Roguelike interfaces [was: Results of prefered graphics poll]
Lucas Ackerman
always been a staple of my serious gaming and I've thought quite a bit
about the ASCII dungeon phenomenon. Here are some relevant thoughts
and analysis. This seems as good a place as any to discuss it, since
it's about roguelike graphics preference. I'm afraid this may become
a bit of a rant, but brain dumps are what r.g.r.d is for :)
One very important thing I've come to realize is that the advantage of
ASCII is in its symbolic power. When I look at an Angband or NetHack
screen, I instantly know what the situation is, and I don't have to
learn a new set of image symbols when changing games. The other part
is that since ASCII's expressivness has (possible) near instantaneous
transmission and interpretation, roguelike games can be played at the
speed of thought instead of realtime (you never have to wait) or even
normal turn based time (since turn actions happen instantly).
Animation can be nice, but it can also be boring filler if you just
want to get onto the next action.
ASCII does have limitations though, primarily its limited resolution
(and the resulting effect on relative scales), and lack of spatial
flexibility, most notably a 3rd dimension. I've considered some
various ASCII alternatives to the fixed 2D-grid, such as higher ASCII
resolutions, multires/scalable ascii tiling, 3D in ASCII, and ASCII in
3D, but none of these quite address the problem, which is to increase
spatial expressivness and information quality.
The last sentiment there, ASCII in 3D, is almost right, but it would
be more of a nostalgic holdover than a real step forward. What I
think might really tip the scales though is a more general 3D
graphical symbolic interface. World structures can be drawn spatially
(pretty artwork or not), objects (items, creatures, dungeon/world
features, etc) can be all represented simultaneously (even with
graphical depiction *if useful*), and game information (notes,
inventory, stats, history, etc) could be overlayed in relevant and
meaningful ways.
The primary objections I've always seen to "lets make a 3D roguelike"
is that the traditional approach of 3D representation, with realistic
graphical depiction, is limited in all the ways that symbolics are
not. If perfect graphical realism were possible, I'd say go for it,
since you could glean all useful game information by inspecting the
reality-quality graphics, but this isn't the case (and won't be for a
looong time, because even when the technology catches up, content is a
harder problem). Thus, conventional graphical games have *always*
provided extra information in some form or other (health bars, game
stats, or whatever), but sacrifice the roguelike capability of
symbolics and speed-of-thought play for the sake of accessability.
Roguelike accessability is an important point though, since it's the
single biggest hurdle to all new players. The learning curve can be
very steep because they have to contend with what tend to be deadly
and unforgiving game worlds while simultaneously learning both the
ASCII and keyboard interfaces. A good symbolic roguelike graphical
interface (new acronym?: SRGI, RSGUI, ...?) could address this too
though, by presenting "layers" of extra information for new and
intermediate players, which could be set by preference. Keystrokes
relevant to specific objects could be displayed concurrently, as could
additional descriptions, statistics, rule information, and so forth.
Focus-based information could keep down the clutter, by scaling up
info display as you zoom in to examine an object with the mouse, or
just scaling up local screen area for info display on mouse over, or
proportional to distance from character (the immeditate surroundings
of @ could be larger and present expanded info), or proportional to
interest or immediate importance (since @ probably cares more about
the incoming black dragon than the newt at his feet).
What the best system is can't possibly be known until we've tried
some, but the room for improvement is pretty substantial. Gaming fun,
I feel, is directly tied to learning via play, and the less of a
hurdle the interface poses (both input and output), the faster you can
get into learning a game and playing, and the more @ fun you'll be
having.
The final word on RSGIs (I'm already tired of typing it), will
undoubtedly be in the gameplay department, since roguelikes have
always been about hardcore gameplay and very little else (witness our
current interfaces). As mentioned earlier, fixing some of ASCII's
shortcomings is a prime motivation for moving to 3D, and improved
resolution, spatial flexibility, and 3D depiction could make roguelike
gameplay just that much better. The details here are obviously game
dependant, but moving to a more seamless, scalable, and natural world
depiction should certainly improve game play and expressivness of
actions. It will almost certainly be a topic of huge debate, and one
I rather look forward to.
With all this said, I haven't the slightest notion as to when or how
this will happen. I guess the bottom line is that someone has to sit
down and do something like this. It will entail a certain transition
phase, as the next roguelike wave comes on and existing roguelikes get
RSGI upgrades (and even dual interfaces, so the same game can be
played via RSGI or traditional ASCII), and will be a little weird for
players at first, but it's one of those things that seems like it will
just click one day and have everything fall into place. One of the
(theoretical) interfaces most similar to traditional ASCII roguelikes
is that of the symbolic streaming character displays in The Matrix,
they must seem weird to non-users, just as roguelikes do, but serve a
deeper purpose. RSGIs could be an important step in symbolic
graphcial interfaces in general, since it's a largely unsolved
problem, and apt solutions would have widespread applicability.