so, what other elements can a voice have besides being a Waveform?

I have other questions, as well but I can’t really verbalize it now. Hahah I’ll bother you more next time! :)

Thanks so much!

I didn’t quite get the Keybanks part. So in the example of the S700, how are the samples distributed to Keybanks?

When the S700 acoustic piano was sampled just about every other note was sampled: C7, B6, G#6, F6, D#6, C#6, etc. and they were sampled in stereo. So KEYBANKS if you understand them to be note ranges… the C7 sample (highest note on the piano is mapped C7~G8. The B6 sample is mapped A6~B6, the G#6 sample is mapped from F#6~G#6, the F6 sample is mapped E6~F6 and so on.

Also, what are Elements, then? Can only ONE waveform be stored as an Element or all the waveform the voice uses are stored as one element?  If so, what other elements can a voice have besides being a Waveform?

An Element in the Motif XS/XF is always multi-sample. One Waveform can be stored in an Element.

Because Yamaha, in particular, makes, or used to make multiple technology synthesizers (original Motif, Motif ES, EX5/7 etc)… the term Element is used. In sample-based instruments the Element is always made up of multiple samples. But in FM synthesizers (DX) the Element is made from 6 operators, in Virtual Analog (AN) synthesizers the Element is made from a set of virtual VCOs, in a Virtual Acoustic (VL) the Element is a physical model of a column of air in motion. But in samples - the Element is a multi-sample set of KEYBANKS.

Wow...I always knew Bad_Mister was good, but man, Bad_Mister is really good.