Crunchy demos

It's been a long time- I don't even know where to start, so we'll do this in a categorized rather than chronological order...

1. The Demo: My last blog was posted a day before my birthday, on September 19th, and during that time, I've accrued a team of very talented and dedicated guys and girls who are putting equally nice artwork into the game. Using contacts that I made at Austin GDC, I'm also able to bring motion-capture animation into the game through Mixamo. Having met a really good concept artist in Austin, I was also able to create and distribute a thousand fliers at VGXPO in Philly three weeks ago, bringing in a modest bump in web traffic. This effort will be replicated for the various Comic Cons that roll through New York next year, each one bringing in thousands of people (as opposed to VGXPO which brought in maybe one or two thousand in total).

At the same time, I've moved the database to a test server- an XP machine that sits on my desktop and does nothing else, except maybe drip liquid coolant onto the RAID array now and then (that reminds me, I need a CPU fan). We've had team members from as far away as Ireland log into the game and test while, ironically, one of our guys just 50 blocks south of me in Brooklyn wasn't able to get on. This allowed me to crush a bunch of bugs that popped up only under online conditions.

Next, we've started moving the behavioral AI into the game and testing that, which continues today. We have the beginnings of a rotating overhead mini-map (not ready as a resource yet), and we've almost solved the mouselook-and-select-objects issues that dog many people (that is also not yet ready as a resource- it's kind of spongy right now). Drag and drop functionality is back in and we'll be adding some crafting recipes and materials this week coming up, and mission generation is also coming online.

And finally... Outsourcing. Because of the time-crunch and the failure to get as many 3D artists as I would have liked, I've resorted to outsourcing some tasks. It's not because I have deep pockets, but because the demo tasks need to get done. In any event, I have some advice for would-be outsourcing/contracting people reading this: Don't promise anything you don't know how to deliver. I'm not about to mention names, because mistakes become past mistakes quickly and though I wound up canning the contractor, it doesn't mean their work doesn't get better. In fact, the modeling and texturing was done very well, but the client doesn't need to hear about your internal problems. Be a black box for me, would you? Thanks.

So, we're in the final two-week-stretch here, and I'm starting to work from about 9am to about 1am every night, with a few hours dedicated to my girlfriend most nights, but otherwise huddled over the laptop beating my head against walls until they crumble. And far from technical, I need to work on budgets and project timelines for the prospective publisher as well- they do like to know how much money and time a project will cost them. If any of you wannabe MMO creators out there hate business, then you better learn to love it, and fast.


2. AI: What can I say? Interrogative is no more. In fact, it is more. Having started to morph from a feature for the MMO into something that can be used generically for games, into something that I'm intending to throw at the Turing Test competition, and even beyond (shhhh), the name Interrogative is no longer proper. Hierarchical opinion-based knowledge representation systems using generalized pre-compiled dialog along with dynamically-created dialog with interrogative and natural language processing front-ends is no longer what Interrogative is or was. It's not even tested yet, but the theories are there, and having run it past a number of industry heavies, I can confidently say that I'm on the right track with my experiments.

At the lower level of things, Interrogative is alive and well in Epic Frontiers, and we have about a dozen new conversation keywords and dialog sets to throw into the system for testing. After that, proper changes will be made to support the Infidel opinion-based architecture we want to implement, which will allow the game to support the players changing the minds of AI, as well as AI reactions to players and other AI who do not share their own opinions (hence the name Infidel, which means "non-believer" and can be applied to anyone who does not share your belief in whatever subject you feel passionately about).

With all this AI tech springing to mind, I'm thinking about spinning off a company to hold all that technology, so that I can continue to develop it without interference from video game contracts and such...


3. GameX Summit: Well, the last conference I attended for the year was just great! Met up with Dave Mark, Mike Worth, and many others and had a great time. For a first-time summit, the mistakes were minor, and the recovery from them pretty swift.

Unfortunately for me, the students that showed up were almost all design students, and not artists, as I was hoping, for recruitment sake. That said, I also met a lot more people interested in AI, and attended a couple of mind-blowing sessions by people such as Clint Hocking, Dave Mark and Kevin Dill, Damian Isla, and Mike Worth, whose audio session simply rocked.

The only real downside to the conference was getting lost. I don't know what it is about the Pennsylvania hillside, but I'll be damned if I didn't wind up in the Valley Forge National Park at 11:30pm one night staring at over twenty deer in my headlights. And it wasn't that "ZOMG what kind of car is that" stare, but the "dude, this park belongs to us after dark" stare... Interesting area. But the food out there is pretty good.

That's about it for the blog for now. If I don't manage to get another blog out by the 15th, then my next will be a post-mortem of the Epic Frontiers demo. Back to the grind...