Oh, my, i23 is out, and it's time to pay the piper, Arcanaville!

It took a couple of hours to test things and set them up and then record them, and I didn't want someone walking into frame and forcing me to reshoot something. Even if they don't walk into frame, walking past can trigger all sorts of dialog from NPCs that generates pop up text which can also mess things up. I basically wanted an empty Pocket D to video in. And who goes to D on test in the middle of the night?

Pretty much as soon as I was pointed to the thread, which is why I tried to get onto the beta server over the weekend it was down.

My first thought was a screencap, kind of like the screencap I posted. But then I realized that it would be difficult to properly convey what was going on in a still frame. I could add thought or speech bubbles, but I'm not fond of those in general. A short video sequence of that screencap to convey motion made more sense, but then I started wondering if that was enough to convey the entire story of what was going on. So I scripted a sequence involving walking up to Null, the "why me" emote, and then the worship emote. At that point I still wasn't sure how to convey Laz in there, but I figured I would figure it out when I had a chance to get on the server and experiment. I shot a bunch of sequences on Monday night, but I decided it just wasn't right without Laz so I asked him for his costume file, and then reshot it on Tuesday night and encoded and uploaded on Wednesday morning.

If the servers were up that weekend, I might have taken a bit more time, but I decided by Monday that it would be a one-day thing to get it out quickly, so there's two big glitches. One or two segments have mouse cursors visible: that's an artifact of mixing up some test footage and the final version. And when Laz ports in, I should be in the why-me stance and not standing, but I filmed that out of sequence so I made a continuity glitch. There's also a moderate editing error: Laz "glares" about two seconds too long: I made a mistake on the edit marker there.

All told, its about two hours of goofing around and trying to find the right angles and emotes, and editing down some, and then about twenty minutes of shooting it, cutting out the segments I wanted, and then splicing it together and encoding it for upload. I wasn't exactly sure how long I needed in advance, but that's a long time to hope a live server's Pocket D will be completely vacant without even a passing-through traveler.

Someone used to give me free copies of Premiere (as in: a friend who worked at Adobe, not someone who pirated the thing) but since I don't use it professionally (or at all) I never bothered to load it on my current platform, nor can I find the last copy I got to load it. So what you see is actually done with Fraps to record, and then Virtualdub to mark out and save the segments I want, crop and splice them together, and then rerender into divx for upload to youtube.

How do you get a picture of an AI? Virtual composite male/female? Pic of a mainframe computer? Since we all know that Arcanaville is an unchained AI and not a fleshy. And, oddly enough, one who has caught the ire of Mod9. Giving away too many secrets Arcana?

Fuegocito