Blog

Leading edge, bleeding edge, or just edgy?

Siggraph is enigmatic, sphinx like, all knowing, all seeing, and unknowable. Is it a trade show, an academic conference, a big R&D lab, a screening room, a job fair, an artist’s colony, or a gathering of the clan? Yes, Siggraph is. As such, it’s probably a little disappointing to some that it’s not more of whichever aspect they want. It’s ...

Robert Dow
This year we had a couple of gate crashers, William Shatner and Dick Van Dyke (Source: Neil Trevett)

Siggraph is enigmatic, sphinx like, all knowing, all seeing, and unknowable. Is it a trade show, an academic conference, a big R&D lab, a screening room, a job fair, an artist’s colony, or a gathering of the clan?

Yes, Siggraph is. As such, it’s probably a little disappointing to some that it’s not more of whichever aspect they want. It’s literally a shared experience, a happening, chaotic, spontaneous, shocking, surprising, sometimes loud, and annoying. It’s a mix of spiked neon hair, grey hair, no hair, and long shiny hair, but regardless of the costume and artifacts, no one who attends Siggraph is dumb, slow, or inarticulate. It’s the smartest group of people you can rub up against and randomly engage in an interesting conversation. Not at all like CES, NAB, or even E3. Siggraph is a gathering of scientists, really smart programmers, creatives, and big shots—and amazingly they all get along together and seem to seek out each other’s company.

At one company presentation we were told that there is no more magic in graphics, and that the last ten years have been a decade of maturity. No magic … Perhaps a poor choice of words, and no doubt the speaker meant that there don’t seem to be any major inflection points like texture or bump mapping on the horizon, and all we’re doing now is refining the known processes.

Not by a long shot. Yes, there will continue to be improvements in the techniques of today, but the never ending increase in processing power, lower costs, higher and faster storage capacities, and enhanced displays will fuel the imagination of the next generation of developers who will amaze and delight us, and sadly maybe leave us behind because we can’t quite figure out what they’ve done. This is the Singularity—its coming whether you like it (or its name) or not.

Algorithmics, systematics, and semicans

I’ve been thinking about some of the generations and given them names (an idea I got from Bubba Lombardi at THQ), and we’ve had three generations of developers: Algorithmics, Systematics, and Semicans.

From the seventies to the early nineties we had the Alogrithmics. The Blinns, Catmulls, and Whitteds to name a few of the founding fathers.

Then we had the systematics—the SGIs, E&Ss, and HPs. They gave us our first machines to work on that didn’t require a government agency to buy.

At the turn of the century we saw the emergence of the semicans, the Nvidias, ATIs, Intels, and a bunch of now gone innovators and explorers. These are the folks who think the Semi CAN, and will, change our lives—and they’re right.

The Semicans will give us the engines at ridiculously affordable prices which in turn will allow the Systematics to build boxes and displays that a new generation of Algorithmics will use to give us procedural AI simple UI tools for things like facial simulation and/or recognition, robotic control in a virtual rig or a hardwired machine, and natural language realtime translation conversations with peers and machine.

I used to think the future could be seen through the eyes of the Emerging Technologies curator or jury. This year they had 130 submissions and selected 26 of them, and you could have spent a day looking at and trying to understand them all. And there was no discernable (by me) theme—except that there was no theme. So new technologies, new explorations and developments, as exemplified by the Emerging Technologies demos, are chaotic, and borderline anarchic—as they should be. And, due to competitive and paranoiac attitudes of some companies, possibly even niftier stuff is labs at those organizations.

Leading edge or edgy?

So Siggraph is a peek at the future, though a fly’s eye—looking in 20 different directions at once. Matured, asym­ptotic in development, flat-lined—I don’t think so. You better put your seat belt on, we about to have the ride of our lives.