• Advertisement

Tony Li

  • Content count

  • Joined

  • Last visited

Community Reputation

22 Neutral

About Tony Li

  • Rank

Personal Information


  • Twitter
  1. R&D AI team vs Player team

    I was thinking of mentioning Behavioral Mathematics in my earlier post. Dave Mark's book is a great resource for devising efficient utility functions, among many other things, because it really all comes down to math. It sounds like you want the AI to act as a team, so you'll probably want to read up on multi-tiered decision-making in which each agent makes some decisions but they also coordinate their actions as a team, which is a larger architectural issue.
  2. R&D AI team vs Player team

    A good utility function will help the AI team prioritize threats. Check David Graham's Introduction to Utility Theory from Game AI Pro. If you can get your hands on any of the AI Game Programming Wisdom books, they were published in the heyday of squad-based shooters. They're the single best resource for this kind of AI. They have entire chapters on tactical AI and group actions. You can search up PDF versions of the books, but I don't know if they're legal or pirated scans, so I won't provide any links. If you're affiliated with a university, you can probably get legal access to digital or print versions through your school.
  3. Or a space between "XP" and "Counter", as in "XP Counter". It's always a good idea to check for null whenever you use a function that can return a null value, such as GameObject.Find or MonoBehaviour.GetComponentInChildren. It's part of an approach called defensive programming, and it can help avoid a lot of hard-to-find bugs. As a side note, since you're using the Dialogue System, there might be an easier way to implement the quest than replacing the entire NPC. When the player picks up a fish, you can set a Dialogue System variable. For example, in a C# script: DialogueLua.SetVariable("HasFish", true); Use a single conversation on the NPC, but branch the conversation based on the value of the variable. If HasFish is false, follow one branch of the conversation. If HasFish is true, follow the other branch.
  4. Game I'm making is not fun

    Or consider the opposite approach. Temporarily remove all the bells and whistles, and replace all your graphics with solid-colored squares. This will let you see the core game loop more clearly. If the core game loop isn't fun yet, it's unlikely that adding glitter on top of it will make it fun.
  5. Help With Story-Driven Game Concept

    Very true. Great game. There'll always be exceptions. I didn't mean to imply any hard-and-fast rules.
  6. Help With Story-Driven Game Concept

    Hi, That fact that you're differentiating between video games and written novels already puts you ahead of many writers. Sid Meier said "a game is a series of meaningful choices." In other words, your story (the kid, the dreams) are only important insofar as they support the player's story, which are the choices that the player makes. Can you refine your synopsis to focus more on the player's choices? For example, can you cut some of the set piece descriptions and provide more details on how the player's actions in the real world affect the dream worlds, and vice versa? Theme is also important, and you're starting to develop one, but I think you could go bolder with it. But keep in mind that players will also find their own meaning (as you can't ram theme down players' throats), so you'll want the game to support that. Another good choice. You may have noticed that in first-person games the story is about the world, since you don't even see the protagonist. Games with character-based stories usually pull the camera back to third-person so you can see the main character.
  7. Player interaction in a MUD

    Many MUDs do realtime combat. When a character enters combat mode, it automatically attacks its target on a set frequency, such as once every 5 seconds, until it exits combat mode, the target dies, or the target flees in a way that the attacker can't follow. But this isn't very sophisticated. It's as if the two characters are standing toe to toe alternately bashing each other with clubs. You could do this but add modifiers if the player submits positioning commands during combat, such as hiding behind a table. Or, if combat isn't the primary focus of your game, you could resolve the entire battle in a single action. This would allow you to assemble a dramatic description based on the properties of the combatants.
  8. Writing great villains

    Many traditional writers mistakenly think writing interactive fiction is the same as writing traditional fiction. There are vast differences, but there are also a few commonalities. Writing villains is one of them. The key is that there are no villains. There are only other characters who are sure they're the good guys of the story. It may be that you've come across interesting villains in games and didn't even think of them as villains because they were rounded characters. They're not villains because they're evil but because their beliefs oppose the PC's, and this conflict is the fuel of story. As mentioned, there are some AAA games with good stories and villains, but these larger games are such big technical productions that often the story gets tacked on, or at least heavily modified, at the very end. This just leaves opportunities for indies! :-)
  9. Community College or Online School?

    It's a two-year degree, often with a technical or vocational focus, although it's sometimes used as a stepping stone to a four-year university degree.
  10. Linear Level Design Guidelines?

    You might just be answering your own question there. Why not add some constraints? A good first constraint is to define what emotional experience you want to engineer for the player. For example, in a shooter, do you want the player to feel like they used tactics to stealthily outwit their opponents? Or hulked in like a reckless badass and just tore the place up? etc. Then there are the scenario constraints, such as the game's overall setting, where they are in the story / game progression, and what needs to happen to move it forward. Constraints like that will help set requirements on how you graybox your level. For example, a stealth level may require lots of places to hide and snipe, and maybe some set pieces where an attentive player can eavesdrop on clues, or where they're prone to have close calls with nearby enemies. If you're making a mech game, maybe the player is on foot at some point, and has to sneak across an exposed roadway patrolled by a guard in a mech. If the mech has a blind spot underneath it, the player could try to time her crossing so she stays in the blind spot while also avoiding foot patrols. As a player, I'd feel pretty sneaky accomplishing that, and it would be exciting and harrowing while I was attempting it.
  11. Writing question for RPG developers

    What kind of dialogue? Branching? If so, you might want to use a branching dialogue editor such as Chat Mapper, articy:draft, or even Twine. Long ago, some writers would use Excel and specify links to different rows. It's a low tech and manual process, but everyone has access to some kind of spreadsheet software. I've also worked with some studios that have used the Aurora toolset family (available to modders in the Neverwinter Nights and Dragon Age editors), exported to XML, and then imported the XML into their own projects. The best dialogue writer I worked with usually started with a whiteboard. At the top, she'd write down the purpose of the conversation to keep it in mind as creativity took her attention down different paths. Then she'd brainstorm different topics and goals for the conversation in a list on the side. After that, she'd write each topic on a post-it note and structure them in a dialogue tree, drawing arrows between them on the whiteboard. As she expanded each topic, she'd replace the post-it note with a large boxed area on the whiteboard containing more-detailed post-its. The process was similar to mind-mapping. After snapping some pictures in case someone accidentally erased the board, she'd usually replicate the same thing by hand over multiple pages in a notebook, filling out the the full lines of dialogue, and then transfer them to whatever writing software the project was using, such as Chat Mapper. Chat Mapper and articy:draft are nice because they have simulators in which you can playtest your conversations, and they can export different formats including screenplays for producers and voice directors. What you don't want to do is sit down at a blank screen in your editor and start writing your branching dialogue without a plan. It's fine to write stream of consciousness to explore different ideas as prep work, but branching dialogue needs to be structured to make sure the links all go to the right places and that you properly cover all the required topics.
  12. In the games industry, even good people in good jobs hop around often to different companies, but it's usually by invitation or networking, not by application. I'm afraid I have to agree with frob that it's likely these applicants are spamming applications out because they're desperate to find a job.
  13. What's so big about World of Warcraft ?

    It's not all about the endgame. Blizzard is skilled at making genres approachable to newcomers. Hearthstone and Overwatch are two more examples. To maintain a large player base, you need to first attract a large player base. WoW has a low barrier to entry, and it provides a reliable, consistent, polished experience for individuals as well as social players.
  14. Adventure Game Story Writing

    There's nothing wrong with talky as long as it's fun. Here are some thoughts: For pacing, consider 20-minute sequences (puzzles or story events), give or take. (Notice that RPG side quests are usually this long.) Ultimately, make it just long enough to engage the player, give them a sense of challenge and accomplishment without dragging on, and entice them to tackle the next sequence. You could structure it like film (although games are not film), where each sequence has its own setup, conflict, and resolution, and is composed of a handful of scenes, each of which moves the sequence forward by introducing a meaningful change, and where each scene is composed of beats, which are individual decisions such as picking up a cooking pot or climbing into a cannon. In a game, player actions drive most beats. Unlike film, you'll have to get a feel for how long the player will take on each beat. Similarly, test your larger-scale puzzles so people can solve them within 20 minutes to keep them from reaching for a walkthrough. If you ever decide to invest in a paid product to replace draw.io, take a look at articy:draft. It's really nice, and it comes with a few example projects including a Maniac Mansion-style adventure game project. (It's also 30% off on Steam right now.) You could also try Twine, which is free. Twine is a branching story tool, but you can use it like a flowchart program.
  15. Writing for a game: Case file 4:20

    That's a fine premise, which I mean as a compliment because there are plenty of games that have weak premises for whatever reasons -- they're sequels saddled with baggage, studio ownership got passed around like a hot potato, whatever The identical twin thing might be a little overplayed in film and TV, but I don't think it's been overused in games yet. But nothing in the gameplay makes it distinctly different from GTA or LA Noire, so execution would have to be significantly better than either of those games. That suggests a $100+ million budget in the hands of a team of experts. What gameplay changes would make it unique, where someone on Steam would decide that they have to play this game instead of GTA? Take Her Story, for example. If it were done like LA Noire but made on Her Story's budget, I don't think it would have had a chance.
  • Advertisement