© 2020 Stefan Nussbaumer.
Creative Commons Lizenzvertrag

VideOSC – a (slightly un-planned) next alpha

Some of you who are reading these lines will possibly already have taken notice of my long-term amateur project VideOSC – an application for the Android mobile platform that allows experimentally-minded users to control music (or other media) through the color information coming from their device’s inbuilt camera. Already five years (or even a bit more) have passed since I wrote a first prototype on a weekend (using the great processing environment) .

Now, quite a lot has changed since that and one can truly say the world is not the one anymore that it was back then. My euphoria about my quickly hacked together toy has given way to a more thorough examination of my concept, and it has become evident that any new freedom in expression at the same time means new limitations. Not only is my time limited, it’s concept of the instrument itself (if I may dare to call my app one) that reveals a so far unrecognizable relationship between the acoustic and the visual, it also demonstrates our limited capabilities to keep control over the relationship between two orthogonally aligned spheres of our perception.

Ironically, in times where limitations imposed by a tiny virus hit the whole world hard, interest in my inventions seems to rise considerably. Yes, downloads have increased quite a lot during the past few weeks. And even better: I’m getting feedback from real humans. People are really testing my app. And, yes, of course there are bugs in VideOSC.

In this current case it was a user who left a comment in Google Play who had obviously observed that values coming from a specific pixel were not in sync with the visual appearance of the regarding pixel. In other words, the pixel changed its color, but the value stayed the same. Now, I have been thinking about the whole process of generating and sending OSC messages, respectively I’ve already had a new model for OSC communication in the pipeline. This should’ve gone in a larger update of VideOSC but that would’ve taken a considerable amount of time. So I’ve decided to put this new OSC communication model into a new release.

What’s new? Not much, at least nothing obvious. Instead of sending pixels’s values in one OSC message each, i.e. in a 35 pixel resolution 35 x 3 distinct messages were sent, OSC messages now get collected in bundles and for every frame in the stream only one bundle gets sent. I.e. at a frame rate of 15 frames per second only 15 x 3 bundles will be sent per second. That should reduce network traffic considerably. Moreover, for any new value a new OSC message is created, not like in previous versions where OSC messages were cleared and then filled up with a new value. I am somewhat optimistic that this has been the reason why values sent to the client were not updated properly – the messages weren’t cleared properly in VideOSC and hence the old value was sent over and over. This might cost a little bit CPU-wise but doesn’t create memory leaks which seams to me to be the most important aspect. Users should not experience any performance degradation.

So, if you’re curious (or have already installed VideOSC) get the latest update on Google Play or have a look at the updated documentation (that was another point of critique. Yes, that should’ve happened much earlier…).

Post a Comment

Your email is never published nor shared. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>