New Orleans (day 3 of 7)

(This post has been backdated, from 2009-08-06.)

Tuesday was more productive than Monday. It's also more abbreviable, so let me first lay out SIGGRAPH's general schedule for clarity.

Every day has essentially four segments: I 8:30 to 10:15, II 10:30 to 12:15, III 1:45 to 3:30, and IV 3:45 to 5:30. Some sessions span I and II or III and IV; and there are dozens of smaller gatherings or sessions or activities that don't respect the segments listed above, such as FJORG! or the various BOFs, or the Computer Animation Festival showings, which mostly come after IV.

Anyway, for me, I was occupied by Real-Time Illumination for Dynamic Scenes, which was interesting and useful; it primarily focused on screen-space methods, which aren't of much use to me in my current job (we try for photorealism, although we only occasionally succeed) but at least had shiny algorithms.

Contrasting this, in II I attended a presentation of technical papers grouped together as Perception and Deception Depiction. I admit that toward evaluating lighting design interface paradigms for novice users was useful -- not in the particular, although I cede that the concept of a light-painting interface was interesting`{::}^1` -- but in the fact that it actually collected hard, reproducible data about HCI, which hardly ever seems to happen, and that it laid out its method and methodology for collecting the data. However, I still probably shouldn't have attended, since it was a Technical Papers presentation-collection, and those are all included on the DVD I got with the full-conference admission ticket. (Well, that my employer got with my ticket, anyway.)

So leaving that session early, somewhere in the middle of evaluating human color perception in high-luminance conditions, meant I got to head to the Exhibition Hall and collect swag for the rest of II and the lunch break. Alas, there was little swag to collect -- or, from a less self-centered position, the exhibition hall was strangely small compared to last year's SIGGRAPH, with fewer interesting exhibitors. (Also those exhibitors that were present were much more aggressive about scanning the barcodes of my badge without first asking permission to flood me with their e-mail advertisements.)

... also I had what purported to be a muffaletta from the exhibition hall food dispensary. I have officially learned my lesson concerning food from within the convention-hall (that lesson being DON'T).

III and IV were taken up by a course on OpenGL, updated to use OpenGL 3.1. Alas, I am not sure what its intended audience was: the OpenGL 3.1 material did not make up much of the course, that being instead almost entirely taken up by a rehash of my college Computer Graphics course -- very basic 'how rendering works' stuff. I thought it was intended for artists who wanted to get into programming, but an one such artist considered it beneath him, and a second artist (less interested in programming) found the entire talk to have been over his head. I'm still not sure what the intended audience of the talk was. It may have been a set of people unlikely to attend SIGGRAPH.

---

`{::}^1` However, the primary conclusion of the paper was that 'interesting' doesn't imply 'useful' here, so don't get too enraptured with the concept. Even manually positioning the lights turned out to be more usable than that.