Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Soulwynd]

I could easily program prototypes for the data structure and game play. But meh, seems like I got completely ignored.

Hey not at all. Your ideas are quite interesting but a little bit complicated. But if you can come up with a working prototype showing your ideas working we would appreciate it.

Complicated? You should have seen the DNA handling for the initial design. It's just object-oriented handling, if you have any C++ knowledge you should know the concept and how it works.

Speaking of prototypes, my only win32 compliant one that still exists is a path finding one. I shall quote just to give an example.

Path Finding

Aural

Algorithm Theory:

    * Cast an 'aura' from the initial position to the destination noting the distance it covered then trace back to it using the shortest path.

Meta Code:

Code: [Select]

      Aura:
      point = start
      for each non-calculated direction from point
          assign it a value higher than the value at point
          recursive for the point at the direction found

      Trace Back:
      point = destination
      while point is not start
          Find the lowest value from the adjacent points
          if more than one is found
              follow the absolute direction to start
          point = point with the lowest value
Example:

Code: [Select]

                       Aura         Path
      +----------+  +----------+  +----------+
      |       #  |  |4345678#67|  |   567 #  |
      |    #  #  |  |3234#89#56|  |   4#8 #  |
      |    #  #  |  |2123#90#45|  | 123#9 #  |
      | @  #     |  |1@12#01234|  | @  #0    |
      |    #   * |  |2123#123*5|  |    #123* |
      +----------+  +----------+  +----------+
Good:

    * Small maps are the best for this kind of path finding;
    * Tight corridors and rooms also make the calculation way fast since those are ignored;
    * Can path find a labirint easily
    * Works on any kind of maps
    * Works great with terrains that cost more to travel, just add the cost to the number assigned to each step and it will simply find the fastest (cheapest) way there.

Bad:

    * Open terrain takes a long time to calculate, but you can do a directional aura if you know there wont be much blocking the path;
    * Time to calculate an area is squared;

http://www.geocities.com/kudex/pfind04.zip

Symbols:

    * @ = Origin
    * * = Destination
    * # = Wall
    * . = Aura
    * o = Path



I have a rogue-like database handling prototype, but I have to dig it from an old HD.

[Mephansteras]

To follow up on my AI idea, and be a little more specific, here are some psudocode ideas for the AI to keep the other creatures interesting without needing sophisticated AI

Individual Creature:
  Is Hungry -> Go to nearest food
  Is Thirsty-> Go to nearest water source
  Danger is Near -> Go away from danger until danger is no longer near
*Danger needs to be relative to the creature. A mouse is afraid of a cat, a cow isn't. Etc. The cat will be afraid of the dog, etc.
  Being attacked, attack back if possible.
    If hurt/scared - run away (This allows prey to drive off predators)
  Otherwise stand around doing nothing

Group Herbivore:
  No one is moving much (a few small steps to graze on more grass is ignored) - do individual
  Someone moves farther from the group - follow them until they stop.

Group Predator
  Alpha isn't moving much - do individual
  Alpha moves farther from the group - follow them until they stop.
    Alpha has a target - attack target unless attacked by something else, then defend yourself


This isn't as fully fleshed out as it'd need to be, but as a first run AI for creatures it should make the game interesting enough to keep people playing while develop it.

One thing I've noticed in game development, the AI doesn't have to be good to be fun. It just has to be good enough to keep things interesting.

[Dryn]

After reading through these posts the image I have in my head is like this: (Just gathering ideas, adding a few of my own)

You view the immediate environment around a 'nest' of your creatures (this could be as cells or more developed forms). You can see around them some environmental features; things that they need to survive and/or harness for a purpose. You have are able to place markers/waypoints (could be either generic or perhaps split up based on actions (eg collect, explore) or needs (eg hunger, thirst) that need to satisfied. I would rather specific guidance to make it more of a game rather then a sim) thus guiding your creatures actions, letting them evolve (as detailed in wallishs post). Each individual evolves on its own, rather then informally.

As your population increases (or perhaps as a result of exploration) the scope of your environment expands bringing with it more environmental elements (resources, other creatures). From time to time you will have to manage changes, like new predators, that you will have to evolve to accommodate in order to survive.

Perhaps once your population becomes quite expansive you will no longer be dealing with creatures but communities (no necessarily more advanced technologically, just large groups of your creatures). Evolution may be handled on a community basis, rather then individuals so that it can be managed easier processor-wise (talking out my arse?).

I also like the idea that your dudes will not only evolve on a biological level but a cultural and technological one. Given a change in environment some interesting interactions may arise. When you start out you are focused on biological evolution but as the game progresses cultural and then technological evolution is also important. Note that I am not using the word 'progress'. The idea is that tech evolution will flow from how they have culturally evolved which will flow from their biological evolution. In this way the game changes with a sense of progress and the player has new challenges. Of course this may be outside the desired scope.

Anyway, I think we need a seperate place to talk about this. That way we can split out the different discussions at hand so that things can be more organised and discussion will be more focused. (and people won't be ignored!)

[Fualkner]

Anyway, I think we need a seperate place to talk about this. That way we can split out the different discussions at hand so that things can be more organised and discussion will be more focused. (and people won't be ignored!)

www.forumer.com

Free place to get a forum. Works great, good software.

I'd love to work on this a bit, if help is needed or whatnot.

[Boksi]

Some name ideas, including already-made ones:

Spogue(Spore+Rogue)
SporeHack(Spore+NetHack)
The Twelfth Evolution(Bay 12 games)
SIEVE(Spore Inspired EVolutionary Engine)
NotSpore(pun intended)
BATRICS(Bay12games' Attempt To Recreate the Ideas and Concept of Spore)
DarwinHack(need I say anything?)




I failed to include Sporgue on purpose. It's a silly joke, but an ultimately annoying name. At least in my opinion.

[Areyar]

I always figured that "you can always go back to earlier stages, when you have unlocked them" in the pre-release interviews meant that you could at any time go 'slumming' as a creature: get out of your UFO and leave behind some germs to bootstrap an ecology.
Then use the godlike evolution tools to fast-forward evolution there.
I wondered back then, how they were going to realistically integrate the various stages...turns out they didn't, each stage stands seperate. 'go back' now means 'start a new game from any stage'....
so I agree with the no-stages sentiment.

What would be fun, is some mini-games in which you can test the viability of your germs real-time (or fast forward) in a petridish.
Have them compete with the local bugs of an alien ecosystem so you can alter the biosphere and thus the atmosphere of a planet.
interact indirectly by dropping in food, biocides etc.

yeah: start out as a UFO-driver, with little knowledge or tools, then slowly gather usefull organisms and genes from ecosystems you visit.
Purchase or invent/build new toolz.




How about 'Sim Spore'? ;P

[Duke 2.0]

 ERogue. Evolution Rogue.

 I would like to see at least a basic framework for all this before any fancy gene stuff gets in. Perhaps just starting with a player that can travel through a small world with food being generated?

[Zemat]

 ERogue.

That could go so wrong (Wikipedia link but maybe NSFW).

[Cthulhu]

 ERogue.

That could go so wrong.

I clicked that before I realized what it said.  Thankfully I'm fast with the back button.

[Fualkner]

I would like to see at least a basic framework for all this before any fancy gene stuff gets in. Perhaps just starting with a player that can travel through a small world with food being generated?

Agreed. Pygame is very easy to get started up, I've seen some code. I can ask Keiseth if you're interested. He's managed a random map gen and a basic roguelike movement scheme in one day.

[Nonanonymous]

I think that completely random creatures would be a bit un-fun.  I like the simulation-heavy evolution idea, but you should at least get to choose between things like mammalian/reptilian/insect and terrestrial/flying/swimming.

[Areyar]

how about taking controll of any creature you desire to controll?
(if a mindcontroll device has been acquired)

then have it jump off a cliff to feed your favored omnivore tribe of diggers.

ideas:
ecological niches besides just the general herb/carnivore Level1/2/3;
such as scavenger, pollinator, seed gatherer, filter feeder.

more variation in nesting; surface, underwater, burrow, cave, treetop, constructed (anthills/homes).

[Dryn]

I would like to see at least a basic framework for all this before any fancy gene stuff gets in. Perhaps just starting with a player that can travel through a small world with food being generated?

Personally I think scope and focus needs to be agreed upon first before coding starts.

[Fualkner]

I would like to see at least a basic framework for all this before any fancy gene stuff gets in. Perhaps just starting with a player that can travel through a small world with food being generated?

Personally I think scope and focus needs to be agreed upon first before coding starts.

I think that he's talking about the actual game, and not the scope and focus. Basically, he's referring to making a tech demo of moving about and generating a food object.

[Deon]

OK so this is a nice imagination thread but who wants to take the lead. We may need a few leaders and they should decide it honestly. I.e. I know myself and I can't be a project leader for now: I can disappear for some time without warning and I don't want a project to be halted because of it. However I have enough fanaticism to sit at the computer days and nights when I have an inspiration and ignore family and daily needs to finish the work I'm intended to do so I can be a nice helper. So let's decide the core team and start a small separate forum.