If you're not familiar with Moore's Law, it's the idea that computing speed will double roughly every two years due to factors such as smaller and smaller transistors being used in chips. That's great, but like with nearly anything else, there has to be a limit to the process:

If components are to continue shrinking, physicists must eventually code bits of information onto ever smaller particles. Smaller means faster in the microelectronic world, but physicists Lev Levitin and Tommaso Toffoli at Boston University in Massachusetts, have slapped a speed limit on computing, no matter how small the components get.

Physicists estimate that we won't reach this limit until the 75 year mark from now. But others have suggested that it's actually far sooner, only 20 years away. Heck, I've heard five years at some point. Whatever the time frame is, I'm curious about is what the actual speed limit will wind up being. (Mind you, I'd love to see it hit during my lifetime, too. Just to know for sure.) [Live Science]

Photo by Intel

It's just a concept piece for now, but the three metalheads behind the project seem like they're intent on selling some or all of it, some day. It uses an Arduino controller to process the input from the wigs accelerometers, then outputs it to a custom game written in the Processing language.

If some sort of jeweled headband controller doesn't find it's way into the next version of Rock Band, I will be disappointed. [Headbang Hero via Kotaku - Thanks, Mike]

Yes, it is final exam time for Shiffman's Big Screens Class—at 6PM on a Friday night, with free wine—and I am standing with a couple hundred other likeminded art techies in the lobby of the IAC Building, a curvy glass Frank Gehry creation on the West Side of Manhattan. We are in front of a 120-foot screen that's the equivalent of six 16:9 displays arranged end-to-end, and we are doing what it's telling us to do. We are obeying it.

It tells us to clap, and we clap. Then we stomp our feet and say "la la la." Then we send text messages to it, filled with the anticipation of influencing what appears on its glowing greatness. We clap to shoo white birds off a power line that's strung across its great length. We do it while drinking and taking pictures of the action, and it is good—a techie church for bigger screens, always bigger! We kneel!

Shiffman and his students have the IAC people, in part, to thank for their classroom. Rather than put in a garden or expansive, empty lobby, Barry Diller's IAC conglomerate—which owns several web-related businesses like Ask.com, Ticketmaster, etc—decided to build one of the world's biggest indoor video walls. It's made up of 27 vertically oriented projectors, linked into a single display by software from Spyder and shined onto a translucent screen to create a massive projection image:

For the Big Screens class, the wall is powered by three dual-head Mac Pros, each driving their own pair of 16:9 aspect-ratio screens (splitting nine projectors for each head), for a total resolution of 8160 x 768 pixels.

The class is part of of NYU's Interactive Telecommunications Program (ITP), a two-year graduate degree they've offered since 1979 and the source of all kinds of geeky curiosities. Shiffman, a wizard of the graphical programming language called Processing that many of the students use to fill up the screen (a few others use openFrameworks, another visual language) has taught this class for two years now. Processing has been used in tons of music videos, data visualizations and interactive video art and is popular for its relative simplicity as a way to turn code into amazing visuals.

Talking to the students, it's apparent that such a unique medium can barely be classified as a "screen" in the traditional sense. The immense size, when paired with such an extreme aspect ratio, turns the screen into more of a physical space than anything resembling a TV (even one that's 150-inches). Besides, it's not about resolution, in the home-theater sense. Sure, you can do a lot with 6 million pixels, but it's not why you come to see this 120-foot screen.

Interaction is the key, as you can see in the following videos. Mooshir Vahanvati created a massive 120-foot stretch of powerline with birds who perch when it's quiet and scatter when microphones pick up a loud noise:

Vikram Tank created a six-panel conductor that synced up the crowd's claps, snaps and la-la-las:

Matt Parker's "Caves of Wonder" took a video feed of the crowd from an IP camera and twisted it into a craggy landscape with Processing—part iTunes Visualizer, part Grand Canyon on Mars:

And Alejandro Abreu Theresa Ling combined silohouettes on screen with the shadows of real actors behind the screen to create three vignettes of Chelsea's seedier past:

Shiffman works the controls at the back of the room with a gigantic smile; he is perhaps the only person that could teach this class. He's the primary author of the "Most Pixels Ever" library for Processing, which allows projects to sync up across multiple displays seamlessly without delays—and not just your dual-head monitor. Most Pixels Ever is amazing because it can handle the 6 million pixels of IAC's video wall without blinking, and without it, this class would not exist in its current form. All the art-tech nerds thank him as we file out the door.

"For the students it's just such a completely unique experience—it's unique for anybody, whether you're a grad student or a professional designer. Few people in the world have a chance to work on anything of this scale, and what's great is that I can say to them you can do whatever you want," he says. "You learn a ton about technically producing the work, and also what it means visually to work on that scale."

"I can't imagine that when IAC build that wall that they imagined performances on it with actors casting shadows behind the screen, so that's fantastic."

The rest of ITP's classes are having their semester-ending show this week in NYC; find out more here and look for our coverage starting later this week. [ITP on the Big Screen]

The recording was made in 1949 by a student at a concert in Newark, N.J. When it was eventually found and played recently, the ancient magnetic wire had stretched and twisted and was so frail it broke often. It took 36 hours of work to just get the audio safely off the wire and into a computer, and even then the tracks were peppered with holes, slowed-down sound and missing high-frequencies.

By finding rhythmic sounds buried in the recording, and using mathematician Dr. Kevin Short's signal processing algorithms, the team carefully pieced together the tracks, interpolating holes and correcting for distortions and speed-shifts. The resulting album, The Live Wire, was nominated for the Best Historical Album category in the Grammys. You can listen to tracks showing just how nifty the processing was via the Science News link. [Science news and University of New Hampshire]