elexis

The mentioned commit fixed it. It should work well for nomad, players get 4 extra minutes before the first attack occurs. Each gaia attacker group is ordered to go to a random CC/Wonder/Monument (foundation or finished building) and then patrol random points of the map. Depends on the victory condition? The game is considered finished if you wipe out the enemy as usual on Conquest. 0AD barely performs acceptable if there are 8 players with 150 pop, but the map is designed to have same amount of gaia attackers simultaneously. There isn't much that we can do besides writing some more new pathfinders and fixing some bugs to need less gaia attackers. Does that replay contain a bug other than infinte lag?

The logs don't really contain any information that indicates why a connection failed. Can you establish any connections between the computers at all? ICMP uses IP, so that's good. But can you get any UDP (or even TCP) connection at all? (The game uses UDP). I'd suspect it's a network configuration, hardware or firewall issue. You could have at the actual packets sent and received using wireshark. I've never seen two computers connected without a switch or router. Do you actually use an https://en.wikipedia.org/wiki/Ethernet_crossover_cable which seems to be needed for that? Two clients on the same LAN joining through the lobby has often been reported to not be working. Dunno why

It shouldn't be too hard to extend that support however. In Anno 1602 you had many resources, only the most important ones were shown in the top panel, the rest in a menu:

Do you have it in the replay too? Can you reproduce the issue somehow?

I'd be surprised if I was the only one connected or the only mod connected. I didn't ban any seigneur account, so maybe @user1 knows.

No idea how this could happen. Can you reproduce it reliably? Does the replay show the same issue?

How balanced is this map without treasure anyway? Fixed in rP21615. Attackers patrol, first target is a random CC, Wonders or Monument, then they patrol random location on the map after they killed everything in sight. So the units can only survive at the docks if the gaia units are all wiped out or can't reach that side anyhow. IMO it's a feature.

replay? Reproducible with a replay? (Don't see how it could occur, but such a thing might happen when one modifies the code while running the game)

All replays are saved, but if they don't start at the beginning of a match (savegame, multiplayer rejoin), then they are not shown and can't be replayed yet.

Most of the tracks? There are like 3 peace tracks that are played for all civs (I don't like that either) and every track is played for 1-3 civs, so that you don't have to listen to the same tracks again and again every 8 minutes. You can find the revision on Phabricator by looking for the revision history of music.js.

Thanks for your feedback, as mentioned it's the mirages which we forgot. Until we have fixed or reverted this thing, don't place foundations in the fog of war.

(It happens when a foundation is placed in the fog of war.)

@wowgetoffyourcellphone reported the same. Did you compile or use the most recent autobuild? Can you send a replay? Which civ are you talking about? (I can't reproduce the issue.)

Oh oh

(Polished and committed it)

Writing trigger scripts is not too easy yet (but it becomes easier with every triggerscript map we add). The tutorial maps have objectives and scripts depending on unit orders and positions if I recall correctly. The random maps Survival/Danubius/Jebel Barkal are mostly about spawning masses of dumb attackers. That would require a player with a locked diplomacy, which is currently only supported for scenario maps, but might be changed in some distant future.

Oh I had never seen it ingame. The hack seemed to work for me. So it's something for @vladislavbelov then. The GUI backwards compatibility for replays and savegames is kind of useless, but for the lobby an attacker could deny the service with broken JSON. But it will be quite a lot of code to address every possible issue that could arise in gamedescription. It would be quite ugly so meh. We never had someone trying to be that funny and if, then he would be banned. Damage would be quite limited (meh)

That would be something for @s0600204. (That language seen on that screenshot is german). About a22 games in a23, we don't have any sanitization for displaying lobby game attributes. The clients might send any kind of garbage and we get JSON errors repeated. Usually we solve this problem by removing the broken game, but it's probably better to add conditions all over the place in the long run.

darn, wraitii had a somehow working patch for that which was never committed! D21

@(-_-) uploaded a patch for the Gaia entities already. I'll fix the mapname. String changes means english strings that are translated on transifex, we have some deadlines with regards to those.

Since there were merge conflicts, it means that at some point in history you most likely changed something in your directory. If you don't care about that anymore, just do a revert and delete unrevisioned files. https://tortoisesvn.net/docs/nightly/TortoiseSVN_en/tsvn-dug-revert.html https://stackoverflow.com/questions/613485/how-can-i-get-tortoisesvn-to-get-a-list-of-files-that-are-not-in-the-repository

unrevisioned files = files that aren't marked as part of the repository. like a new template that someone wanted to add sometime but wasn't actually added but you still got the file around. patched file = file that differs from what is in the repository

Merge conflicts only happen when you at one point patched your working copy, then updated it and the update changed the same files in the places you have edited. Than can include a leftover unrevisioned file. If you want to preserve your diff, you have to carefully proofread it. If you weren't aware that there was a difference, you should revert your working copy regardless of the chosen merge resolve action. Should also look for unintentionally leftover unrevisioned files with the file "status" somewhere in tortoise.

Imarok if the tree obstruction size is increases, then you have people chopping thin air. Same issue exists when there is clearly space between trees to walk through them but an invisible wall is blocking you. Can create impassable forests only by making them so dense that there is no space between the trees / bushes (asking for lag).

We would need a bit more information to interpret the error. Right now it only says "Segmentation fault" without any context. There are some logfiles which may contain more information: http://trac.wildfiregames.com/wiki/GameDataPaths A callstack would be what we need, to see which line of the code exactly crashes: https://trac.wildfiregames.com/wiki/ReportingErrors