Rampant Coyote posed a question which Shamus then picked up, “What makes a [computer] RPG great?”
I responded in Shamus’ comments with something to the effect of, “We can’t know.” — Yeah, a cop out, but a true one.
A lot of the other responses seem to be geared towards, “I want a world with real choices.” Although I tend to enjoy freestyle games like that, the more open-ended the game the weaker the story usually ends up being. That, to me, tends to make me lose interest. For example, I loved Fallout, but I’ll never finish it. Same with Baldur’s Gate. These games are too long, and their main storyline is too removed from the actual play of the game for me to be able to see them through.
I was also on the phone for about an hour today with one of my acquaintances, a guy who pretty much lives and breathes computerized RPGs. Although this guy is primarily a player, he also envisions himself as a designer/developer/programmer. When I posed the question to him he responded with how he was using ELIZA scripts to modify NPC behavior in Neverwinter Nights to create truly dynamic NPCs. Though I’m more than a little bit skeptical of treating ELIZA as anything more than the facade of intelligence, the approach has merit in at least creating the illusion of dynamism in NPCs.
From that point he started telling me about how he wants to simulate internal states, likes and dislikes, and all this other material relating to NPCs — This is done with The Sims, so is not impossible, nor is it a bad idea. Then we got on about how a game should process speech, and NPCs should react to your speech as if you were actually talking to them. And then we got on the subject of NPCs reacting to speech they aren’t programmed to be aware of, like talking about a “revolver” in a game world where that technology doesn’t exist. It was posited that NPCs wouldn’t have knowledge of that device, but that in a truly great RPG you could assemble the parts in the proper way to create a revolver, or draw one, or any number of things, and teach NPCs what revolvers are.
In other words, we need NPCs with internally defined personalities, with a robust method of generating language, a spectacular text-to-speech interface with an all-encompassing dictionary, a dynamic intelligence for NPCs, and physical laws of the game world defined down to the molecular level (or below). This is the danger I see in thinking greatness is necessarily an aspect of “choice.” Where does that end? By this sort of understanding, there have been no great RPGs. In fact, the term RPG might as well be replaced by “alternate life simulator.”
Although the category of RPG is vague and problematic enough to begin with, we shouldn’t keep moving the goalposts until the term essentially requires the creation of an alternate universe. Though we might make it there eventually, with truly sentient AI and lifelike visuals using a completely transparent interface, I wouldn’t expect that in the next twenty years. The programming obstacles are pretty great in a number of areas, and the art-asset creation time itself will be huge (though, theoretically, we may have the computers creating the art assets in the future, or at least doing the heavy lifting). For the time being there are some great games being made — I just don’t know why they’re RPGs, what that is, or what makes them great other than being really enjoyable experiences.
Ignore the ensuing epistemological crisis.
This is why, I would argue, MMOs are so popular. This is an approach to the “Why bother writing a program to do all this, when you could just have a PC playing an NPC?” question.
Amusing: the captcha is “human”