Visualization on tiled displays: new software from the University of Texas at Austin

When in 2005 I saw for the first time a 55 monitor tiled diplay at the opening of Calit2 building, it changed my life. I realized that such displays combined with the massive cultural data sets that were becoming available (for example, Artstor) can offer fundamentally new opportunites for the study of cultural processes and dynamics. Two years later, with the suupport from Calit2's visionary leaders Larry Smarr and Ramesh Rao, together with Noah Wardrip-Fruin I established Software Studies Initiative. We focused on development of "software-based research methods and next generation cyberinfrastructrure tools and resources for the study of massive sets of visual cultural data, asking theoretical questions that are important for humanities."

In 2009 we developed the interactive visualization application for explorations of image collections which was running on 287 mexapixel display at Calit2, made from 70 30-inch displays. (The software is described in Yamaoka, S., Manovich, L., Douglass, J., Kuester, F., Cultural Analytics in Large-Scale Visualization Environments, IEEE Computer, 11/2011.)

Our app allows for interactive manipulation of thousands of images of any size. However this interactivity currently has a price - the sofware only works on multi-monitor dispay systems which run CGLX (A Cross-Platform Cluster Graphics Library); the development of new apps requires experience with CGLX.

Now Texas Advanced Computing Center at University of Texas at Austin released Most Pixels Ever: Cluster Edition (MostPixelsEverCE), a library for extending Processing sketches to multi-node tiled displays:

TACC Develops Visualization Software for Humanities Researchers

Most Pixels Ever: Cluster Edition

MostPixelsEverCE - software

Created by artists Casey Reas and Ben Fry, Processing is an established platform used by numerous artists and designers - a very important factor in the further development of humanities visualization. Artists and designers pioneered innovative visualizations of cultural data almost ten years ago (for example, Manyeyes, 2003). Later, Processing artist Daniel Shiffman developed the original MostPixelsEver library, which inspired University of Texas researchers to develop their software. (Daniel teaches at ITP in NYC and he recently published a book Naure of Code funded via Kickstarter).

We are looking forward to "sketching" with the new software on scalable multi-monitor displays walls at Calit2. While our latest walls have lower resolution than U. of Texas's amazing 328 megapixel Stallion, they use new monitors with very thin besels:

Tour (04-11-2012) 23
Lev Manovich demonstrating interactive explorations of image collections (168 paintings by Mark Rothko).