This is the mail archive of the xconq7@sources.redhat.com mailing list for the Xconq project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: Non flat maps (use pentagons and septagons on maps)


Jakob Ilves wrote:
>
> For instance, there were a few strategy games written in the
> 80s (Task force games' "Cerberus" and
> Game Designers Workshop's "Traveller") which uses hexagonal
> maps of planets where a few hexes were
> replaced with pentagons, resulting in a playfield which
> topologically was a icosokaeder (20 sided
> die).  That would be cool to use in Xconq (or any other
> computer strategy game as well).

Oh fuck.  Dude, don't take this personally, but I want to seriously save
you from yourself now.  The primary reason I'm all but bankrupt is
because my 4X TBS game "Ocean Mars" is based on exactly such a hexified
icosahedron.  I thought with 11 years of 3D graphics experience it
wouldn't be any big deal to build such a thing, and it was.  YOU DO NOT
KNOW OF WHAT YOU SPEAK.  TURN BACK NOW.  Seriously, you can lose the
roof over your head going down this primrose path.  I've got $60K+ worth
of credit card debt to prove it.

The whole reason I went looking for other people's projects was to find
a more doable, interim prototyping effort to test some game design
ideas, before getting back to Ocean Mars sometime next year.

Now granted, it might have been more doable if I had adopted a very
coarse hex scale, instead of trying to model Mars at 10 km/hex scale.  I
was shooting for a map of 1.6 million hexes.  Much much smaller maps
actually run ok with unoptimized code.  But there are still decidedly
nontrivial issues of geometry projection, terrain texture mapping, and
stitching the mathematics together so that AIs can navigate the faces.
Realize that when traversing the topology of a real sphere, it's easy to
get lost!  Many directions are indeterminate, no matter how you hexify /
pentify them.

> Actually, by using septagons (7-sided "hexes") in clever ways
> it's possible to create a playfield
> with "tubes" protubing out of or into playfield surfaces.
> One can have a few large ikosaeders
> tied together with bridging tubes.

With such topologies, you're talking about a general nodal graph
implementation.  From an efficiency standpoint, you will need to
restrict the size of your map because each arbitrarily flexible node is
going to cost a lot more bytes than it would in an array.  Also search
algorithms have to be general purpose, they can't be optimized for the
special cases of rectangular or hexagonal arrays.  As long as you're
willing to settle for modest topologies, it's viable.  For instance,
Europa Universalis uses a freeform territory approach, and even though
it's fairly coarse the map of Earth feels huge.  One spends a lot of
time zooming in and out in that game, changing from the Eurpoean to the
New World theaters.

> Imagine such a playfield where you have various terrains laid
> out... wow!  Imagine the nightmare
> of displaying the thing in a GUI.... ouch!  It's the latter
> part which I've been scratching my head with.

Please... stop scratching your head.  You are wasting your time.  You
could spend weeks just *designing* this thing, on paper.  And you will
never, ever finish what you started.  You won't even make a dent in it.
I have Been There, Done That, the entireity of the past year.  And I
mean full time, not as a hobby project.  I know exactly how much labor
it's going to cost you.


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

Taking risk where others will not.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]