Monday, November 25, 2013

Dubstep Vortex

Authors:
52919718 CHANG Weifan
51847192 LAU Chun Man



(EDIT: Cannot imbed Vimeo Link. Replaced with Youtube link.)

Vimeo link can still be found here.


Dubstep Vortex is an audiovisual feedback installation. 

However, unlike most other video installations of its kind (in which a video signal simply records itself in an "infinite" loop), Dubstep Vortex’s augmented feedback system—allows not just a greater deal of customization and flexibility for the artist, but also offers an unprecedented degree of interactivity with the audience. 



Instead of just having a camera record its own display, by first rerouting the captured information through a computer (CPU), we are now able to effectively alter the data stream before it is redisplayed.  Prior to this, the only way to alter the stream was to physically interfere with the light signal between the camera and display; with the inclusion of the CPU, more complex maneuvers, such as audiovisual presentations, are now possible. 



Dubstep Vortex takes advantage of Windows Media Player’s in-built visualization engine.  The entire visualization window is made semitransparent, and is layered atop the camera view.  The result is that the audio visualizations become part of the feedback loop, with spectacular results.  Due to the fact that everything is semi-transparent, the screen becomes a pulsating mosaic of various opacities, dancing in tune to the music.



Using a computer also means that the audience is able to easily make adjustments to the setup that can drastically change the style and look of the feedback loop.  Not only can audiences choose what song to play out of the playlist (or even adding their own), but they can also choose (at any time) what kind of visualization they wish to use—they can even easily adjust the camera’s shutter speed and color temperature (via camera real-time display program). 



To make the setup even more unique, the camera is physically tilted at an angle—not only does this make each subsequent loop more and more “off-kilter”, but it also creates unique patterns of overlap unlike any other.   


Lau Chun Man stands next to the camera in our revolutionary new feedback loop setup, in the video art installation "Dubstep Loop"


The new "augmented" loop comprises of: Camera, Computer, and Projector/Screen


The loop warps in real-time, in accordance to the musical cues provided by WMP's visualization engine.  

Sunday, November 24, 2013

Tip of the Iceberg: Technical Difficulties

Now that our concept was finally ready, Edmund and I began the slow process of fine-tuning the system and procuring the various components necessary for its implementation.  Almost immediately, we learned that, although the saying goes that "the idea is half the journey", turning an idea into reality was a lot harder than it looked.  While some troubleshooting an occasional technical bug is inevitable, some of the problems that we ended up running into sometimes made us wonder whether we angered the computer gods.

In fact, problems seemed to plague us from the start.

Most of the initial difficulties were inherent issues that had to be resolved due to the novelty of our design. While adding a new component into a system doesn't sound inherently complicated, unlike the modal design of present-day computers (with plug-play USB devices), we were essentially rewriting the structural framework, transforming a duo-level system into a triple-tier infrastructure (comparable to making the jump from 32-bit to 64-bit; and how 64-bit programs are not reverse-compatible).

The first challenge, therefore, was to somehow link the camera's screen--not to a projector, but to a computer screen.  Modern-day computers do not readily support analog, so we were unable to use the DV cameras provided; instead, we had to go back and use the old, reliable 5D Mk II, which had HDMI and USB support.  Edmund and I also found several programs available online that allowed us to see a camera's display in realtime--and also allow audiences to easily adjust camera attributes (such as shutter speed, focus, and color temperature).


The second issue we had to deal with was the audiovisual presentation.  Originally, I wanted to use After Effects to create visual images that were powered by music scores.  I had already done extensive research into how to link audio keyframes to modifiable attributes via expressions; when I calculated that I did not have enough time to create the audiovisual images I initially wanted.  Undeterred, I found a quick solution--use the in-built frequency visualization in Windows Media Player.  While not as glamorous as I had initially envisioned, both Edmund and I felt that it was adequate to do the job, and that "the simplest solution is sometimes the best."  Not only did we find that WMP came with an entire library of visualization styles (audiences can choose midscene what kind of visualization to use), using something that is familiar will only make the feedback distortion all the more interesting.



Another design hurdle we had to cross was to somehow display the audiovisual stream without breaking the feedback loop.  If the audiovisual content is not transparent, it will effectively block the feedback (because one cannot see the recording screen!)  After a long and extensive search, I found a small dusty little program called Vitrite that allowed users to primitively tweak any windowed program's opacity, in increments of 10%, and this adds flexibility to the video installation in terms of user-friendliness (audiences can now choose how strong the audiovisual information is in the feedback stream.)  


These problems encountered in the pre-assembly design of the installation were not particularly distressing; they were annoying, troublesome, and took time to figure/find a way around, but they didn't make us want to pull out our hair.  On the other hand, once we began assembly, we ended up screaming like lunatics in an asylum out of frustration.  Here were some of the hilarious problems we encountered (and the drastic solutions we had to come up with)


  1. The cable lines were too short.  As a result, the physical proximity of the camera to the projector screen was too close for any meaningful feedback.  
    • Solution: Jury-rig with our own wires.
  2. The audiovisuals were not visible in the feedback loop
    • Solution: Tilt the camera slightly, so that each "loop" deviates away and fills the entire screen.
  3. Halfway through setting up, we were kicked out of the room.
    • Solution: Find another room.
  4. The computer could not detect the camera.  
    • Solution: Give up and Die.  Just kidding.  We searched the internet, deleted computer drivers, and when that didn't work, started messing with/deleting computer registry entries.  Giving up is for pussies.  


Saturday, November 23, 2013

The Feedback Loop

After closely analyzing the handout from class, Edmund and I have come to focus on a second topic for our video installation:

The Feedback Loop.

The simple idea is to make a camera/capturing device recapture its own output, thus creating an endless "loop" of the same signal.  This signal loop can be adjusted with the camera's shutter speed, physical proxmity and white balance, creating loops of different colors, strobes, and depth.

While we did have hands-on experience on how to create simple feedback loops in class, Edmund and I felt that to make a video installation based on the concept of a simple feedback loop would be too generic (as we were certain that most groups would be doing more or less the same thing).

The original feedback loop cycle, in its most simplistic form, is comprised of two components.



The only problem with this simple feedback system is that interactivity is at a minimum, in which the most audiences can do is to alter the settings on the camera or stand between the capture/display devices (to distort the signals).

And thus, herein, was our greatest challenge: the concept of a sound-driven installation was still VERY stuck in our heads.


After a long, constructive discussion, Edmund and I felt that we could try to create an amalgam of the two concepts by slightly tweaking the feedback loop process.


Original/Basic Loop



Revised/Augmented Loop


Not only does the addition of the processor unit open a completely new level of possibilities, the inclusion of a computer also allows greater flexibility and ease of access for audience interactivity.

With this new additional step to the feedback loop, we can now create a sound-driven video feedback loop.

Friday, November 22, 2013

On the Shoulders of Giants... Basic Ideas...

Group Members:
52919718 CHANG Weifan
51847192 LAU Chun Man



From the beginning, we were heavily inspired by the Japanese video artist that Rita showed us in class; as a result, we were determined to make sound an integral part--the critical driving force, of the installation.

The two art pieces by the said artist demonstrated (and confirmed) that audio is a very powerful component of a multimedia experience.  (I remember having my jaw wide open during the entire presentation)  



Plan A (Immediate plan upon leaving the classroom): create a video installation with CGI particles/turbulence powered by sound, similar to the opening sequence of my previous film work: The Gamer.


I have experience in After Effects, so while I knew how to let audio waves "power" certain effects,  the only challenge was to somehow connect all attributes; from the behavior of particles to the animation speed of turbulence systems — to the amplitude of sound waves.