Connexion
Ce forum permet à des personnes du monde entier de communiquer, c′est pourquoi les messages échangés sont en anglais.

Can only hear some notes in this instrument

  • Do 5 0
    Message de Don le
    Hello

    In the attached SF2 I'm only able to hear 0-40 with one layer, and 57-62 with both layers. I also get the same results with Viena SoundFont Editor. I will try Vienna and others to see if it is any better, but can hear most of the SF2 when played via BassMidi in Falcosoft Midi Player. Some help would be great.

    Thanks a lot
    Normal.sf2
  • BO 282 13
    Message de bottrop le
    delete the Sample end offset at the bottom of the Instrument pane'
    regards bottrop
    Normalbottrop.sf2
  • Do 5 0
    Message de Don le
    Appreciate it, yes I set the Sample End Offset from -xxxxxx to 0 and it works now. But what is Sample End and Start Offset and how to set it in Viena or Vienna ?

    Thanks again
  • BO 282 13
    Message de bottrop le
    if you use single samples (like you do) you dont need to set sample offsets, the complete sample will be played.
    there are people in this world who think it is better to have all samples in one file an pick parts setting start and end offsets.
    advice; stay with Polyphone.
  • Do 5 0
    Message de Don le
    I prefer dual samples for stereo instruments (this is not my soundfont) Yes but what is sample offset for and how to set that in Viena and Vienna ?
    Yes I like to have one soundfont or sf2 file store all the samples for an instrument too if that is what you mean. I'm not sure how start and end offsets help here.
    Sure but Viena has a nice look to it.
  • BO 282 13
    Message de bottrop le
    stereo in a soundfont is nonsense. if you have two ears you hear everything in stereo. but if you like to give yourself and your computer twice as much work as necesary, go ahead.
  • CS 73 0
    Message de csw900 le
    I suspect the offsets are referring to the samples and are a number added or subtracted to the sample number. Perhaps useful if your sample has a silent gap at the beginning. An offset could make it start to play when the sound really begins and not play the silent bit.

    I agree with Bottrop, forget about Viena (it has only one 'n'). It is very old and Polyphone works much better. However I do not agree with him about stereo. All musical instruments produce stereo sound and if you do not record stereo samples your sound font will be inferior to one which uses proper stereo samples.

    csw900
  • BO 282 13
    Message de bottrop le
    well CSW, i am always happy to learn something, so tell me how a flute produces a sound for my left ear and make sure that sound does not reach my right ear and at the same time produces a sound for my right ear and prevent my left ear from hearing it?
    regards bottrop
  • ZI 190 0
    Actually, bottrop is right on one hand.
    It is absurd to use the same mono sample separately for both left and right without any parameter adjustments. This means twice as much use of parameters. //plus: each sample you add is counted as a used parameter also.
    In the "sample end offset" sections, there is an attempt to set something up to avoid overlap because the same samples is used twice, but the attempt was unsuccessful.

    If you only have mono samples and you absolutely want to make something stereo looking, some touches can be used for this. one of these is available in the attached soundfont.
    Normal2.sf2
  • Do 5 0
    Message de Don le
    Stereo is not all nonsense in a soundfont. Well to me if an instrument has dual samples it sounds fuller and louder. I am with you that the computer and person will need to work twice as hard.

    This sounds about right as the number is slightly larger than the sample size not sure why it is negative. No way for me to tell if there is a silent gap anywhere in the samples. I held notes of the inaudible 41-56 and 63-127 for a more than a minute but no sound.

    Yes i do agree with some things, but I think you mean Vienna (with two 'n''s) is the very old one and Polyphone beats that. Not so sure on stereo it is a bit tricky. And any sound can be stereo but not on its own and it will need to output from 2 speakers. I guess that must be why stereo samples sound better than mono to my ears.

    I'm not 100 percent sure on this but I think the difference from mono to stereo is that the sound is equally heard in your left and right ears. Of course any sound is going to enter both your ears unless you shield or block your ears in some form or way.

    Yes, some points are sent across well.
    That is true and we can see in the Normal.sf2 file I uploaded each L sample is panned to the hard left with -50 and each R sample is panned to the hard right with +50 also. Yes more parameters but it is useful in here.
    I really have no clue what the "sample end offset" sections are trying to pull off, but it ends up canceling out the sound so that the attempt is not successful.

    Yes most of the time I'm stuck with mono samples only although I never recorded my own soundfont yet. The tricks used in the attached soundfont make it sound louder too.
  • BO 282 13
    Message de bottrop le
    here is the idea of offsets; three samples in one .wav file, all keys use this same .wav file

    keys 59-61
    for sample one start 1, end about 38108 (for some reason end offsets wont stick in my version (1.9), they turn to 0 after enter. i am not going to find out why because i am never going to use this stupid system.
    loop settings

    keys 62-64
    sample start 46464, end about 39000 etc.

    regards
    One4All.sf2
  • ZI 190 0
    If you have real Stereo samples, there are several situations that make them stereo.
    a. They were recorded with two microphones equidistant and separated by at least 1 or 1.5 meters.
    b. One was recorded from where the sound was made, and the other from a different place. // For guitar: one microphone is in front of the sound hole and the other is in the middle of the neck.
    c. One microphone was close and the other far-away.

    It is not recommended to set them as "-50/50" or exactly "L/R", except for the "a" option.
    Because in option "b": sound-hole/neck-sound mix will be made. // A value like "-12/12" or "-6/6" might hold this better.
    In the "c" option, it is aimed to adjust the far-near (timbre) setting. By increasing the attenuation of the near microphone and decreasing that of the other, it can be made to sound farther and vice versa. This setting adjusts the distance to the stage. With the L/R option here, it allows to adjust the right/left position of the place we stand in the hall in 3D.

    If both samples are exactly the same, then we cannot talk about stereo. There are two identical mono samples and it has no effect on stereo.

    In any case, the aim is to trick the brain in line with the desired purpose by giving different samples to the ear. : eg: a sound to the far right, whether the place where we listen to the guitar is near the sound hole or the handle, the sound reflection and absorption properties of the hall walls are different compared to the place where we listen in the hall, etc. This is why stereo samples sound fuller.

    Of course, it is also possible to ignore all the parameters of the auditorium, use the instruments as mono, and adjust and pan the other necessary effects externally during mixing.
  • Do 5 0
    Message de Don le
    This might be the first time I've heard or seen a .wav file with three notes or samples in it.

    middle sample
    I do not see any of this in Viena only Polyphone (I'm using Polyphone 2.1 so it may work in that) the number must be too small or large. Same here I do not get any benefit from this and still do not the use for that.

    first sample
    Not sure where these numbers come from

    I have seen stereo soundfont samples and they are usually red and yellow in Vienna.
    a. Yes this pretty much what I meant by the sound equally hitting the ears and is achieved this way.
    b. Interesting, I wonder how it is decided where the other place is going to be. I see it is always two locations the sound is picked up from as that is the nature of stereo and how it differs from mono.
    c. Yes I bet how far away matters but not too much.

    Let's hope every "-50/50" setting is for samples that are recorded as the way of "a" as you put it.
    Yes I've seen some perhaps stereo soundfonts that are set as "-12/12" or "-6/6" I guess they are like option "b" then

    In many free soundfonts like from my example the samples are exactly the same. Yes it is like 'doubled' mono sound and cannot come close to stereo.

    For sure the mind is played tricks on and that is how the effect of stereo is pulled off in most of the cases. Also, a sound to the far left will work too. Yes it is always worthwhile to know what parts of the instrument will be recorded. True the absorption and reflection of the sound is not gonna be the same when listening over disc. And that is what really makes stereo sound so much fuller and louder.

    I feel doing it in this way will miss out on a lot of useful things. As the parameters are what make this work so well. But if we adjust pan and others to make up for the lost it is fine.

Connectez-vous ou inscrivez-vous pour participer à la discussion.

Polyphone a besoin de vous !

Polyphone est gratuit mais il y a des coûts associés à son site web et à son développement. Un petit coup de pouce aidera beaucoup.

Faire un don
Apprenez les bases Voir le tutoriel
Haut de
page