Sample length

[Raymond62]

In the listening room I mentioned having trouble with trills for the piccolo.
I did some examination and couldn't find a solution for that "stuttered" sound I've got. While trying to get it right suddenly this came up (for all instruments):

the samples have a max length, they must have a minimum length also. What happens when the sequence of notes are so fast that they cannot be played even at the minimum sample length? Does that give me that stuttering?

By the way, the same trill with the flute (at the same moment), but only one octave lower, doesn't have that problem. Is it in this particular case an 'error' in the sample libr for the piccolo?

Raymond

[DPDAN]

Hi Raymond, I hope I can help a bit...

What you are experiencing is a dilema that has to do with the editing/looping of the notes recorded for the library in question. This editing I am talking about is not the editing we do as users, but rather the editing and looping that the library developer does during it's design and development. Depending on the program/library, and how it is edited will give different results.

Sometimes the note is separated into three pieces under the hood, attack, sustain and release. It is imperative that the attacks or notes make sound immediately upon the midi notes' onset.
If this editing is not performed perfectly, and you play a line so fast that all you hear is the attack sample, this will explain the stuttering. It is also important to understand that there is an engine under the hood (so to speak) that is manipulating the sound that will be delivered at the end of these super fast notes.

Since all instruments are edited by hand, (ear) this would explain why one instrument may play the fast notes great, while another one may not.

Each instruments' velocity and sometimes note length, will need to be
altered (usually increased) by the musician using any given library in order to accomplish the desired realistic sound.

Now, if this is indeed GPO we are talking about, and you are doing trills, technically speaking, you should have the sustain pedal depressed just after the first note of the trill and held for the full duration of the trill. This CC64 sustain pedal automaically removes the attack, which is intended to provide a more realistic legato sound..... HOWEVER!!!! sometimes, your trill may sound better without the sustain pedal, you just have to experiment and use your ears to determine what sounds best. That's my experience anyway.

Dan

[Tom Hopkins]

I did some examination and couldn't find a solution for that "stuttered" sound I've got.

Well, I can't be sure exactly what you are getting that sounds like "stuttering" but here is the way trills are designed to be done for piccolo (actually, the same way I would do trills for any other wind instrument):

www.garritan.com/support/PiccTrill.jpg
www.garritan.com/support/PiccTrill.mp3
www.garritan.com/support/PiccTrill.mid

I've just played in the trills from the keyboard and edited the data to to have correct sustain pedal and note overlaps as detailed in the .jpg file. This was done with the default "length" setting of 62 and used the Piccolo Solo V instrument. I don't hear anything "stuttery" here. Hope this helps.

Tom

[Haydn]

Wasn't there an issue with latency on some soundcards causing the legato layer to not work properly? This occured especially with fast notes if memory serves me right.

Jim

[rbowser-]

Raymond, you have gotten Three Stellar X-pert replies to your great question. I know this is an extension of what came up as an issue in The Listening Room re: your new piece.

I just want to make sure that you know you've been indirectly answered when you asked:

"...What happens when the sequence of notes are so fast that they cannot be played even at the minimum sample length? Does that give me that stuttering?.."

The answer being - no. The duration of a sound is determined only by how long you hold down a key - how long the MIDI data is for the length. There's no minimum required. A Gated sample is different - The best example of that being a percussion sample. When you play a Snare Drum sample, it's going to play entirely no matter how briefly you hold the key. That's gated - The entire sample is going to play no matter how briefly or how long you hold down the key. But with the more sophisticated pitched instrument sounds, the samples are Not gated - They will play only for the duration of your MIDI length event.

I'm glad to see that DPDAN echoed what I said on the other thread - that you need to experiment on a per-case-basis for what method is going to make a trill work. It depends on the instrument and the tempo. Sometimes adding Legato works, sometimes it doesn't. Sometimes you need to edit velocity so the temporarily held trill note is as loud as possible - sometimes you don't.

And understand that the "stuttering" problem you're talking about with the Piccolo trills isn't the same as the "machine gun" effect that crops up sometimes with Strings, for instance, during a passage of repeated notes. That's usually a matter of the velocities being too much the same, and the onset of the notes being too perfectly quantized.

Be sure to look at what Tom posted - The links are very instructive.

Randy B.
(rbowser)

[Raymond62]

Thank you all, folks. I certainly will look at those useful links.

Going for a thrilled piccolo,

Raymond