Mixer, sends…?

I started work on overhauling how the audio routing happens inside HourGlass, in other words added a mixer object that can more sanely handle routing of the audio. Sending audio from the fragments into a separate effects chain appears to be more or less working. The nice thing about this is that it allows to do things like probabilistic processings for the fragment streams. (ie, 50% of the fragments can get processed with a reverb or delay etc, and 50% won’t…) The downside of this change is that it probably makes it necessary to also come up with a GUI for the mixing/routing, which may become a complex thing to design and code…

Sound example in link

http://soundcloud.com/xenakios/todnaksend2

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

5 Responses to Mixer, sends…?

  1. S-252 says:

    Cool program, bro! I’ve been thinking about another small enhancement: what if the Source Position could be spread randomly? So each grain would be randomly offset from the Source position set by the envelope. The higher the spread, the further randomly away the grains would be picked from the source position. At maximum, the randomness would pick anywhere in the full range between the Start and End points of the sample. At small values, it would pick grains only in a small area around the Source Position.

    This should create quite an interestingly varied carpet of sound, no?

    • xenakios says:

      You can actually achieve this already by using the fragment scripting.
      -Show the Fragment scripting pane (View->Fragment scripting)
      -Check the Enable fragment scripting checkbox
      -Copy paste the following line into the script text area :

      offset=offset+rand(1)*meta_a

      -Press the script pane Apply button

      Now the Script Meta A slider or envelope controls the amount of random spread of the Source position.
      This however only randomizes forwards to the current time, but I can think later of an alternative script text that does a plus/minus type of randomization too.

      • xenakios says:

        Here’s the script program for the plus/minus type randomization (replace the program line given in the above comment with this):

        range=meta_a*0.5;
        offset=offset+(-range+(range*2.0*rand(1)));

        You can change the 0.5 to something smaller or bigger in the first line to adjust the randomization range.

  2. S-252 says:

    Just tried it out, and it works. Awesome! Thank you for the tip. 🙂

  3. Skaven252 says:

    I played with HourGlass for hours yesterday. The randomness works great in two ways: First, you can randomize all over the sample to create a long “carpet of sound” from source material that’s too short in itself – for example, it can turn a splash of water into a river.

    Second, small randomization values can reduce “static waves” if the Source Position doesn’t change. So if you ‘freeze’ a voice, it still gets a bit of variety from the randomization. I took a sample of a jet engine revving up, and was able to make a “virtual jet engine” that could be revved up and down by shifting the Source Position. When the value was not adjusted, the randomization added variety to the sound so it didn’t sound ‘stale’.

    This is an awesome, and a very useful program. If developer further and interfaced with VST so that it could be hosted in a DAW (with automation support, so you can use the DAW’s envelopes too, maybe note-triggered pitch support with polyphony on top of autoplay), I’m positive many sound designers and musicians would be willing to pay for it.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s