Loading ...
Sorry, an error occurred while loading the content.
 

Re: [gamedesign-l] region based commands

Expand Messages
  • Wayne Imlach
    On clicks per game: Don t judge this by tiles, but by number of features on the map that might require a re-assessment of the initial click. In most games I
    Message 1 of 3 , Feb 2, 2005
      On clicks per game:

      Don't judge this by tiles, but by number of 'features' on the map that might
      require a re-assessment of the initial click. In most games I may traverse
      from one side of a map to another with a single mouse-click, assuming there
      is no blocking obstacle along my path.

      Even in a turn-based game, there is no reason that a command that may take
      several turns to complete cannot be given with a single click, with the
      order automatically being followed on each subsequent turn until fulfilled
      or changed.

      Unless you are talking about a game where the player has to interact very
      specifically with *every* tile in the game map - something I see little
      benefit to for such a large scale game. I every tile really going to
      introduce some new piece of information that causes the player to re-assess
      each order?


      On flag interfaces:

      One idea I toyed with briefly in Startopia (although it wasn't the game to
      really use it) was the idea of flags with 'priority' values that on a simple
      level split up the ratio of forces assigned to each flag. So putting down
      two flags of equal priority would split your forces between them 50:50.
      Increase one of the flags to priority '2' and leaving the other at '1' would
      produce a 2:1 split of forces. Each new flag added instantly re-calculates
      the distribution of forces on the game map. The interface for each can be
      simple as well - a player may add a flag type (from a list of possible
      types) and then can either increase or decrease the priority of existing
      flags. Decreasing the priority below one removes the flag from the map.

      Additional functions could include changing a flag type from one type to
      another without affecting priority.

      So, rather than seeking out units and directing them to a location or task,
      the player merely sets a flag at the destination or task, and assigns it's
      priority over other flags to 'attract' units toward it.

      Of course, this then leaves the split of forces between the flags to the
      AI - though with some simple weighting based on distance to and
      effectiveness for the type of flag appropriate units should be assigned to
      appropriate flags.

      Beyond this are the specific functions of the flags, and other factors such
      as duration, maximum numbers of flags, maximum priority that can be assigned
      etc.

      ----- Original Message -----
      From: "Brandon J. Van Every" <vanevery@...>
      To: "Gamedesign-L" <gamedesign-l@yahoogroups.com>
      Cc: "Don Bernstein" <adiemus57@...>
      Sent: Tuesday, February 01, 2005 1:22 AM
      Subject: [gamedesign-l] region based commands


      >
      > After playing too many games of Freeciv, it dawned on me that the unit
      > of activity on a Freeciv map is roughly the area surrounding a city.
      > Let's say that's 5x5 = 25 squares. I believe a standard Civ map is
      > 64x32 = 2K squares. So, one could abstract a Civ game as being on a
      > map of 2K / 25 = 80 squares. Where each square can contain a city, and
      > is otherwise the "atomic unit" of military and infrastructure.
      >
      > Let's say I'm making a "plop down a flag" interface. Let's say that the
      > area a flag affects is 1 city. Then, I would need to perform at least
      > 80 mouseclicks to finish a Civ game. Many more, if I must perform
      > multiple tasks on a square. For instance, conquer, colonize, build
      > roads, develop infrastructure, build units, etc.
      >
      > Earth has a surface area of 512M km^2. Holland is roughly 33K km^2. I
      > have standardized on Holland as the smallest geopolitical unit worth
      > dealing with. On the argument that once upon a time, Holland was a
      > major maritime power. It took over a noticeable chunk of the Earth's
      > surface, primarily in Indonesia, with South Africa contributing, and a
      > few other colonies here and there. So, by the "Holland standard," a map
      > of Earth would be 512M km^2 / 33K km^2 = 15,500 squares.
      >
      > 15K is a lot of squares! Assuming an average of 2 mouseclicks per
      > square, how long do you think it's going to take to perform 30K
      > mouseclicks? Well, if each one takes 0.5 seconds, that's 30K seconds =
      > 500 minutes = 8.3 hours. And it would be quite a clickfest! I don't
      > think anyone with the patience to play 8 hour games actually issues a
      > mouseclick every 0.5 seconds. I would wager that people can keep it up
      > for at most 1 hour, and more likely 20 minutes.
      >
      > Now, maybe it isn't necessary to click on every square to colonize an
      > entire planet. If one assumes that to colonize a given square, one must
      > blaze a path from one's empire to the colonization site, then clicking
      > on a square could imply lotsa intermediate moves.
      >
      > There are 3 problems with this approach, however:
      > - the intervening territory may not be worth colonizing
      > - oceans permit free movement to colonial sites, without the need for
      > intervening conquests
      > - dropships permit free movement to almost anywhere on the planet,
      > without the need for intervening conquests
      >
      > This implies that a "drop a flag on a square" interface doesn't scale
      > up. It could work if the game is only about 80 squares. (Which,
      > incidentally, is the density of an Axis & Allies map of the Earth.) But
      > for maps with more squares, we need flags that can cover a variable
      > number of squares. One could, for instance, set the radius of a flag to
      > cover the entire planet. Then just sit back and watch the AIs conquer
      > the Earth on a "default" setting.
      >
      > Should flags just have a selectable radius, or should they be fully
      > customizeable regions? The regional approach occurred to me as I
      > remembered DungeonKeeper. One would specify regions of a dungeon to dig
      > out. Then imps would come along and do the digging. The regional
      > specification interface was extremely simple. Click individual dungeon
      > squares on or off, or click and drag to select a larger rectangular
      > region.
      >
      > I am recalling, however, that DungeonKeeper maps don't have that many
      > tiles to them. Probably no more than a Civ map. Builder games like
      > SimCity or Pharaoh or Zeus are similarly smallish. These click-and-drag
      > interfaces work ok at the level of a builder game, but even then, it can
      > be time consuming. I've observed in the past that a Builder game is
      > very much like a paint program. One is in essence "painting" one's city
      > or dungeon onto the map. So do we want paintings with more detail than,
      > say, 256x256 ?
      >
      > For Builder games, I've wanted the ability to drop entire predesigned
      > neighborhood chunks into my city. It is tedious to draw each
      > neighborhood layout the same way every game, one house at a time. I am
      > wondering if similar command structures might be devised for military
      > units? An "attack plan?" That might be too complicated to impose over
      > arbitrary terrain, however.
      >
      >
      > Cheers, www.indiegamedesign.com
      > Brandon Van Every Seattle, WA
      >
      > Taking risk where others will not.
      >


      --
      No virus found in this outgoing message.
      Checked by AVG Anti-Virus.
      Version: 7.0.300 / Virus Database: 265.8.3 - Release Date: 31/01/2005
    • Brandon J. Van Every
      ... In 4X TBS, I find that specific orders or build queues (i.e. go exactly here, build exactly this) are invalidated by game circumstances in 10..20 turns.
      Message 2 of 3 , Feb 2, 2005
        Wayne Imlach wrote:
        >
        > Even in a turn-based game, there is no reason that a command
        > that may take
        > several turns to complete cannot be given with a single
        > click, with the
        > order automatically being followed on each subsequent turn
        > until fulfilled or changed.

        In 4X TBS, I find that specific orders or build queues (i.e. go exactly
        here, build exactly this) are invalidated by game circumstances in
        10..20 turns. Depends on how much is happening in the game. At some
        points in the game, it's simply not possible to queue your orders.

        The only semi-reliable "subsequent turn" order is an auto-pilot. A lot
        of auto-pilots are dumb. For instance, in Freeciv the Workers tend to
        walk into zones of enemy activity and get killed. They will retreat
        from visible enemies and won't approach them, but if the enemy isn't in
        plain view, they have no concept of "this zone over here is typically
        crawling with enemies."

        Some auto-pilots choose safety over efficiency. For instance, the
        Freeciv exploration auto-pilot doesn't move a Trireme as maximally as it
        could to search for land bridges. This keeps your ship from
        gratuitously sinking when something unexpected gets in the way, but it
        also keeps you from discovering that other continent sitting right next
        to you.

        AI stupidity means the jobs don't really get done, so you end up having
        to do a lot of manual labor anyways.

        > On flag interfaces:
        >
        > One idea I toyed with briefly in Startopia (although it
        > wasn't the game to
        > really use it) was the idea of flags with 'priority' values
        > that on a simple
        > level split up the ratio of forces assigned to each flag. So
        > putting down
        > two flags of equal priority would split your forces between
        > them 50:50.
        > Increase one of the flags to priority '2' and leaving the
        > other at '1' would
        > produce a 2:1 split of forces. Each new flag added instantly
        > re-calculates
        > the distribution of forces on the game map. The interface for
        > each can be
        > simple as well - a player may add a flag type (from a list of
        > possible
        > types) and then can either increase or decrease the priority
        > of existing
        > flags. Decreasing the priority below one removes the flag
        > from the map.

        In Startopia, I didn't find the battle mechanics to be controllable /
        subtle enough to warrant the placement of different flags with different
        weights. It's a RTS, and at least against the computer, you're best off
        bringing all of your units into a huge mass to destroy any resistance.
        First you gather and stage outside of enemy range, then you advance en
        masse and destroy. The main combat determinant is covering fire from
        security towers. In practice, the priority interface didn't give
        results any different than DungeonKeeper's "Flag Horde" interface. In
        DK there's 1 flag, you plop it down where you want your horde to go, and
        that's it.

        TBS might make all the difference. There's time to set up more
        elaborate tactics. But, that in and of itself might become a bog. I
        don't know how many juggled priorities a player really wants to contend
        with. It might turn out to be a small number, like less than the
        fingers of 1 hand.


        Cheers, www.indiegamedesign.com
        Brandon Van Every Seattle, WA

        "The pioneer is the one with the arrows in his back."
        - anonymous entrepreneur

        --
        No virus found in this outgoing message.
        Checked by AVG Anti-Virus.
        Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
      Your message has been successfully submitted and would be delivered to recipients shortly.