elexis

The number should be some timer. Someone else had the same error message before with a custom scenario map (#3015), but there can be various reasons triggering that exact message. You could backup your map (XML file), and then remove the bottom half of the file, check if there is a bug and repeat that process until you found out which entries are bugging. A similar approach would have to be done for the gamesetup attributes at the top of the file to find the buggy entry blindly. You should probably post your map here. Perhaps Stan finds the bug immediately when looking at it

Map: English Channel (Naval Blockade) Players: (arcadius, Hannibal Barca, lou_diamonds, elexis) vs. (Finch, bb, Vercingetorix, dkaiser92) Summary: Example for a successful naval rush (garrisoned ships at minute 5), resulting in an all-encompassing naval blockade, ending the game strategically in the first 10 minutes. The blockade ensures that enemies will never be able to build docks and ships. Uncontested maritime superiority allows continuous lodgement, regular interception of adversary trade and the build of Civic Centers on the undefendable territory. Heavy warships garrisoned with catapults finally wipeout the last attempts to fortify the landing area. english_channel_naval_blockade.zip

It's likely a missing game attribute. Does the error state in which file it occurs?

The code looks performant enough. The Math.pow() operation is only done when SetBuildMultiplier is called (i.e. whenever a builder is added or removed). Repair and GetRepairRate are likely called more often on average (and these functions are trivial, just 4 divisions). Useful 0 A.D. math includes computing when economic upgrades are worth it in comparison to creating more workers, or comparing the profit/cost ratio of trade carts with gatherers, or computing how many units survive when outnumbering the enemy using Lanchester's laws.

Should be possible to implement your approach, with or without prior wall placement refactoring, mostly lacks manpower though (or womanpower for that matter).

Might have to warn you that continuous investigation of code behavior might lead to bugfixes. http://trac.wildfiregames.com/wiki/BuildInstructions http://trac.wildfiregames.com/wiki/GettingStartedProgrammers I suggest though to lookup the components and templates in binaries/data/mods/public/simulation/ as it's much faster than a blackbox analysis. The buildtime depends on the number of workers, on the units that work individually (at least the code supports that) and the buildings HP. Also auras and technologies can affect the build rate - for example the female inspiration aura (10% faster males in the range of females). Most importantly, the more units are assigned to a building, the slower they work individually (hence some pro players build each structure with only one unit, so that the construction finishes just when it's economically feasible to use it at capacity). http://trac.wildfiregames.com/browser/ps/trunk/binaries/data/mods/public/simulation/components/Repairable.js#L74 Repairable.prototype.GetRepairRate = function() { let cmpHealth = Engine.QueryInterface(this.entity, IID_Health); let cmpCost = Engine.QueryInterface(this.entity, IID_Cost); let repairTime = this.repairTimeRatio * cmpCost.GetBuildTime(); return repairTime ? cmpHealth.GetMaxHitpoints() / repairTime : 1; };

But did you rule out that it aren't the new maps? It is unexpected that the performance-heavy change would be introduced in Alpha 20 or 21, while Alpha 19 was the one with a performance sensitive change (new pathfinder). Perhaps @mimo also got some idea whether there was some change that could have influenced petras performance.

If the code is inherently slow, the fastest CPU won't help. Most of the players have 100% workload on one core. What you could try are replaying games (Menu -> Tools & Options -> Replays) and check whether the old games are still 'performance-lag'-free while the new replays do lag. In that case it were confirmed that it's a performance (not network) issue and that it can be determined with the gamesettings.

Well if it's really the AI, then it must be a performance problem. If you have more than 5 FPS and if the game isn't freezing every few seconds briefly but only showing network issues, then I suspect that the AI consumes too much performance on the other computer. Indeed a smaller mapsize (at most "normal", better "medium"), a map with less trees, less population (at most 8x150, better 8x100), less AI players should significantly help. If it's obvious that it isn't a performance bottleneck, then I'd doublecheck that it isn't a network issue (you could for example keep a "ping <ip-address>" running in terminal while the game is running). It is important to note that rejoining games with AIs doesn't work correclty (it will trigger an Out of sync error, players end up in different game states). So only play multiplayergames via IP if you're sure you won't disconnect. Also you are welcome to play games in the lobby with human enemies if none of it helps.

WIFI?

The correct phrasing would be relative vs. absolute bonuses (and mark my words with the chariots ).

#4431

If you have downloaded the source code using SVN, you can simply do an "svn revert -R" to reset all changes you made (with the exception that it doesn't remove new files added). @wraitii upgrading to sierra currently, so perhaps he can keep you updated.

Broken in 19043, autobuild from r19046, reverted in r19054, refs #3785. Triggering a new autobuild now.

See also (pre)reviews by FeXoR and me in #4354

Also to me it sounds like @Libervurto is looking for the ceasefire option.

Mostly satisfied with the game mechanics. Agree that we need more (than zero) technically skilled visionaries compatible to the developers and players pov. The pace of the game seems slow already as one needs to reach city phase, build a fortress and construct a ram and destroy the enemy civic centers in order to win (which was possible with citizen soldiers in a17 and possible with champions in a18-a20). If we deincentivize early attacks, the game would become extremely boring to competitive players. Singleplayer games target casual players who might play less games but expect longer ones. So this is mostly influenced by the aggressiveness of the bot. There are plans to allow setting the behavior of the bot #2550. It could become defensive by default. Also if there is a singleplayer campaign, none of this will apply anymore.

Don't have access to OSX, unfortunately only @wraitii currently has. Can only link the ticket #4408 which has a patch attached, which apparently is ready to be tested (see also BuildInstructions#OSX).

Doesn't look like SDL2 supports them: https://wiki.libsdl.org/SDL_Keycode https://wiki.libsdl.org/SDL_Scancode https://wiki.libsdl.org/SDLScancodeLookup

Oh wait he could connect once - in that case it sounds like he becomes disconnected in the loading screen as the map was too big or something (the better the ping, the less patient the server will be disconnecting clients).

If the connection can't be established, the port forwarding rule is wrong (probably for the wrong IP address) but there might also be a local firewall blocking things.

Valid keys are defined in KeyName.cpp#L40, not sure what World64 is supposed to be

If you see the replay, I could rebuild my entire base and thus could have helped killing enemies as usual, I was just too slow as I forgot 200 stone. I believe it should be possible to lose only 7 minutes when executing this correctly and perhaps stacking more resources, doing upgrades and building more houses early (as they will be recaptured). One possible application might be regicide games where one might receive an ungarrisonable hero elephant. Presumably one can rebuild the better heroes after defeat and thus avoid the possibility of becoming gaia in the middle of a match. (Still the player won't be able to change the fact that it's defeated afterwards (which was actually possible last release)). Nerfling slingers, perhaps, needs some playing dev and another who agrees.

Map: Cantabrian Highlands (Surviving Defeat) Players: (Vercingetorix, siole, yunk2, elexis) vs (arcadius, Valeranis, matialejandro12, marson126789) Summary: Post-defeat comeback demo This replay demonstrates how a player can become defeated and yet rebuild an entire base (#4384). Since being defeated doesn't help after all (team chat disabled, resource tributes impossible, needs small GUI code change or cheat to train units and the player will be still declared as a loser of the match), but it was fun to try. Thanks siole for testing, this was the fourth try (still messed up as 200 stone were forgotton). 2016-12-17_defeat_survive.zip

Flying Wonder: #3913