elexis

I have a body that wants to make more of that (more or less now, see oppidum map). It should become player 9 however, not gaia, so that the enemy has territory as well and possibly diplomacy interaction. (Also full blown campaign mode assuming one wants to live as a human-computer-interface user for probably a year or longer)

This map is really good and we should hence probably have it in vanilla. I didn't look at the code yet, but here some improvements from the match I had played with you and from what I can see. Forests way too dense (hard for units to pass). Most likely too many trees overall (danger of lag). Random animals, mines, berries, actors, possibly bumps and hills missing The paths end harshly (red arrow), the path should appear more organic. It would make the map more traversable in strategically interesting ways if there was a small area just above water level near the cliffs (orange) where the player or his neighbor could build a dock, possibly gather some resources. The other side could be reached through shallows (green) and a small steep path to that mountain (black). This path could be so small that only 1-2 units can pass at a time (possibly no siege engines). This way it would be easy to defend but a possible attack through the backdoor if left unattended. It would allow trade without going through the middle of the map (which will be most often unfeasible unless having won already).

Noticed, we have improved the ship ungarrisoning mechanism even. It also occurs on other maps (Hannibal mentioned Hellas, although I dont recall that being buggy). If I recall correctly, the problem were maps with waterdepth that is greater than very small. The shoreline became to steep or something. I think we don't have a ticket for that yet, do you want to write us a reminder? http://trac.wildfiregames.com/

The map was created by @(-_-). As far as I recall the things were rather used as decoration than story elements.

Such balancing tests can be done in the Atlas map editor (which you can start from the main menu). You can place some units for player 1 and some more units for player 2, set the ones of player 2 to Passive stance, set the diplomacy to enemy if it isnt already, then click on the Simulate or Run button or whatever it is called, and then they will fight. You can click on reset to restart the fight.

If the same key is assigned to multiple hotkeys, all of the hotkeys are triggered when the key is pressed. Whether the hotkey does something is up to the code assigned to it. Building hotkeys for instance should only be triggered once the GUI is in "building placement mode" after pressing that hotkey; i.e. B for build, H for house. So this particular example would unlikely occur unless the user configured it particularly weirdly. But in many places of the current code that's already the case and also the reason why there's the "shiftlag" issue.

It could be B -> build, S -> special building, S -> next special building and the user could specify which special buildings S binds to.

Depends on the implementation. As (-_-) mentioned it's very weird if you have the same hotkey configuration but it equates to a different unit or building for each civ. It should be something like B for blacksmith for all civs. But since all civs have different civ templates, one can't use the template name as an identifier. It would either have to be an Identity class - or what I came to conclude - a hotkey identifier in the template names, such as "hotkey.building.tower" or "hotkey.building.gate_ishtar". (It's what I intended to code after the hotkey editor which I coded in 2017 but never released.) This way one could control have the game with keypresses.

All hotkeys are listed in https://trac.wildfiregames.com/wiki/HotKeys but just enabling it in the ingame options should do.

On the legal possibilities of having GPL licensed games on Steam, consider their documentation: https://partner.steamgames.com/doc/sdk/uploading/distributing_opensource. GPL requires modified versions of GPL licensed works to be licensed under a GPL compatible license too, but Valve does not want to have their Steamworks API code released (also their application key): https://www.gnu.org/licenses/old-licenses/gpl-2.0-faq.en.html#MereAggregation So GPL licensed projects can only use Steam by not using most Steam features and adding odd workarounds to use anything of that. Some of the features seem to be unusable entirely (for example the GPU drivers, screenshots, probably voicechat) since they would have to communicate to the separate wrapper process. So to implement video or voicechat using Steam one would have to implement a custom voicechat protocol, but then why not reuse an existing p2p voicechat protocol without reinventing it and being independent of Steam? It's many small steps from not using anything of Steam and making Steam our hoster and software API and having main features of the game depend on Valve. Other GPL licensed project examples: Warsow was submitted to Steam Greenlight on February 9, 2013and was greenlit on September 18. https://steamcommunity.com/sharedfiles/filedetails/?id=125965803 but never released. Xonox is not submitted to Steam because they want to make it more stable first. They also rule out to use the Steam multiplayer services because they can't want to ditch their own platform: "There’s just no way currently we could do two way synchronisation between Xonstat and Steamworks - extracting data from Steamworks might be possible, but injection from Xonstat is impossible. In the end we might have to choose to use Steamworks and ditch Xonstat - obviously this is totally unacceptable, it’s out of the question." http://xonotic.org/posts/2013/about-xonotic-and-steam/ Blender and DosBOX - presumably don't use any of the Steamworks services, are downloadable free of charge at Steam and hence must be there for the popularity. They are also automatically updated, can only be used with Steam being started and Steam submits usage statistics to the developers (which seem like weaker reasons to use that platform in comparison to the popularity). https://store.steampowered.com/app/365670/Blender/ Worthy to note that Valve asked Blender Foundation to come onto Steam: https://www.reddit.com/r/Games/comments/1kbn7y/valve_wants_to_integrate_blender_into_steam_games/.

There is currently only one plane waterlevel. People have worked around this by placing water actors from a mod: https://wildfiregames.com/forum/index.php?/topic/17326-your-screenshots/&page=13&tab=comments#comment-346126

The supertuxkart Steam state wasn't clear to me. Here a summary from their blog and github: February 23, 2017, SuperTuxKart is on Steam Greenlight, http://blog.supertuxkart.net/2017/02/supertuxkart-is-on-steam-greenlight.html March 1, 2017: "WE ARE GREENLIT!!!", http://blog.supertuxkart.net/2017/03/weare-greenlit.html April-May 2017 Steam-Wrapper repository: https://github.com/hiker/steam-synchron-manager Steam TODOs on github: https://github.com/supertuxkart/stk-code/projects/2 July 8th 2017 on #supertuxkart as quoted on https://wildfiregames.com/forum/index.php?/topic/16476-steam-greenlight/&tab=comments#comment-333867 August 24, 2017 Steam Release Date? https://forum.freegamedev.net/viewtopic.php?f=17&t=7591&p=74286&hilit=steam#p74286 November 20, 2017: SuperTuxKart 0.9.3 released http://blog.supertuxkart.net/2017/11/supertuxkart-093-released.html Summary: supertuxkart is not available on Steam yet, the GPL licenses are incompatible with most Steamworks API features, can partially workaround it with a separate non-free standalone process, the team needs to register a legal entity and figure out how to apply the donation policy to Steam. I doubt one can have the Steam linux GPU drivers that were wished in February 2017 in a GPL licensed application due to the stand-alone process workaround. Commentary: At least some of the supertuxkart devs seem to wish to use the cloud, useraccount and leaderboard service. It seems a bit against the spirit of the GPL that one could not host a multiplayer lobby oneself with free software unless implementing the Steam features as a networked service oneself. It would be a bit nicer to be able to get the benefits (more players, donations?) from Steam without getting too many disadvantages (putting them in charge of the project directly legally or indirectly by service or money dependency). The next post will show in how far Steamworks API and GPL are compatible or not.

Isn't it a team aura too powerful then instead? Also consider that the nr of bonuses may increase further if the max number of players possible would be 16 or more eventually. We can observe a similar phenomenon if the gamesettings are set to 13 relics.

Mess with the elevation tool in atlas for a long time, or write a rmgen script that for all X and Y coordinates sets the height of a sphere at (X,Y)

A map in the current pyrogenesis engine is a 2D grid of X * X tiles, each vertex can have a different height. The heightrange is big, so you can create a curved map, but it's not possible to have two different terrain locations at one place (overhang) in the terrain. We're updating our privacy policies for GDPR until we start alpha 24 development.

You didn't create a new random map, those are created with code and are saved in maps/random/. When generating an existing random map in atlas, it computes one specific map of all the maps that script can generate, and that can then be saved as a Scenario or Skirmish map and you find it only by the map Title (not by the filename) in the ingame menu.

So it seems the bug is becoming more relevant with those zombie defense maps and should be addressed with a higher priority than before. At some point it would be cheaper to fix the issue rather than answering why the bug is there. Anyone already eager enough to fix? At this point thanks for the individual members of the WFG programming department who have actively fixed a number of formation and garrisoning bugs that became evident when playtesting JB. There are different possible implementations to address the issue, #2001.

Square maps have several problems: square maps of size X have area X², but circular maps of size X have the area X² * pi/ 4, i.e. square maps have 25% more size which yields unintended consequences like more lag and more cost to attack the enemy, more trade income minimap is made for a circular maps squaremaps have corners which are special locations that are hard to reach. that impacts the gameplay unless its a special case (like corsica vs sardinia or latium). On circular maps, the greatest distance between two points is X, on a square map SQRT(2) * X², i.e. 40% more walking distance, while still most of the relevant locations of gameplay are within the circular part Don't know if there are remains of atlas square map support. But you could probably hack it by generating the latium map and then deleting the contents and work with the square thing. About the 6 year old terrain flattening patch, that is not only outdated but also incomplete, because it lacks serialization. It's not a feature because the implementation wasn't finished, maybe yet.

Ah that, partially yes. They're probably beneifitial in general, but they can't be used to create the arbitrary paths as seen on the screenshot between buildings, unless the inner oppidum is one single actor (but then the collisions are weird and houses should be destructable or capturable individiually). In that case one could create the oppidum in atlas and place that as is in rmgen. Still would be most flexible to have to have an algoirthm doing that, it should be attempted once.

This should be considered a bug not a feature that rightclicking doesn't open the dialog for every item. Only researching the tech / training the unit should not be possible when there aren't any resources.

Sorry I missed the update. A patch / diff file is really easy to apply, one can execute "patch -p0 < filename" on linux or click on "apply patch" in tortoise svn on windows. But the patch linked above is 6 years old, that most likely doesn't apply anymore (and likely didn't implement something ready to use. From reading it it seems like it could work except for multiplayer and savegame loading). Possibly, there may be technical bottlenecks and difficulties that are only seen during implementation phase; but possibly. The question is whether I will try to, then whether I can comes after that. You can assume there will be 2-6 weeks going down the drain when also implementing a fun map or similar interesting applications.The petra AI would also likely need some changes at which point we may consider involving our last AI dev again. Compiling will be the easy part however.

Who are the pogs? LordGood can you upload the paths (see filenames above) as a file attachment in the forum post? I guess the textures are already committed for other actors?

If anyone has access to the notes, it's Omri. I had used the SonicVisualiser in order to determine some notes of the guitar on that catchy Elysian Fields riff, but it's a bit timeconsuming.

Right click works if and only if leftclick works, left click works if and only if the button is not disabled, and its disabled iff. there aren't enough resources. So we'd need disabled_left and disabled_right or something like that.