agentx

get: ==== Building ActorEditor (release) ====test_root.cpp/bin/sh: 1: sdl2-config: not foundLinking test/bin/sh: 1: sdl2-config: not foundHow do I install sdl2-config? sdl-config is present.

What is the apt-get install for sdl2?

No, starting over....

elexis wrote: -Ubuntu 14.10 -compiled 0ad against SDL2 ??

Atlas starts, can copy paste actors, open save maps, anything special to test? Was SDL2 automatically integrated?

thanks, compiling....

I put the diff in root, called patch and got: ax@T3000:/Daten/Projects/Osiris/ps/trunk$ patch -p0 < fix-atlas-sdl2.diff (Stripping trailing CRs from patch; use --binary to disable.)can't find file to patch at input line 5Perhaps you used the wrong -p or --strip option?The text leading up to this was:--------------------------|diff --git a/source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp b/source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp|index 54b8463..26b6b11 100644|--- a/source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp|+++ b/source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp--------------------------File to patch: What does it want?

> Patch your revision with it How does that work with RapidSVN?

I've updated to b16355, what do I have to do to test the diff before running make? Have Ubuntu 14.04.

> I like the "streets" concept Thanks, I wish the bot could pave them.

> range placement ??? > rules I want the cities look organic, like they have grown over decades and somehow blended with the environment. Mostly to avoid all look the same and to give opponents more challenges. Running a real constraints based layout procedure (some algos exist, there are great js libs for axis parallel boxes) would cost to much time or don't work with rotated rectangles, so I'll stick with a tiny set of fake rules, so far I have as input: civ and building size and as output angle and border. Currently I'm looking into building blocks of 4 houses with no border for the heles. So they occupy as much space as a fortress and provide some barrier. I've never seen real street fight with 0AD and I want the bot to hold and defend positions inside the city. The API's obstruction map is extremely important to detect positions worth to defend. > satellite villages I like them, especially because houses are not only build near the centre, but appear everywhere on the map. However, there is a bigger concept behind: Lion's last exit dock, military satellites, fortified miner villages, etc. all fit in this category. Sometimes a bot is forced to build a new one (centre lost), sometimes it has economical sense, sometimes it is just to fortify a claim. For sure these satellites have a main building and Hannibal is prepared to work with multiple main buildings, but still I have the satellites not really thought through... When is it time to give them up, when to defend at all cost? What are the criteria? PS: > blacksmith and barracks Have a logical relationship, but a bot doesn't profit if they are close.

Lion, good tips, ..... for humans.

I now have some grip on map analysis and play around with village layout. This is the current status: Each centre has a no-build here zone with a radius of 40m. Also Hannibals trys to detect possible streets. It selects 48 points from an inner circle and 48 points on an outer circle and runs the pathfinder on each of these pairs: Cells visited by the pathfinder multiple times qualify as street-cells and block buildings. Hannibal launches 4 builder groups for houses. The groups build the next house near the last one, but may start from different points, hence the "clumps" in the maur village. The find-a-free-place algo allows to define a border with of size 0, 1, 2, 3. A border of 0 doesn't work with random building angle. In the villages above the maur build houses facing north and a border of 0, all other face centre with border 1. The spart have always face centre and border 1. Also interesting on small maps is facing map centre. In mind a have an agora for the greek cultures, may have a monument or a tower in the middle. Also, idle units can met here or healer do their stuff. I think, a few simple rules might be enough to layout a complex looking village. Ideas, proposals below

I think the Thomas Theorem applies here: If men define situations as real, they are real in their consequences. In other words: When men believe in gods, there are real and physical consequences. On the other hand, looking at Europe right now, I'm happy for every little space where gods don't play a role. Even if that place is called 0 A.D.....

It's peanuts/1000. Computers are fast with math. There's little memory involved and probably SpiderMonkey caches n values. Open the console of your browser here (F12) and copy/paste this code: var i = 2400, r = 0, t0 = Date.now();while(i--){ r += Math.pow(0.9, i);}[Date.now() - t0, r]i is the amount of units on your map, worst case is all are fighting. What you get as milliseconds?

> How can i move some soldiers That's pretty simple. I copy the API (entity.js): Engine.PostCommand( PlayerID, { "type": "walk", "entities": [this.id()], "x": x, "z": z, "queued": queued }); If queue is true, it means the AI has "shift key pressed". PlayerId: a integer, 1-8 are meaningful, entities expects an array with integers and x, z are map coordinates. > or get a list of all the soldiers? A bit harder. Assuming you mean units which are part of class "infantry", you loop through all entities, check the classes, collect the ids in an array and feed it into the code above to send them around.

> A timeline of unit counts Check out the charts mod: http://wildfiregames.com/forum/index.php?showtopic=18763

> ...yet another argument for shared libraries for simulation and RMGen IMO I wish, I could use my JS helper lib for simulation, rmgen, trigger and bot...

When I reach pop max, I pile resources and prepare for combat. Then I put the rally point of all trainers to somewhere close to the selected battle field and order another 100 units in batches of 5 or 10, ignoring blinking pop counter. Each unit lost in the battle is then auto replaced and appears where needed. 3 barracks * 3 lines * 8 batches * 10 units sounds epic.

I like the idea of ranking the most outstanding features of the civs. Probably the most useful information is hidden in the design documents. I've consulted them often, but still fail to memorize their content. Also, with random civs the first click is usually the diplomacy dialog to find out which civ I actually have. Well, except Mauryans, their blue colors + unique architecture + mobile dropsites make them easily identifiable. A feature rank would help me to easily make a quick plan, because it avoids selecting units and find out what they actually can build. The upcoming structure tree will definitely help in this regard. Even better would be a cheat sheet with all the relevant information on a single page ready to print out and put next to the keyboard. Out of thin air I'd suggest the civ name, a visual indicator to link to the screen and the top three unique features, if exist. The elephants are a good example because there is nothing to counter that feature. Many civs have specials which can neutralized with some other trick. I think, this kind of information is useful later in the game and once a player has some experience. Since a structured civ feature table will help me to make my bot project more interesting, count me in. PS: Another line of thought at game start is: Wow, lots of stone here. Does it help me with this civ?

No, I'm interested in your thoughts whether area expansion can be somehow balanced against health points and shooting power.

Bots too can apply Panando's no-wall-span exploit: Is this cheating?

wgoyc: How do you deal with TerritoryInfluence of towers in your mod?

> Nice analysis Indeed. That's probably all you need to know about towers. Is there a best of forum category?

I've found random trigger maps are superior to manually designing maps for bot testing. I can easily change number of bots and civs from the commandline without editing XML files. But that's the random map part. With the trigger part I even managed to change the resources before bot initialization and force research of technologies at game start. So a bot can build fortresses right away from start instead of collecting all needed resources and raising a town first. I'm sure I'll detect more useful possibilities soon and add them to 'brainland_triggers.js' as from trigger map code the simulation is available and Hannibal's groups scripts will make it a good choice for smarter map natives. If interested grab the files from git: https://github.com/agentx-cgn/Hannibal/tree/master/maps/random