May 28, 2017, 01:19:00 PM

Author Topic: Formicarium: Dwarf Fortress meets Ants&Genetics  (Read 4002 times)

Bersaelor

  • Newbie
  • *
  • Posts: 4
    • View Profile
Formicarium: Dwarf Fortress meets Ants&Genetics
« on: November 18, 2014, 02:13:53 PM »
Hey guys, please don't ban me for spamming, but I'm a big fan of DF and Gnomoria and hells yeah, I also had the innovative idea of coding a new take on DF, yay ;)

I had the idea while playing DF, when I thought that sometimes it's less Dwarf fortress but ant hill , with all the little guys scurrying to and fro...
So here's a first screenshot (from the underground management of one's ant hive):


And here's another one of the surface with bugs, beetles, plants and butterflies (everything has attributes and code and reproduces including the plants):


So, if you're interested we are currently running a kickstarter campaign:

« Last Edit: November 22, 2014, 11:50:30 AM by Bersaelor »

Merry76

  • Moderator
  • Hero Member
  • *****
  • Posts: 3448
    • View Profile
Re: One more game inspired by Dwarf Fortress
« Reply #1 on: November 18, 2014, 10:16:50 PM »
I (like many here) am a sucker for Dwarf Fortress - likes. I have been burned by kickstarter in the past, so I became careful and untrusting. Its not your fault, but it is your problem, unfortunatly. This also belongs to the "Other Games" forum, where it will be moved after this post. That being said, let me tell you what I think...

Interesting concept, I also thought of simulating a few hundred ants once. But it was more of a vertical ant farm (but with physics - because everyone loves physics, right?).

What rubs me the wrong way however is how low you set the kickstarter goal. I know you already spent the last 6 months into the project, but I dont think 20k will take you that much further. Where I live, that goal would sustain two developers 3 months tops (on a very modest salary that would fit someone without any coding skills). I dont really want you guys to live off instant noodles in a back alley when you are developing a game I want to play. This is why I would prefer a "X thousand dollars would allow us to work on the project for a year. We need a year for it, but if we manage to hit X*2 thousand dollars we will devote the next two years into the project" kind of statement. Way too few projects state anything like that, hit their targets and you never hear of them again because they run out of money a few months later and are still not finished.

How finished is your prototype? You didnt write that in the Kickstarter page. You really should :)
Have a problem or a fortress so awesome it needs to be shared?

Well, go on, dont be shy! Use the GnomeworldPool Dropbox account!
How to share Savegames

Bersaelor

  • Newbie
  • *
  • Posts: 4
    • View Profile
Re: One more game inspired by Dwarf Fortress - Kickstarter
« Reply #2 on: November 22, 2014, 11:50:04 AM »
Hey Merry76,

you're making a good point. To be honest, I'm a well paid freelance iOS Consultant so I don't have to worry about my own living expenses.
I'd be ok if I get out of this project with a all my investments recovered by the sales, as the 20k are mostly to cover all the expenses for the next 6 months. ( I already paid my designer for 6 months, have to pay her for 6 more, paid for the video, server fees etc so the kickstarter is really just making sure I don't have to pump any more money into the project after it's funded at kickstarter).
I'm really eager to build a game that uses genetics and a level of detail that come close to Dwarf Fortress while being visually slightly less .. weird, so I don't mind If I do the programming work for free (also the work I do in the entertainment industries is usually rather boring so it's a great balance).


The current prototype is quite far advanced, in the dwarf fortress forum I explained that the metabolism of 1000 ants and 4000 plants can be run without the frames dropping, I'm hoping to finish the first playable demo next week and will update the campaign-description with a youtube link to the current game then.
Pathfinding works, making ants dig tunnels according to player will works, Queen, Worker and Soldiers are implemented and we have Beetles and worms as herbivorous npc's. 2 different spider species are animated but not in the game yet (I still have to balance how much they attack) and 2 different pollinators pollinate flowers and distribute flower plants.

Here's the Q&A from the DF-Forums:

First: "I like the concept. What about reproduction? Spreading new colonies? Inter-colonial warfare? Anthill coalitions?"

Let's talk a little about scale&scope.
As you all know, Dwarf Fortress' FPS drop once you have 200 dwarves (+ assorted NPC's in the map amounting to maybe 600 individuals).
Our current prototype can handle about 500 ants + 1000 npc + 1000 plants and 10k liveless objects without dropping the framerate on an iPad. A Desktop PC doubles this. Since I only spent a year worth of saturdays and 3 months devoted time to it yet, there's still a little room for optimization, so let's add another factor of x2 and we arrive at 2000 ants and like 40k total objects.
Now you can all do the math about how many colonies we can have in one game given those numbers.
Also on average each animal/plant has a double set genes + a ton of variables governing it's metabolism so that saving all the individiums of more then 2 colonies would put stress the memory of modern computers.

1. Will there be different species? Like Slaver Ants?, Army Ants? and polygyn (a hive with many queens) and monogyn (hives with only one queen?)

There will be different species, but in gameplay terms I am highly sceptic whether it will be fun to only give the player only one species to play with at a time.
I'm tending towards letting the player start with your run-of-the-mail garden ant with a Queen that just landed from her nuptial flight. And then you can unlock more genes/castes/abilities as you discover more males that can mate with your queen. Some army ant species do this, where the queen(s) mate with multiple males during their lifetime that get carried to them by the workers.
Think about Plague INC, it would be pretty boring when you start out as a Norovirus and can only unlock abilities that Noroviruses have in Real life. Instead you get to progressively unlock more Symptoms/Infection pathways that are never possible in one real life pathogen (thank god for that!)

@Queens: I believe I might include some tolerance gene/ability that allows you to have more then one queen. So "polygyny" will definitely be a part of the middle/late game.

2. Will it be more scientifically accurate with different species only having different ants and specific mechanics (aka leafcutters can grow fungus but slaver ants can not, but in return get to keep slaves)? Or will it more likely be a freeform system?
I will have both species abilities but I'm tending towards freeform, see reasoning above  :)

3. Do you intend to include different seasons?


Yes, as you have seen in the video we already have a day/nightcycle (hooray for OpenGL Lighting), and we can make leaves fall already. Since our plants get pollinated and different phases (flower bud, flower, pollinated flower, seed head ..) seasons will be the next step.
In cold climates we might skip the winter though and just make the workers die and the queen sleep at the moment of the first snowfall?

4. Will humans make an appearance? ( the big foot out of the sky...run for yer lives!)

For starters I was thinking of shrews attacking insects/ants. At the scale that you look at the game they'd already be quiet scary.

5. Will you be able to invade other Ant-Hives and take their hives?

Maybe, but then I have to write AI for the central management of a hive. See multiplayer

6. Will there be a multiplayer option?

Unfortunately, synchronising games, sending package updates back and forth between players etc is a ton of work. The first game I ever wrote was a hexagonal chess-like 3D Game and I worked 3 months on the multiplayer alone. Therefore in our current quiet restricted timeplan that we have after the kickstarter-campaign it will not be included but it is in the strech-goal's.

7. Will the game in scale be like Dwarf Fortress, or will you be able to actually expand into a farch stretching "Ant-Empire"?

Very much like Dwarf Fortress. For the mentioned reasons above, I'd like to simulate each individual living being with attention to detail. If you have an ant-empire with many hives that would not be feasible with modern computers. If the project is succesful and maybe we have computers with 10cores in ten years, who knows, maybe we can have one hive per CPU?

8. Is there going to be a campaign?

Yes, but I haven't worked at it so much yet.
You know this feeling when you observe an ant hive, and everything first seems so chaotic yet every ant seems to have some reason for whatever she is doing at any point? I want to capture this feeling and let players enjoy and control this flow. Currently working on the logistics of an ant-hive, making individual ants transport goods from outside to stockpiles and from there to the queen/nurseries.

9. What about "animal husbandry?" Will ants be able to domesticate lice and milk them for their sweet, sweet sugar?

That was planned since the first time I had the idea for this game in my head. I was reading Jared Diamond's Third Chimpanzee a year ago where he talks a long time about the fact that humans totally didn't invent animal husbandry ;)

10. I hope ant empire functionality works - it could be like that when you create new colony, you are given choice between which one you want to control and hand other over to the AI... otherwise I assume player burden will be massive.

I think the player burden would not be different then in dwarf fortress. Most processes are automated, and you only have to plan them. Also we will have a Dwarf Therapist - like menu right in the game  ;)

11. As a geneticist (I guess that the way to make it into a title) by trade and training, this looks so. cool. Definitely keeping my eye on this. At first I was a little lukewarm, but the more I thought about a modern Sim Ant-ish sort of thing where I can actually fiddle with genetics, the more excited I got. You could even indirectly select for better plants by making sure the bad ones get taken out by your ants.
Thank you for the compliments and interest. The genetics are basically the same simplified system that Randall Munroe (creator of xkcd) describes in  his latest book 'What If' page 158. In case you haven't heard about it, as a scientist you should check it out. (And tell me what you think about this system!)
The idea with ants weeding out the inbred plants (you know, we have asexual and sexual reproduction) sounds amazing, thank you for that!
« Last Edit: November 22, 2014, 12:02:48 PM by Bersaelor »

seronis

  • Hero Member
  • *****
  • Posts: 600
    • View Profile
Re: Formicarium: Dwarf Fortress meets Ants&Genetics
« Reply #3 on: November 22, 2014, 12:37:02 PM »
I recommend you keep a build mode prepared for 'stress test'  that will every few seconds add a huge batch of ants to the population and check computation times.   Rerun the stress test mode every time you make changes to your AI so you can see  how many ants it reduceses your total ability to handle

Bersaelor

  • Newbie
  • *
  • Posts: 4
    • View Profile
Re: Formicarium: Dwarf Fortress meets Ants&Genetics
« Reply #4 on: November 22, 2014, 04:24:04 PM »
Thanks for the idea :)

What I'm using atm is a method that does a thousend-tick testrun with a random environment but always the same amount of creatures/plants. Usually I let it run n-times and take the average. This number I use to gauge how much a new function affects my performance.

Merry76

  • Moderator
  • Hero Member
  • *****
  • Posts: 3448
    • View Profile
Re: Formicarium: Dwarf Fortress meets Ants&Genetics
« Reply #5 on: November 23, 2014, 10:38:41 AM »
I am a "raw data man", so I kind of need that part of the information. I didnt think of reading up the DF forums, but you conveniently pasted it here - thanks for that.

It looks like you could pull it off even without kickstarter, but with kickstarter it will "soften the blow" of launching the game. I can rally behind that.
Have a problem or a fortress so awesome it needs to be shared?

Well, go on, dont be shy! Use the GnomeworldPool Dropbox account!
How to share Savegames

Bersaelor

  • Newbie
  • *
  • Posts: 4
    • View Profile
Re: Formicarium: Dwarf Fortress meets Ants&Genetics
« Reply #6 on: November 23, 2014, 03:12:38 PM »
I am a "raw data man", so I kind of need that part of the information. I didnt think of reading up the DF forums, but you conveniently pasted it here - thanks for that.

It looks like you could pull it off even without kickstarter, but with kickstarter it will "soften the blow" of launching the game. I can rally behind that.
Well other thing is to just gage peoples interest. If there are only 50 commited people who would play it, I have to think hard if I really want to spend another year on this (having some holidays would be nice too, haven't had proper holidays in over a year).


After 2 days of implementing the digging, I finally managed to make it stable&clean.
If you like, check it out on youtube.

Lobstarooo

  • Jr. Member
  • **
  • Posts: 73
    • View Profile
Re: Formicarium: Dwarf Fortress meets Ants&Genetics
« Reply #7 on: November 25, 2014, 01:59:42 PM »
This is really interesting. And to be honest, since my girlfriend is really fond of ants and was also looking at DF to go and play it, this'd be an extremely good combo for her. For that alone I'll back it.  :)  Looking forward to seeing more about this!