Vangard

True, but kings can also fall if you know what I mean :stuck_out_tongue:

The basic idea for a political AI is that it is smart but can never win. Weaker nations always band together against the winning AI so an equilibrium is always maintained. Also larger kingdoms are less cohesive. It should lead to a delicious state of perpetual warfareā€¦ Of course it means the title of High King is elusive and at best fleeting.

This sounds like a pretty huge project. Is this just a toy project or are you planning to take this much further?

Iā€™m expecting to work on it for the next two years. Iā€™ve been experimenting and thinking about this for the past three years.

Whoa, you kept that quiet. I thought you just decided to make this game on a whim! :open_mouth:

It is the same game as my earlier Fog attempt; I just changed the plot and threw out the graphics. From Fog I got a number of ideas for AI and the general form of gameplay I wanted, but I suppose 95% of the code is new, from the start of this thread. The main thing I learnt from Fog was to be less ambitious and to limit scope.

Spent the evening playing with the look of paths.

I doubt that very much young ags1ā€¦
:wink:

This game might fit on mobile after all. Here is a Galaxy S6 compared to my development laptop:

http://www.headline-benchmark.com/results/c312470e-5153-4db2-8264-128c0fa25ede/9c86c50d-1f80-4029-b47e-6e45a2e379e6

If I scaled the map down to a quarter, it would be able to run on the S6 (probably). But it would be one of those games that makes the phone hot and eats the battery.

I felt tired tonight (had to put in overtime in my real job) so I just made a start on new GUI elements: health bars and alliegance badges. All these people are at full health and belong to the sky blue nation.

These individuals look very sympathetic to me :point:

In a good way, I hope!

Put together a list of things to get done for the first tech demo, in order of importance:

ManufactureItemBehavior - new simplified behavior
skills - load from career files
refactor#37 - load all characters from career files
spawning - smarter spawning that respects town planning
experience - tweak experience levels
leveling - add simple AI for leveling up, add UI for player controlled leveling up
trader - implement trader career
smith - implement smith career
hunter - implement hunter career
servant - implement servant career
eating - fix eating behavior
alehouse - add alehouse model
feasting - add evening feasts at alehouse
temple - add temple model
cottage - add cottage model
longhouse - add longhouse model
hall - add hall model
combat - integrate combat code; make wild predators hostile
hoards - place initial hoards and guardians on map
tree/rock/frill geometry batching - integrate and benchmark batching code (not dependent on graphics lib)
javafx migration - 'cause I want to use JavaFX
full screen mode - obviously
test/resolution issues - text will be too small on screens with high DPI; make font sizing proportional
Radial menu - add some demo code

Realistically, this will take me a while :frowning:

Not in the demo:

Finishing town planning
Entering buildings
Lands of the alfar and giants/goblins (alternate planes of reality)
goblin strongholds
politics, intrigue and warfare (I have some simulations already)
magic system
dialogue
social interactions (chess, gambling, wrestling, song, poetry)
building construction and settlement establishment
etc :frowning:

Youā€™re gonna love JavaFX. Itā€™s way better than Swing :smiley:

Also, thatā€™s crazy. That is crazy. How are you going to create a game on that large a scale? Itā€™s a virtually impossible feat to accomplish on oneā€™s own without giving up, donā€™t you think? A magic system, social interactions between villagers and suchā€¦itā€™ll take another year at least.

Iā€™m estimating 2 years at least. But finishing is not a major goal; I just like working on this project.

Also, I do not think of this as a very large project - I prefer to think of it as a large number of small projects :slight_smile:

Sometimes building a game is the game.

Arenā€™t we all a little crazy right when we start thinking about making making games our job?

Thatā€™s a very good point.

I simplified some behaviors and of course everything broke. It turns out I switched on equipment logic and forgot to give anyone their tools. The poor little things were running in circles helplessly.

Iā€™m curious as to how youā€™re going about the entities. Is it an entity system? Or are you using inheritance / composition? Whatever it is, it seems very, very cool!

There is a single Entity class. Each instance of Entity holds a list of Property objects, where a property could be a behavior or an attribute. For example, trees do not have any behaviors normally, but there is no reason I couldnā€™t give a powerful ancient tree all the properties of a druid (except GotoBehavior of course).

All properties store their working data in a common map stored on the entity, so saving the entities to file should be very simple when I get round to that - just write out the map and write out the list of property IDs. (I have some legacy properties that still store their data locally, got to spend an evening fixing thatā€¦)

To avoid looking up properties repeatedly from a map or set, I also have a class that hardwires direct references to the most important (mostly) common properties. It should make the code execute a bit faster (I havenā€™t tested that), but it sure makes coding easier.