You can install Waveforms and Samples on the FL1024M Flash Memory Module.

You do not install Banks of sounds on the Flash Board, just the Waveforms and Samples. To understand how this works, requires you understand the Synthesizer architecture of the MOXF.

A VOICE is the basic fundamental playable entity. If contains an Oscillator - a set of Filters - and an Amplifier. LFOs that can be applied to modulate Oscillator Pitch (vibrato), to modulate the Filters for timbre (wah-wah) and modulate the Amplifier loudness (tremolo). Effects processors, EQ, etc., etc.

The Oscillator in a sampled based synth is an audio recording. The data size of the Oscillator is huge in comparison to the size of all the other settings combined and multiplied by large factor.

Example, all the Voice parameters (not counting the Oscillator Wave data) might be about 2KB (2,000 bytes). Meanwhile, the Waveforms and Samples that form the Oscillator could be in excess of 400MB (that’s 400,000,000 bytes) take a meticulously sampled Electric Piano like the Chick Corea Mark V - it’s approximately 400MB of audio data to 2KB of Voice parameters that shape it.

Sometimes Kilobytes and Megabytes don’t translate to musicians... substitute dollar$ to get a clear picture of the scale of magnitude:
You’ve probably held $2,000 in your hands; but it is unlikely you could even carry $400,000,000 by yourself... although you’d give a good try!
You could walk around with $2000 in your pocket, you’d need a truck for the $400 millio!!!

While the rest of the Voice data, literally, loads in a blink of an eye, you don’t want to have to wait while the oscillator’s audio data blinks.... its two hundred thousands times more data. Leave it to say Load times for the huge Waveform and Sample Data is measured in like 20-30 minutes. Even at speedy load times... it is simply that much larger in size.

The idea / concept behind FLASH is to install the large audio data onto non-volatile memory so that you only have to do it once. The Voice data that uses it will be instructed *where* exactly to look to find the data that it needs.
Then you can load entire banks of Voice data (sans the big audio data)... you can load a bank of Voice data in, literally, seconds.

The Voice data will address the MOXF’s Waveform List to know *where* the necessary audio data is located. The Waveform List is like custom “catalog” of what data goes with which Voices.

Example... at the Factory Yamaha installed to the Factory Wave ROM, the Waveforms and Samples that make up a Yamaha CFSIII Concert Grand Piano... you don’t need to load this data, it is always there. How many Voices can be made with this data? It’s truly unlimited. You can load 384 Normal Voices at a time. And they will be quick to load (seconds) because you don’t have to wait while megabytes of data are loaded. That data is in ROM (Read Only Memory).

The Flash Board allows the user to created their on User Wave ROM... you get to install the Waveforms and Samples of the instruments you want to build Voices from. You can then create Voices that point to and use that data.

So to answer the question... install another Bank?
You don’t, you install only the Waveforms and Samples.
How many Banks of Voices can use that Waveform and Sample data? It’s unlimited.

A User Bank can be loaded to the USER 1 (128), USER 2 (128), USER 3 (128) and DRUM USER (32) locations.
The Voice data loads in seconds and because the audio data that it will use has already been “burned in” on the Flash Board, you will not have to wait the 20 to 30 minutes it would take to load.

Across a week’s time this is a significant amount of time. I think I did the math once.... I had about 28 Minutes of Load time data in my Basic MOXF ... Across a year it saved me several days of just “waiting” for data to load.

Why is the audio data so big... recording at 44,100 samples per second, 16-bit, Stereo would use approximately 10MB per minute of audio.

Here are the limits:
A maximum of 2,048 Waveforms
A maximum of 8,192 Samples
A maximum of 1,022MB of total audio data
It works on a “whichever comes first basis”

Once you reach 2048 Waveforms it will matter not that you have not used all 1022MB of room, you’re out of memory
Once you reach 8,192 Samples it will matter not if you have not used all 1022MB of room on the board, you’re out of memory.
If you reach 1022MB without using all 2048Waveforms, you’re out of memory.
You’re out of memory as soon as any of the limits are reached.

Hope that helps.

Hi,

OMG ! I do not know how to thank you enough for this perfect and incredible explanation.
Everything is clear now - I won't bother you anymore !
Really, many many thanks for your time and knowledge.

Best regards.
G.

One other limit - once you have loaded voices in all of the USER voice locations, the only way you can load more Voices is to overwrite some that you already loaded previously. My recollection of Inspiration in a Flash is that on Motif it had 4 full user voice banks. They modified a MOXF version, I believe to account for the fact MOXF only has 3 USER Voice banks = 3 x 128 Voices each.

What this means for you is if you did load the full Inspiration library, you'll have to decide which Voices you love and want to keep, and which you don't want to keep. You can't keep them all and then load more Voices. Once you make those decisions ... you'll need to some transfer/cleanup if you want to free up a complete USER bank

@David,
If you install the Waveforms and Samples of Inspiration in a Flash to your FL1024M, then any time you want to use the 384 Voices you could load the Voices *without* the Waveforms, *without* the Samples, Load time will be a few seconds.

Even if you decide to load just one Voice from that data, or even if you decide to load 1 Bank of Voices, or all 384 + 32 drum kits because you installed the large audio data, you don’t have to wait for them to load ever again.

You can create new data assembled from several Libraries, please read my explanation again, you’re still missing a major piece of the puzzle. The concept is not to never have to load data, the concept is never having to wait 20 or 30 minutes to load data after you’ve *installed* the large audio data. You do that once, and all subsequent loads take seconds, not minutes.

The trick is never install Waveforms and Samples you don’t use. Why take up valuable room on your Flash Board with Data you are not going to use. That only makes sense.