I’d like to know what the suffixes in the waveform names mean. While editing sounds, I run into names like

ap: cf3 stretch soft st +

ap = acoustic piano, of course.

cf3 groups samples for a particular piano, just like ep1/ep2/ep3 does for different versions of electric pianos.

What does stretch mean? I guessed that this might be a version of the keymap with wider intervals between samples.

Soft indicates the velocity layer (soft/mid/hard). I realize also that there are sometimes soft1/soft2 and hard1/hard2 versions. So, ep2, therefore, is a sample set with 4 layers contained in ep2 soft1/soft2/hard1/hard2.

Sometimes soft/mid/hard is missing for an instrument, and I find “sw” instead, or the sw version is there in addition to the individual layers. The sw waveform seems to be a multivelocity sample that trades off smaller sampling intervals for multiple velocity layers in one waveform. However, I also thought that “stretch” meant this, also. Will someone please explain?

“st” and “mn” obviously indicate stereo and mono.

However, I’ve not been able to reason out what the “+” and “-” mean when they follow a waveform name. The + and - waveforms seem to have a different tamber, but I can’t reason out the functional intent. They sound like other velocity layers, but, if that’s true, then I don’t understand the difference between the layers in soft1/soft2/mid1/mid2/etc versions of the waveforms.

If any other suffixes come to mind, please bring them up.

Re: Understanding how ROM samples are named

+ and - are different mappings of the same group of samples.

+ waveforms have the samples shifted up in mapping and - waveforms have the samples mapped down in mapping. 

So why do we include different maps ?

First as you mentioned mapping the samples differently providesa slightly different timbre.  - maps will be slightly darker and + maps will be slightly brighter.

Second, because these different waveforms are specifically mapped so no samples overlap this allows you to layer a normally mapped wave and a shifted wave without any phasing. 

Finally, with waves that have vibrato in them , you can layer
two waves and get different rates of vibrato.  - maps will have a slower vibrato than the normal map and + maps will have faster vibrato than the normal map.

Stretch refers the fact that the samples are stretch tuned.

SW refers to samples that have velocity switches built into the waveform itself.  A single waveform can theoritically have up to 128 note ranges and 128 velocity ranges.

Good question and hopefully that covers most of the abbreviations and explains why we made them and how to use them.

Re: Understanding how ROM samples are named

Athan,

What’s the difference between “Sw” and “Kb Sw”? Those suffixes coexist in guitar waveforms…

On a slightly unrelated subject - what does “stretch tuned” mean?

Re: Understanding how ROM samples are named

I think I can explain stretch tuning, at least a little.

Pianos are sometimes (often?) not tuned to mathematical octaves.  I believe the reason has to do with imperfect harmonics.  The harmonics of strings would not naturally line up with mathematical octaves, so the harmonics of one note could clash with the fundamental tone of a note one octave up.  Stretch tunings try to compensate for this.

Check out wikipedia for the gory details.

As to exactly how this is done on a synth, I would not know.  I would assume with proper multi-sampling, the synth would sound the like sampled piano in terms of it’s tuning.

Re: Understanding how ROM samples are named

Sw and KB Sw were really just for the programmers .  KB Sw was a designation so the programmers knew that these waveforms were meant to be able to duplicate the megavoices and were the equivalent of keyboard megavoices on the ES.

MichealJHuman was right about Stretch Tuning.  It a method for tuning pianos to compensate for the mathematics of playing in all 12 keys.