• Register
  • Help
Results 1 to 8 of 8

Topic: Z-Velo Sampling : How to use

Share/Bookmark
  1. #1

    Z-Velo Sampling : How to use

    I came up with this idea a couple of years ago . It's a very simple concept but very effective .

    When you programm a Soundfont or Giga instrument , usually you have one layer of samples covering a velocity range and the rest velocities are either automatically filtered or have different layers with their own LPF filter values .

    With Z-Velo , you can use the same samples on different velocity ranges but with altered pitches. You can even eliminate the "Machine Gun Effect" without having to buy a more expensive sound set that offers alternative samples .

    Below follows the original tutorial as i had made it back then , i have added some comments with *.
    I hope you find this trick helpful , it has helped quite a lot :

    --------------------------------------------------------------------

    "Z-Velo is a simple Layer Generator concept based on transposing
    Samples a few keys depending on the nature of the intrument.
    That means that we leave the Basic Layer unchanged and we transpose
    the upper layers higher and the lower ones lower, so the result is lower
    layers sounding warmer due to their stretching and the higher layers sounding
    "pointier" due to their compression. (*Layers that have close velocity values will give different samples for eliminating the Machine Gun )

    If a Soundfont (*Or Giga file, etc) already has more than one layers with *DIFFERENT SAMPLES* inside
    them and not just the same basic ones LP Filtered. The soundfont is better to be left alone , though you can follow a different pattern of transposing in order to get 2 ,3 or even 4 times more samples for each key .

    Warning: If a Soundfont/Gigafile is based on only a few samples
    ( or even one perhaps!) Z-velo Will not function at all if the transposal is not more than
    the keymapping that the samples use.

    A : By trasposing the layers up and down, what you get is the same samples
    as the original soundfile but with higher or lower playing rate and different
    keymapping. Which means that when you press a key with velocity that responds to LAYER 3 and the
    same key on LAYER 5, the one on layer 5 will sound more bright because it
    is transposed upwards, that means that a MIDDLE C on layer 3 could be a
    MIDDLE E on layer five (same sample) .

    HOW TO :
    Open your Soundfont/Giga file and choose it's main samples layer. Afterwards Select COPY and paste the layer
    in the same patch 4 times (or as many as you want).
    Now you have 5 layers all exactly the same ( * don't try playing with the instrument at this point, the DB's are gonna get huge ) .
    Choose layer 4 and all it's samples and perform a Global Sample move (move it
    more to the right on the keymapper),
    afterwards choose layer 5 and move it even MORE to the right.
    With layers 1 and 2 you will do the same but with negative values
    as shown below. We leave Layer 3 unchanged since it is our original samples.

    ! : Before Starting editing. Make a Copy of your file
    ( just in case things get messy

    $##########################################
    #----------------------------------------------------------
    #-----3 Velocity Layers Mapping Structure (Grammical)------
    #----------------------------------------------------------
    # 0 - 42 Layer 1 -
    # 43 - 84. Layer 2 0 = MIDDLE LAYER ( Unchanged Mapping)-
    # 85 - 127 Layer 3 +
    #----------------------------------------------------------
    #-----5 Velocity Layers Mapping Structure (Grammical)------
    #----------------------------------------------------------
    # 0 - 42 Layer 1 -
    # 43 - 63 Layer 2 -
    # 64 - 84 Layer 3 0 = MIDDLE LAYER (Unchanged Mapping)--
    # 85 - 105 Layer 4 +
    # 106 - 127 Layer 5 +
    #----------------------------------------------------------
    #-----7 Velocity Layers Mapping Structure (Grammical)------
    #----------------------------------------------------------
    # 0 - 18 Layer 1 -
    # 19 - 36 Layer 2 -
    # 37 - 54 Layer 3 -
    # 55 - 72 Layer 4 0 = MIDDLE LAYER (Unchanged Mapping)--
    # 73 - 90 Layer 5 +
    # 91 - 108 Layer 6 +
    # 109 - 127 Layer 7 +
    #----------------------------------------------------------
    $#########################################

    Transposing Examples :

    $######################################$
    #--------------------------------------#
    #-------Grand Piano (5 Layers)---------#
    #--------------------------------------#
    # Layer 1 @ - 4
    # Layer 2 @ - 2
    # Layer 3 @ 0
    # Layer 4 @ + 2
    # Layer 5 @ + 4
    #--------------------------------------#
    #-------Harpsichord (7 Layers)---------#
    #--------------------------------------#
    # Layer 1 @ - 3
    # Layer 2 @ - 2
    # Layer 3 @ - 1
    # Layer 4 @ 0
    # Layer 5 @ + 1
    # Layer 6 @ + 2
    # Layer 7 @ + 3
    #--------------------------------------#
    #-----Classical Guitar (5 Layers)------#
    #--------------------------------------#
    # Layer 1 @ - 2
    # Layer 2 @ - 1
    # Layer 3 @ 0
    # Layer 4 @ + 1
    # Layer 5 @ + 2
    #-----Pizzicato Strings (5 Layers)-----#
    # Layer 1 @ + 2
    # Layer 2 @ + 1
    # Layer 3 @ 0
    # Layer 4 @ + 1
    # Layer 5 @ + 2
    #--------------------------------------#
    $######################################$

    Q's & A's

    Q : 5 layers???? Doesn't that mean five times more the size?
    A : No. Five layers in Z-Velo means that we use the *same* samples 5 times. But it results in different samples per key at the end if it is done succesfuly

    Q : What about the LPF Filtering that the file used to have before
    i deleted the old Articulated layers?
    A : You can keep the LPF filtering. That makes things even better! But
    Remember that now the values must change since the layers have
    a brightness of their own without articulating them.

  2. #2

    Re: Z-Velo Sampling : How to use

    Picture :

    This is how a Single layered-chromatically sampled instrument will look after you have done all the editing :



    11 Layers with different sounding samples . The lower layers automatically play softer and warmer , the higher layer play brighter .


    Mp3 Info :

    Noise bursts seperate the "chapters" in the mp3 file .

    Noise 1 : Original file, Machine gun notes
    Noise 2 : Z-velo file , Machine gun notes -> No longer machine gun
    Noise 3 : Jingle Bells -> listen to the fast same notes playing .
    Noise 4 : Arpeggio with Z-velo -> No machine gun
    Noise 5 : Arpeggio with normal file -> Lots of machine gun

    Demo File - "ZVelo Ns Glocken"

    ( scroll to bottom of page to download the file, click on MP3 to download the high-bit rate 0.6Mb mp3)

    Good luck

    PS :In case you liked the sound :
    The original Glocken.Gig file was downloaded for free from http://www.worrasplace.com/ where a free Honky Tonk piano is also available .

  3. #3

    Re: Z-Velo Sampling : How to use

    Great idea! I think the result depends on the instruments' timbre though.

  4. #4

    Re: Z-Velo Sampling : How to use

    I'm glad you like it ! As you said about timbre, it usually works great with simple sounding instruments ( Non vibrato strings, piano, guitar, pizzicato, harp,glocken) for the vibrato and more expressive instruments it reduces a lot from their character

  5. #5

    Re: Z-Velo Sampling : How to use

    Is Kontakt allow users to build round robin range alterations for ADSR, etc... ?

    Each sample would'nt sound the same at all and memory+cpu+wallets could benefit of this.

    SergeD

  6. #6

    Re: Z-Velo Sampling : How to use

    Quote Originally Posted by Theodor
    I'm glad you like it ! As you said about timbre, it usually works great with simple sounding instruments ( Non vibrato strings, piano, guitar, pizzicato, harp,glocken) for the vibrato and more expressive instruments it reduces a lot from their character
    I agree. With pitching the vibrato speed would be altered as well, and that mostly sounds bad. I have some AKAI samples with only one layer (rare medieval instruments) that contain old harps, bells and lutes. One of these days I'll try it out on them.

  7. #7

    Re: Z-Velo Sampling : How to use

    I suggested this to Roland when they released the S770 sampler.
    You're actually doing two things at once:
    1. Shifting a layer so that, for example, the unity note is moved up a semitone
    2. Retuning that layer in the opposite direction, so that the layer is still in tune - but stretched slightly, so that only the timbre is changed.

    Depending on the direction of the stretch you get a warmer or thinner sound.

    I noticed the other day that Roland now have a knob in one of the Edirol modules which does just this. Simply turn it one way or the other and you get a very different timbre.

    It's a great way to avoid the machine gun effect without using more samples.
    What would make it better is if you could do it by increments smaller than a semitone (as with the Edirol knob, which is continuous).

  8. #8

    Re: Z-Velo Sampling : How to use

    Quote Originally Posted by Chadwick
    What would make it better is if you could do it by increments smaller than a semitone (as with the Edirol knob, which is continuous).

    Hi Chad ,

    How does the quoted work ? You re-pitch a sample with the knob to +25 cents upwards for example, but then to play a note you will have to downsample it again to -25 , won't you get exactly the same sample and colour again ? Do you mean the edirol does time-stretching with the knob ?

    Thanks

Go Back to forum

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •