• Advertisement


  • Content count

  • Joined

  • Last visited

Community Reputation

134 Neutral

About Mull

  • Rank
  1. AI in MMORPGs

    Quote:Original post by WeirdoFu Let's just assume that all low level npcs that are killed spawn from some sort of boot camp, which teaches them strategies. So, everytime you kill something, this "boot camp" learns something new to teach to the next generation. So, technically, as the game "ages" the NPCs will become more and more experienced. So, here comes the game balance issue. This is something I've considered. Do you think it's conceivable that training the agents based on their observations/experiences may be counterproductive? Initially many MMO*s suffer from a particularly steep learning curve. Having agents that learn from what are essentially "dumb" players that are muddling through their first experiences of a game may result in less than optimal training. Using one of my early MMO experiences as an example, the space-sim Jumpgate, the learning curve was incredibly steep and people often died to the very basic AI (not to mention splashed themselves against asteroids and stations thanks to the pseudo-Newtonian physics engine). Would training the agents based on this sort of behaviour result in them learning bad strategies, especially as the player's own skill improves quickly? Andy
  2. Begining AI Programing

    I'll happily plug Mat's books but then I'm a shameless fanboy ;) I started getting interested in AI through writing some simple artificial life apps years back. That got me interested in closed systems, behaviour and how simple rules can create (cue buzz-word...) emergence. If you've an interest in biology I'd recommend reading around a-life as a way of piquing your interest. But yes, it's a huge area to "get started" in - I'd recommend getting a good book that'll ease you into the subject area. Mat's books are good and I'd also recommend Alex Champandard's "AI Game Development", especially if you're interested in AI from a first-person-shooter perspective. Andy
  3. Ai Programming

    Matt's book is pretty good as well and you have access to the source code for the examples in the book. If you want to avoid writing the game and get more involved in modifying the logic behind its AI, I'd recommend picking it up and having a play.
  4. Halo 2

    I agree with the points on gameplay. Having spawn points and triggers that are carefully played by the level designers can serve to provide (as pointed out) respite from attacks and a way to drive the player through the game level. However, Quote:well there's actually no way around having triggers that spawn enemies, especially on console platforms. I disagree. Despite processor and memory constraints, there is no reason whatsoever that a game can't consist of a world populated by autonomous agents, each with their own behaviours and constrained by the same mechanics and physics that the players are limited by. I say this because it's pretty much the way the AI behaves in the game our team is developing. Game agents are autonomous, scripted entities that follow the same rules (down to what packets they send to the server) that the game clients do. This model suits itself more to MMO games. I'm hoping to see more and more of these sort of techniques used in games in the future. One game that I recall using something similar is Saga of Ryzom where, if I recall correctly, large groups of agents wander in packs (herbivores, etc) through the plains and grasslands of the game. -- Andy
  5. Quote:Original post by johnnyBravo I actually played the beta of that, it was pretty good at first, but the physics although let you drift, had air resistance, so you would slow down after a bit. Which did you play? I spent a few years playing Jumpgate and still love it. Played Vendetta recently and I've got to be honest, I don't really like it. Without going completely off-topic and getting this moved to Game Design, what's your (no pun intended) resistance to drag? Our engine implements drag and a capped max velocity, both for gameplay reasons. Personally, I love true physics but (and our team debated this for a long time) for a commercial MMO, the barrier to adoption for new players is WAY higher if you don't implement some form of assisted physics. Jumpgate has a steep learning curve and a lot of the deaths new players experience are down to drifting into asteroids and the like. edit: fixed tags :)
  6. MMO space sims that are current and active (sort of...) include: Jumpgate EU or US Vendetta I'm not sure about SWG's Jump to Lightspeed. It's definitely twitch but I'm not sure about the physics. BlackMoons SpaceSim Oh, and ours. But that's classified ;) edit: Added BlackMoons sim in development. [Edited by - Mull on November 6, 2004 9:27:32 AM]
  7. It arrived today. A quick skim-read and it looks pretty good: a lot more dense than "AI Techniques..." and I just love UML diagrams ;) Just wondering if fup modelled for the wireframe bloke on the front cover... -- Andy
  8. Looks good fup. I enjoyed "AI Techniques for Game Programming" and just sneaked the new book onto the missus' latest Amazon shopping spree ;) Good luck with it! -- Andy
  9. Are you a member of IGDA?

    Yep, two on our team are paid up members (second year running). Heck, it's worth it for GD magazine (not the easiest mag to get around here) and the funky membership card :) Never made it to a chapter meeting yet though.
  • Advertisement