In other news....
Saturday 5th April 2003
Okay, so I've been busy lately. Sadly, the story isn't quite as long as the event was.
This is the diner.
Personae The Superchick. Leftenant of Gophertown PD's CARPS BetaUnit. Heavy Weapons Moron. Sergeant; CARPS BetaUnit Field Medic. Sergeant; CARPS BetaUnit Captain of CARPS BetaUnit; DoubleAgent for the Parasol Corporation The GoTard CEO of Parasol; mascot of GHB Ministries' FaithHealing Network Bioengineered Badguy; Parasol tool designed to eliminate CARPS Members Living Individual Simulation Android; biomechanical assassin for Parasol Head of GHB Ministries GHB's Little Lamb Himself Field Reporter for the National Prevaricator And so on. There are a few other characters, but I'm still working out who and what they are.
For the most part, I've been concentrating on the preproduction for Deadache. And there's a hell of a lot of that. Sometimes, I make the mistake of announcing that I'm going to do something before I've worked out how--or even if--I can do it. This has been one of those times.
On the bright side, I've worked out both how and if I can do it. Which is to say that I can.
The how part of it is a bit complex. I'm trying to make it less complex. And faster. Which is tricky, since making it faster is complex. It works like this.
The longest part of a project like this is setbuilding. There are some theoretical shortcuts you can take there. Like using stuff that's already been built. Which would be easy if those things had been built for a real programme. Unfortunately, most prebuilt sets are built for things like Poser5, Bryce, 3DSMax, and other toys. So now we have to add the step of converting all these prebuilt files to work in something a little faster. Like LightWave. And that can take longer than just building the damned thing in the first place.
It is, however, worth it.
Originally, this thing was a Poser5 file. Rendering the original in Poser5 at 2.4GHz took about ninety seconds a frame. Converting the thing to LightWave7 and rendering it on the same laptop cut the time to 1.4 seconds. This is good news.
Of course, the LightWave version is lacking a bit of detail, but that's not a big deal yet. Right now, the only texture in the LightWave version is the floor. And I haven't quite decided to keep that yet. Adding jpegs to the surfaces won't do much to the rendering times; it'll just make the thing look better. The important factor is that, once I've added a few people to the scene, each adding about a fifth of a second to the rendering time, I can film this thing at about two seconds per frame, which works out to roughly two seconds per minute for actual filming.
Which is to say that the final show, being filmed at 12FPS, will take less than ten hours to render from start to finish. To put this in perspective, it would take twice as long or more to film this thing on celluloid, or even DV, and edit out all the shit we don't need. That's not even factoring developing time for the film we're not using.
These numbers all rely on the show being filmed at 640*360, which is twice the size of a standard television. Mostly because computer screens are a hell of a lot better than most televisions. Filming at 320*180 would cut the ten hours to two hours, but I'm not quite thrilled with the look of it yet. At some point, I'll film a few minutes with enough objects per frame to make it interesting at both 640*360 and 320*180, smash each into .wmv, and see if there's a severe difference when playing them back fullscreened. If not, I might film this at 320*180 just to save eighty percent of the rendering time.
Meanwhile, I save the actual scenefiles for later. If this turns out to be worth watching, I can always refilm the entire thing at 1920*1080 for DVDs. That, of course, if a big If.
Which relies on other factors, of course. We know from Final Fantasy: The Spirits Within that the ability to produce a film entirely in CG is meaningless if the story sucks. So I'm trying to avoid that. To some degree.
Obviously, the story has to suck somewhat, or this won't be a parody of exactly that sort of thing. Engrish videogames. Which presents a new problem. Because all the characters in Resident Evil, Silent Hill, Grand Theft Auto, and so on are one-dimensional weenies who just happen to appear in three-dimensional renders. This is boring. And, ten million lusers writing Resident Evil SlashFic on the 'net notwithstanding, there's no way to write a story with them.
So the other half of preproduction involves working out who and what these characters are. And that's not exactly easy to do. Mostly because I have no frame of reference to begin with.
So I'm dropping the entire thing, and starting over from a better perspective. I'm eliminating the characters from the games, creating my own, giving them some sort of reason to live [or, in cases, to die], and then dropping them into the situations of the plotlines. And that's working a lot better. The only problem is that it sets us back a bit while I play deity and create these creatures from dirt. Or sand. Or silicon. Or whatever you use in the twenty-first century. You get the idea.
For those actually involved in this, the changes won't be that major. The names and vocations of the characters will remain the same; I'm just rethinking why they do what they do. For those not actually involved, here's the Dramatis Personae to date....
Dramatis
Jane Reynolds
Arnold Bush
Debbie Parker
Alec Bilderberg
Bucky Witherford III
Antagonist
LISA
The Reverend Doctor Lee T Roberts, Esquire
Katie
Youth Pastor Bill
Ace MacKenzie
The good news here is this: I'm consolidating a few things into this. Obviously, GodHatesBrains.com and NationalPrevaricator.com are becoming involved. Which is not necessarily to say that those sites will stop being what they were originally intended to be; it's to say that the original intentions take a lot of unintended time to deal with, and this is an easy way to give them some content relevant to more recent projects. They were never really meant to be taken seriously anyway; this just gives them a little more depth than they've had to date.
Meanwhile, I've got a few new URLs for this too. Oh--and one oldish one which the longstanding regulars to gremlin.net will like a lot. I'll incorporate those into the entire Deadache thing as we get closer to release. I don't know how successful this thing is going to be, but it's going to be big either way.
Somewhere between working out the characters and the graphics, I'll have a few dossiers made up sometime this month. I hope. I'm also concentrating on the elements of Gophertown needed for a fullmotion trailer; I'm hoping to have that online sometime this month too.
Which just about wraps up the Deadache thing for now.
In other news--in case you're wondering why this took two weeks to happen--there have been some other projects. Hunter's Daily Paranoid Delusion has been reworked. I modified the messageboard here at gremlin.net to work over there as a way to update the DPD; now it's a sort of cross between the board and these What's News: the updates are just plotted in through a Perlscript instead of written out in WordPad. Which is probably a bit lazy; but the end result is that the DPD should be updated roughly daily from now on. So that's a good thing.
And there are a few other things. But, like most of the things I do, they're not really ready to be announced yet. I'll get to it. One of these days.
More Later....
--Gremlin