Ever walk into a room and get stuck on a table? No? How about in a video game? If so, Paul Wedgwood has a solution for you.

Yesterday, I posted about the greatest feeling in video games as determined by Splash Damage's Paul Wedgwood. I explained how he hopes his studio's next game, the team-based first-person shooter Brink will deliver more of it.

But when I interviewed Wedgwood at QuakeCon 2009 last week, he also told me about something that bugs him.

We were sitting at the far end of a hotel meeting room in Dallas that contained a long table surrounded by chairs. Wedgwood got out of his chair and approached the furniture to explain what I'd call the Table Problem.

"If I walk up to a table and the level-designer made it an inch higher than I'm able to jump, that's it, I can't get over the table," he said. He and I sized up the table he was standing near. It wasn't that tall.

"So even though I'm 200 pounds — maybe 210 and somewhat chubby — I can vault that table if I ran at it right now and get across that table," he said. I agreed with him, but hoped he wouldn't try it. There wasn't much room for a running start, and who knows how much weight a hotel table can support.

"Walking into this room, I know to avoid the tables and chairs," he said, reasonably. "I don't get stuck on geometry. But in a shooter, I can't see my feet. Even if I rendered the whole model, I can't see my feet in this view I'm using." As a result, people get stuck bumping into things in shooters that they'd never walk into in real life.

Well, Wedgwood first shared what the old solution has been: "What level designers do to get around this is having boardrooms that don't have furniture in them. In multiplayer games you have to have super-smooth clipped routes."

And if there's something to climb, maybe a wall or a table that's in the room to vault, a game programmer inserts what Wedgwood called an "entity," a programming instruction that produces a signal to players that that given wall or table can be clambered over. "If he forgets to put an entity there — or if the designer didn't want you to [climb] — you suddenly have an invisible barrier. And, bam, your immersion's gone. You're out of the game because you find that so frustrating. Why can't I climb up that wall, because the icon shows up, but not that one? Worse still: When you hit the button, you enter a canned animation until you get to the top. And that's it, done."

Wedgwood was exercised about this and described to me what Brink would do different. "We wanted a system that was real-time, dynamic, blended animations, full trace of the geometry around you, not faked, not clutched. In other words, if I decided that I'm going to mantle up that wall, if it's a height I could climb or reasonably jump to, I can, irrespective of what a level designer wants. If it's there, I need to be able to climb it. And, as I'm climbing it, as my first hand comes free, I want to be able to start shooting. As my second hand comes free, I want to be able to start re-loading. If I want to stop and take my finger off the button, I want to drop back down to the floor. If, as I'm dropping I hit jump, I want to kick away from the wall. It must be a completely dynamic, fluid system. It's not on auto-pilot, but it is smart, which is handy because it stands for Smooth Movement Across Random Terrain [laughs]."

Credit Bethesda marketing for the acronym, he noted.

And he continued with what sounded like a furthering of the freedom of movement given to first-person gamers in last year's Mirror's Edge:

"The idea behind this is that, when I want to get where I'm going to, I'm holding down my sprint button, because I'm going there as quick as I can. At that point, I've given the game permission to interpret, intuitively what I want to do, which is to vault, step up, jump or do whatever. But I want some control over that. So, if I look up, I want you to go up the route that mantles me over something. And, if I look down, I want you to slide me underneath it. So, if I hit a security center, I can make that choice in real time. I'm steering, I'm turning. And, if I let go, I stop what I'm doing at any given point.

"So, all it does is solve the problem of you not seeing your feet in an intuitive way that would work if you walked into this room, which is why you didn't bump into any of the chairs or tables, because you can see them because they're in your view. And this is great because our level designers can now flood our maps with tons of crap that makes it feel much more realistic. And the little routes and things are not based on complex player-clipping. And that just makes a huge difference."

I saw some of this, particularly the player-character wall-scrambling, with my own eyes when Wedgwood played the game live on stage at QuakeCon the next day. I can't vouch for the nuances he described about free hands being used to do things the moment they are free.

But I can vouch for the real Paul Wedgwood not bumping into any of the furniture in the Dallas hotel room. He avoided all of it.