elexis

Correct. If heroes can't be trained again and have a global aura, they will be parked in safety. @fatherbushido and me addressed this last alpha for the most important heroes already (in particular Boudicca and the roman +20% attack guy).. Apparently there are still many left though: So we should change athen_hero_iphicrates, brit_hero_caratacos, cart_hero_hamilcar gaul_hero_brennus, gaul_hero_britomartus. (with the range of the size of a city), mace_hero_demetrius, pers_hero_darius, ptol_hero_ptolemy_IV, rome_hero_maximus, sele_hero_antiochus_great, and not change: athen_hero_pericles_2, athen_hero_themistocles, mace_hero_alexander, ptol_hero_ptolemy_I_2. Also @wowgetoffyourcellphone, if you create an account on code.wildfiregames.com, you can propose, review and accept patces like this one too.

@fatherbushido wanna commit your diff then? Health + cost changes at D102. Couldn't convince myself to make catapults 250HP, a quick test with 400HP has shown that a single sword unit takes really too many hits to destroy it (I believe catapults should melt when encountering sword units). Catapults can easily become OP if they are too hard to destroy.

Thanks for the screenshoooots! Those incentivize players much more to download your content. Some quick remarks: No matter what's the deal with the map, there should be 20-40 initial trees and 1 stone + metal mine There are whales / sharks and fish? Giant maps are just too big Terrain shapes look interesting Avoid repetition and empty areas. The eye should always have something to look at, independent of the viewed area. The forests on Acadia are too dense, give the players some space to build a city. Perhaps some gaia settlements could fill the void and provide strategic incentives And yes, 0 A.D. lacks an online map repository! It is too hard to get custom maps played in the lobby.

Map: Canyon (2nd try mountain climb) Players: (wang_wei, Hannibal Barca, siole, elexis) vs (fpre, Vercingetorix, Grugnas, FeldFeld (later bb)) Summary: 100 minutes gametime (actual game much less). This time it was a game with only two teams, so walling and trade was way more effective. Unfortunately one team took out the middle of another team, so it was gg at that point. That however didn't bother me as my mission was to build a nice base on that mountain. Enemies delivered messages in wall codes or mostly left the game due to boredom. While there were some tries to reach our city, only bb who rejoined as FeldFeld was succesful at defeating us. From what I recall, he didn't use buildings but a ram to enter the mountain with some champions, captured a house, built a barracks and eaisly took our civic center. So people on the mountain are not untouchable, nor hard to defeat at all if you know what you have to do. Also I repeated a mistake here, which is not garrisoning the towers after losing the civic center at the bottom. Hence my own towers on the mountain attacked my trade carts! The toughest challenge is to use the few space available on the mountain wisely. One corral was used with batches of 15 sheep or more, the rest went into houses. I should have really built a temple though to train champions. In particular I lacked the health regain upgrade. At the right side is siole again. Notice it is possible to get down from the mountain the same way you get up there. Just that there are less viable places. 2017-01-23-canyon_4v4_loza.zip

Map: Canyon (1st try mountain climb) Players: (wang_wei, Hannibal Barca) vs (Merov, SaltySeyi) vs (siole, elexis) vs (Grugnas, Liberty (aka ZhengfuZhe)) Summary: 90 minutes gametime. On this map, each player is surrounded by mountains. The middle is valuable due to the forests. But as small passages connect allies, trade is safe alternative. Walls were allowed and utilized. In this replay, I made my first try to relocate my city onto the mountain: The other replays following this posts are more fleshed out though. In case you ever decide to visit that godforsaken place using a bug, consider that those bumps on the mountains are only reachable from one small path (if at all), so don't block that path by constructing a building at that place! The tiny paths between the bumps can be used for trade if lucky, but that often gets stuck due to the small area: After the winner of the match was clear, he had to get up to that mountain to wipe out us escapists. After finally having succeeded with that and building a fortress there, he found out that the area was disconnected from our city. This resulted in a mexican cheat standoff: He used the developer cheat to delete my civic center, I paused the game and quickly typed the debug command to exit the game for every client without notification before he could delete the rest. What an ending xD Thanks siole for providing the full replay! 2017-01-22-canyon_2v2v2v2.zip

screenshooooots!!

A hero is a unique unit, but the game doesn't reflect that. If it dies, it can be trained again and again without any repercussions. Since the hero is the most tanky unit, it is the first one to be trained when having built a fortress and usually just sent behind the enemy line. Instead, it should have a significant impact if the hero vanishes. It could either A ) be trained only once (i.e. if the first trained hero dies, only the other two hero choices can be trained) or B ) the cost and train time should increase with every death (for example double) This change has been widely discussed a lot in the lobby (ping @Grugnas, @Hannibal_Barca, @Vercingetorix_, @borg- , @fatherbushido, siole) and is with some exceptions agreed upon. @scythetwirler do you agree with the idea? How the change is implemented code-wise is another story (but in both cases, the Player component should keep track of the trained heroes). ("we all got sick of the 10th boudicca dying to mace champs. you kill her and 2 mins later she is there again")

I agree that they are too hard to use outside of the range of ones own fortress to increase catapult and bolt shooter health from 100 to 200 or 250 (since rams have 400 and still vanish quickly) to reduce catapult cost from 350 wood 350 stone to 350 wood 200 stone (as rams cost 350 wood 200 metal) to give bolt shooters 75 pierce + 5 crush linear splash damage, as planned in #4328 but disagree with making them outrange towers making them so tanky that they are not easily destroyed by sword units anymore increasing their accuracy in particular, since all units should receive a higher accuracy Would need an ACK by @scythetwirler or @fatherbushido before changes can take effect.

(don't forget uploading replays when posting bugreports, though not needed in this case)

We already have more hotkeys than keys on the keyboard, some are UNUSED. But it's a good thing, because it allows players to chose which keys they want to use. The hotkeys need to become configurable in a new GUI dialog though, as few people go as far changing their config files. After that, we should indeed add hotkeys for specific buildings in my opinion. It should be possible to assign B to barracks without having to fallback to the group hotkeys which have to be reset manually every game. We have to decide which entities will be available though and how they are going to be linked. There are hundreds of units and buildings available, so likely many useless choices. Which identifier would we use (without adding custom identifiers making the code ugly)? How do we avoid distinguishing the civ specific differences (barracks should work for all civs.) One option would be using the Identity.Classes or Identity.VisibleClasses information, but that also contains useless choices, like Village (for all village phase buildings) for example. Seems like we have to add a custom (sub)set of identifiers :/

Sounds like a graphics driver issue

1. Kick / ban unwanted clients using the /kick <playername> or /ban <playername> command. Notice the rating is part of the nickname, so use the tab key to autocomplete names. 2. If you want to play a private match, hand out the IP address to those that shall join privately and don't enter the multiplayer lobby in the first place (main menu -> multiplayer -> host / join) 3. A network dialog is planned, so that you can kick people by clicking on things.

I'm thinking more of a new dialog that displays a list of all units owned, ordered by their amount. For example 40 Skirmishers, 40 Spear Infantry, 40 Women, 12 Traders, 3 Healers, 2 Ships. Buildings could be added to a separate tab. Perhaps we can add a new dialog for resources, as the resources of allies are currently buried in the tooltip and noone knows about it and since seeing the number of gatherers per resource type is useful for balancing ones economy or team economy.

New autobuild for svn out there -> http://trac.wildfiregames.com/changeset/19172

The proposed fix to this issue has been committed in 19160, see also #4408 and D42. The autobuild has some issue (@Itms), so compiling would be needed if you have SVN to test, otherwise wait a bit for the autobuild (or wait to the next release if you can't use svn).

Rejoining a multiplayergame with AI inevitably yields an out-of-sync error, as that isn't implemented. (Since related code is about to be changed, we could at least move that message box to the javascript GUI and display that rejoins with AIs are not supported at all.)

(Aren't the green cloths in portrait a bit bright and oversaturated? Can't imagine cloths with that tone 2000 years ago. Looks good otherwise. Also wondering if the eagle symbol would have been put there if it were crossed by some thing on top it on the shield.)

Thanks for the report, I marked it for sooner, rather than later #4459

3 hours is a lot though. If you can, defeat the economy of your opponents early on (women only have 25 health!). Citizen soldiers are much more useful in this release now as they are not mowed immediately by towers anymore. In order to completely wreck your enemy, you have to take out the city center with a ram that is defended and repaired by your guarding citizen soldiers. You might also want to take a look at the replay thread to see how lobby players reach city phase in the first 15 minutes and wipe each out each other a bit sooner:

I'm not an AI dev at all, but I'm following development a bit. There is some existing code about personality traits iirc (for example degree of agressiveness) and plans / ticket to extend to such characters that were suggested above. The "lure the bot into your towers" phenomenon is a classic that is considered we are well aware of. It's not always easy to fix things in a performance efficient way, we have to check so many things as an AI.

Either way, the ticket is still open and planned to be implemented, just differently, better.

I was wondering already whether gameboy left

The tech first loads the starting units (4 men 4 woman 1 cavalry) and then recursively all units that they can train, all buildings that they can build. So if phase_town_roma isn't found, that should indicate that the phase can't be researched at those entities, does that make sense anyhow? @s0600204 is the author of the tech tree and the most recent requirements rewrite, so might want to ask him.

> Inside of the market when you obtain Cartography you can view resource counts / population - something pretty important on deciding best allies. Indeed, it's one of the most crucial elements in team gameplay already. Often players have 0 resources left of some type while allies are drowning in that resource. Communicating via chat is hard work. Even when a resource is short for all team members, they can more easily estimate whether they should send over even 100 resources of that type. - There might be another layout to the Diplomacy window which should list who is at war with who, this information is not clear at the moment. That isn't even clear if you're in observermode. Every player can have arbitrary diplomatic relatios to everyone else. Also the diplomatic stances aren't often applied. Sometimes there are enemies who act as friends by fighting the common foe. At other times there are allies building towers and fortresses close to the ally to prepare a preemptive strike. - It would be great if you could "surrender" a structure to an ally or have the ability to build a true "embassy" type of situation where you could own/control territory inside of an ally, etc. The "giving buildings to an ally" part was ruled out, because it would allow cheating too easily and remove the territory restriction. There is a ticket for it.

Fixed for next release: #4146