The Game Developer's Guide to Pwning Second Life

wjamesau writes "How do you create a game in Second Life that earns you thousands of dollars and scores you development deals with outside publishers? One SL user did just that last year with a casual game called Tringo (sort of multiplayer Tetris with gambling). The game became so popular in Second Life that he sold the rights for a Web version, a GBA port from Crave, and coming up, a TV game show. While there's dozens of other games in Second Life, from FPS to RTS to a mini-MMORPG, none of them have come close to Tringo's success. Kotaku is running an article I've written, based on three years helping Linden Lab organize and run the annual Second Life game developer contest: a how-to guide for creating the next Tringo-big hit."

I've made a few games in SL.... it's frustrating

[presearch]

I built one of the winning games in last year's SL contest, and several other moderately complex things as well.

The scripting language is interesting, fun, and somewhat well thought out. If you could use it to write someting that ran locally, you might be able to have something semi decent. but... After it goes thru the server system and out over the net intermixed with all that SL data using Linden Lab's lazy update protocol, you feel lucky to get things to work at all, ending up with everything a primitive compromise.

It's irritatingly flaky. The API calls are at best 80% reliable, terribly documented, and they come and go at the whim of Linden Labs with no standards a developer can rely on. Maybe object messages will work, maybe not. Maybe when a player shows up, all the parts will rez, maybe not. Maybe physics will work, usually not.

In 3 years, there's been no significant improvement in the graphics. It looks very dated, especially the avatars and what passes for a skybox. Everyone walks around like stiff zombies. It's still buggy as hell, especially if you have less than a 1MB line and a $4K PC. Get more than 10 people in the same place and it slows to a crawl for everyone. Can't complain though... Second Life zealots will tell you it's your fault because you don't have everything turned down to minimum settings.

The idea has potential, and Linden Lab has indeed solved some of the harder problems of implementing the Metaverse, but at this point, they just can't scale any further without it collapsing under it's own weight. Time to take what's been learned, pull the plug on Second Life, and build something with modern graphics, open standards, and distributed servers that anyone can run.

Although it's neat to move from sim to sim in the mainland , most "serious" players opt for a seperate island. That costs >$1200 startup and $200 a month.
I'd much rather be able to host my own sim, with a coordinated method of sharing comm channels, directory service, and inventory items with other hosts and players.

Again, so much potential, but at the rate they are going, it'll be dead in less than 2 years, a blip in computer history. I think the real goal of the game, and it *is* a game, is to pump up the number of user accounts, and squeeze in a couple thousand more users "in-world" to make it look attractive for some company to buy them up so Philip "Linden" and the verture guys can cash out.

Re:gambling

[Zerth]

Actually, it's kinda the other way around... With the cheapest items, you'll never have time to do anything else because your items/furniture won't fill up your need meters fast enough.