[Freeciv-Dev] [patch #4224] Max units on tile requirement type
Follow-up Comment #3, patch #4224 (project freeciv): I would like to be able to set different limits for different types of unis. E.g. only 3 tanks fit on a tile or 2 helicopters but only a total of 4 ___ Reply to this item at: http://gna.org/patch/?4224 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4144] freeciv-ruleedit
Follow-up Comment #1, patch #4144 (project freeciv): How about freeciv-editrules, freeciv-edit-rules? ___ Reply to this item at: http://gna.org/patch/?4144 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #20983] single player vs computer
Follow-up Comment #1, bug #20983 (project freeciv): Unless I'm mistaken the researched bulb are not lost when gaining a tech through diplomacy with the default settings. So while buying the tech only saves one turn it also gives a huge leg up for the next tech (since the bulbs transfer to that). ___ Reply to this item at: http://gna.org/bugs/?20983 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #20894] Idea: Mark tiles as do-not-use/force-use in the city map to help with overlaps
URL: http://gna.org/bugs/?20894 Summary: Idea: Mark tiles as do-not-use/force-use in the city map to help with overlaps Project: Freeciv Submitted by: mrvn Submitted on: Thu 13 Jun 2013 09:53:51 AM GMT Category: None Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: None Planned Release: ___ Details: I have 2 cities A and B like this (simplified and shrunk for this example): aaa ab aaAa#Bb ab aaa A is larger and has a city radius of 2.5 while B is small and has a radius of 1. The usable tiles of A and B overlap at #. A has a lot of choices of usable tiles while B is seriously limited. What happens is that A will claim the # tile while having perfectly good alternatives while B is desperate to get it. Or worse, I'm desperate for B to use it while the AI thinks B should use another tile and A should use it. Then every time A and B change size (or some other event I can't pin down) B will give up the tile and A will start using it. Now here is my idea: 1) In the city map of A I would like to flag the # tile as do-not-use, preventing the tile from being used by that city. The CMA agent and implicit reset of tiles on size change would then ignore that tile, take it as blocked by another city even if it is not. 2) In the city map of B I would like to flag the # tile as force-use. This could have 3 meanings: 2a) Must use the tile. Force-use can only be set when the tile is used by the city. If the tile must be vacated due to lack of citizens then the force-use would expire. Don't like that much. 2b) When picking tiles to use first pick all force-use tiles before any other. No Specialist may be picked before all force-use tiles are used. But force-used tiles may remain empty if there aren't enough citizens to fill them all. This would survive a city shrinking because a settler was build and regrowing. 2c) When picking tiles to use first pick all force-use tiles before any other. The CMA/implicit selection may pick any subset of those tiles if it can't or doesn't want to pick them all. So specialist may be chosen instead of using a force-use tile. Comments? MfG Mrvn ___ Reply to this item at: http://gna.org/bugs/?20894 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #20216] Improvements to Cities tab (F4)
Follow-up Comment #3, bug #20216 (project freeciv): One can show the number of units owned by a city. Unfortunately that includes units without upkeep. It would be nice if one could add a column (each) for amount of upkeep food, shield, gold of each city. The column should show the actual upkeep and in () the upkeep without freebies (if different). ___ Reply to this item at: http://gna.org/bugs/?20216 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #20890] Better way to display build slots
URL: http://gna.org/bugs/?20890 Summary: Better way to display build slots Project: Freeciv Submitted by: mrvn Submitted on: Tue 11 Jun 2013 11:35:04 AM GMT Category: client-gtk-2.0 Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: None Planned Release: ___ Details: First the progress bar for how much of a unit is build could show more information. The already build part could be left in a solid color and the amount gained next turn shown in a light color or half transparent. Secondly when a city has multiple build slots the city window displays the progress of building units with a single bar. Instead it could split the bar into x rows, one per slot, and show how much of each slot is filled (will be filled). ___ Reply to this item at: http://gna.org/bugs/?20890 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] updating rules files, wish the server would warn about issues
Hi, I recently updated to freeciv 2.4-beta2 and tried to play a old custom ruleset. Given the age of the rule set it failed to load at all and I had to update it. Updating from version 1.9 (or even older) rulesets to 2.4 was quite a leap so some things didn't work right. The first thing was that the terrain was missing colors and the server refused to load them. That was easy to spot and the error told exactly what needed fixing. I give this as a positive example. There was a problem and the server told me exactly what it was. After fixing those obvious problems the ruleset did get loaded by the server but still didn't quite work. The problems became more hard to spot. For example: 1) No irrigation effect was defined since that simply didn't exist back then. So units were unable to irrigate at all. I think the server could give a warning about effects know to exist but not used. Esspecialy such crucial effects like being able to irrigate. 2) A veteran diplomat defending a city became invincible. Back when the ruleset was written there was no veteran system at all. I've added the default veteran system but did not see that diplomats needed a custom veteran system. This caused veteran diplomats to be twice as good as normal ones. With the way diplomat chances are computed this seem to result in invincible defending diplomats. I can thing about two ways the server could warn about this: A) It knows that diplomat chances are suspectible to overflow and warn if the veteran system is in danger of such. This would require some reverse calculation to find out at what veteran_power level the diplomat becomes invincible and warn if that level is reached or exceeded. B) The code computing diplomat chances could detect that the diplomat has become invincible and warn (once) about it. This type of check should generally be much simpler but would need someone to actually trigger it to become visible. Comments? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [bug #18087] server segfault on a map with 1.3M tiles
Matthias Pfafferodt no-reply.invalid-addr...@gna.org writes: Follow-up Comment #7, bug #18087 (project freeciv): +#define MAP_MAX_SIZE 4048 Why such a strange number? Shouldn't this be 4096? You are right; it was 2048 in an earlier test +#define MAX_DBV_LENGTH (4 * 1024 * 1024) Perhaps 4096 * 1000? Also OK; at the moment the temperature map is quit problematic. Furthermore, there seems to be an endless loop. My computer tried for 5 hours to generate a map of size 2000x2000. Memory requirement was at about 700MB but it was not limiting ... Didn't you run into the problem that detecting continents (flood fill) runs into a stack overflow because it recurses too much? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16777] Having multiple units on a field resizes the large map view
Follow-up Comment #6, bug #16777 (project freeciv): What commit fixed this? ___ Reply to this item at: http://gna.org/bugs/?16777 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17110] Synchronize images for city with city radius
Follow-up Comment #5, bug #17110 (project freeciv): Why? Resize the array as needed. If you hardcode the size then you will always have either way to many entries or not enough. I can also think of having a lot more images than 4 (or actually 8 with city walls). For example risk of pollution can add a smoke cloud (0-4 none, 5-9 light, 10+ heavy) over the city and risk of plague a red cross (0-4 none, 5-9 small, 10+ large). That would require 72 images. ___ Reply to this item at: http://gna.org/bugs/?17110 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17110] Synchronize images for city with city radius
Follow-up Comment #3, bug #17110 (project freeciv): You could simply change the number at the end of the description tag (city.electricage_city_4 = city.electricage_city_3). The city radius, except for the first step, is changed by other things than just size so this won't work. But I see you have already wrote a patch to make city images an effect too. One thing though: +; The maximum number of images is only limited by the maximum size of a city +; (currently MAX_CITY_SIZE = 250). The constant is defined in common/city.h. With images being an rffect it makes no sense to limit the number of image by the size of a city. They are really independent. MfG Goswin ___ Reply to this item at: http://gna.org/bugs/?17110 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17116] Citizens roles become unrecognisable in large cities
URL: http://gna.org/bugs/?17116 Summary: Citizens roles become unrecognisable in large cities Project: Freeciv Submitted by: mrvn Submitted on: Sunday 11/21/2010 at 01:49 Category: client-gtk-2.0 Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: svn trunk Discussion Lock: Any Operating System: None Planned Release: ___ Details: When a city grows the icons for each citizen in the city display begin to overlap and become harder and harder to deciver. Maybe at a certain size it should show a single icon per role followed by the number of citizens in that role instead of showing each citizen. ___ Reply to this item at: http://gna.org/bugs/?17116 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17110] Synchronize images for city with city radius
URL: http://gna.org/bugs/?17110 Summary: Synchronize images for city with city radius Project: Freeciv Submitted by: mrvn Submitted on: Saturday 11/20/2010 at 03:14 Category: client Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: None Planned Release: ___ Details: There are images for cities of different size so bigger cities do look bigger on the map. Currenty the first change happens with size 4. The city radius on the other hand grows with size 3 (at least in my ruleset). I think it would be nice if the image for a city would change based on city radius instead of city size when the ruleset has variable city radii. ___ Reply to this item at: http://gna.org/bugs/?17110 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] memory requirement (statistics)
Andreas Røsdal andre...@pvv.ntnu.no writes: On Tue, 16 Nov 2010, Matthias Pfafferodt wrote: After the increase of the number of players and the map size I did plan a check of the memory requirement of the different freeciv versions. I finished this now and the attached pdf files shows some results. The first two pages show the RAM used by the server as listed by pmap. A savefile with the default settings and aifill = X, map size = Y at turn 100 was loaded. Only one value of X or Y was changed while the other was set to 10. The requirements increase fro release to release; but with 2.3 it goes down for bigger maps. I also tested a game with aifill 100 and mapsize 100 - the server allocates up to 900MB RAM! With 2GB ram and 2GB swap I could not play that. I run out of memory. The last two slides show the size of the savegame. For small number of players and map sizes is is OK. But increasing these numbers the size of the savegame increases and, if both are set to 100 the uncompressed savegame reaches a size of over 100MB. For such savegames the time needed to compress the data is significant. Full ACK. And the disk fills up quickly with autosave games. Some sort of rotating would be nice (last 20 saves) or some sparse keeping (keep games 1,2,3,...,10 turns ago, every 10th turn till 100 turns ago, every 100th turn till 1000 tuns ago). For the client, the memory consumption is increased from 1.14 to 2.2 (xaw to gtk2). After that, nearly no change is visible. The client uses up to 500MB, mostly for the graphical toolkit. If there are questions please ask. These statistics are very interesting. It is a good idea to keep an eye on the memory usage, since this would allow the game to scale better. Another aspect of memory usage that I would also find interesting, is how the memory usage scales when running multiple Freeciv servers on the same host. For example, does running 10 servers at the same time also require 10 times the memory? Basically yes. Although the amount shared is static. The amount not shared goes up map size and players. So if it was 9.5 times before it now becomes 9.7 times if you use larger maps / more palyers. Also the memory usage goes way up with time. More of the map is known to players and AIs, more cities and units are created and big gotos and war maps are created. Measuring the memory at the start of the game might not be the best measurement. Maybe let it play 100 turns and then measure. Or even better, plot the memory usage after every turn. Other stats that would be usefull would be statistics about number of players remaining, number of cities and number of units in the game. Maybe something someone could patch into the server. MfG Goswin PS: I also found that XOrg uses a huge amount of memory when playing freeciv. Is freeciv leaking server side pixmaps? ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17015] Would like to have Go Home command for units
Follow-up Comment #4, bug #17015 (project freeciv): Another feature that is verry simmilar. For wounded military units it would be great to have a Goto nearest city with barracks so they heal quickly. Should I open a new bug for that or is that similar enough to be fixed togeteher? ___ Reply to this item at: http://gna.org/bugs/?17015 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17015] Would like to have Go Home command for units
Follow-up Comment #1, bug #17015 (project freeciv): I would like such a command too although for a different scenarios. At the start of the game I use wariors to explore. At some point the enemy has found me and I need to defend my cities. So I want to slect all wariors and send them to their home city. That way they get evenly spread. Every city that has build a warior gets defended. Then I build up to 3 units in each city for defense (and military law under Monarchy) so over time a nice army gets build. Perfect for selecting all but one per city (all for internal, safe cities), crank up luxury and go to war. After decimating some cities of the AI it then becomes time to go home. Again it would be nice to send each unit to its home town so they get evenly spread. For the scenario initialy described the Goto nearest city (G) command seems sufficient. Or just hit (A), Autosettler move on their own and if you move them away from the cost they will just go back if that is where the best improvement can be made. MfG Goswin ___ Reply to this item at: http://gna.org/bugs/?17015 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17015] Would like to have Go Home command for units
Follow-up Comment #3, bug #17015 (project freeciv): From my experience they do loose the auto flag when they can't find work and then you can hit 'G'+'s' to hide them or 'a' to make them try again to find some work. Anyway, the feature is still wanted even if for different reasons. ___ Reply to this item at: http://gna.org/bugs/?17015 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17006] build raid/farmland with a single command, connect farmland
URL: http://gna.org/bugs/?17006 Summary: build raid/farmland with a single command, connect farmland Project: Freeciv Submitted by: mrvn Submitted on: Tuesday 11/02/2010 at 17:30 Category: client Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99 Discussion Lock: Any Operating System: None Planned Release: ___ Details: The client has shortcuts to connect 2 tiles with railroad but no shortcut to build railroad on the current tile (if no road, build road then build rail). Same thing for farmland and connect with farmland. ___ Reply to this item at: http://gna.org/bugs/?17006 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #17007] Complex goto to connect with road/rail/irrigate/mine/farmland in one go
URL: http://gna.org/bugs/?17007 Summary: Complex goto to connect with road/rail/irrigate/mine/farmland in one go Project: Freeciv Submitted by: mrvn Submitted on: Tuesday 11/02/2010 at 17:35 Category: client Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99 Discussion Lock: Any Operating System: None Planned Release: ___ Details: The client can already do goto, connect with road, rail, irrigate over waypoints. What it can't do is switch jobs between waypoints. E.g. go to tile A, connect with road to tile B, connect with irigation to tile C, connect with rail to tile D, build mine. The goto protocol between client and server is flexible enought o handle this. All it needs is a way for the user to input the commands. ___ Reply to this item at: http://gna.org/bugs/?17007 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16871] Having more AIs than nations segfaults at start
Follow-up Comment #3, bug #16871 (project freeciv): No. So maybe it wasn't the number but just the wrong format. ___ Reply to this item at: http://gna.org/bugs/?16871 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16863] faster goto
Follow-up Comment #10, bug #16863 (project freeciv): The way it currently works is that every unit is computed on its own with the map as it currently is. If there is a boat somewhere the goto will happily move 100 units over that boat to cross a channel of water. Then when the units actualy arrive at the boat they suddenly notice the boat is gone or already full by the first few units using it and abort their goto. Same goes for ZoC or simply enemy units stepping onto your path (unless you are already at war and it is the last move or something). The unit just stops. What would be nice would be to assign the unit a destination and have the server or client recompute the path when the unit gets stuck. ___ Reply to this item at: http://gna.org/bugs/?16863 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16904] Minimap has vertical scrollbar even on large main window
Follow-up Comment #2, bug #16904 (project freeciv): I'm playing with hex tiles so that will change the critical size somewhat. I'm assuming you used square tiles. I would say the minimap should expand if there is space. There really is no reason for the scrollbar if the window is large enough to allow a full mini map and the unit infos. If you fear a reoccurance of the grows to big bug then the size should be adjustable. Like the main view and message window are seperated by a grip one can drag to resize just add one of those below the mini map too. ___ Reply to this item at: http://gna.org/bugs/?16904 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16863] faster goto
Follow-up Comment #4, bug #16863 (project freeciv): If I had to solve this I would first build a list of unique units from the selection. Units that have different tiles, movement points, capabilities, ... Then goto needs to be compute only for those units. This could be still improved by only using the fastest and slowest unit with comparable capabilities on each tile. Similary when moving units build the same list. Move all units on the list visibly, then move all remaining units invisibily. ___ Reply to this item at: http://gna.org/bugs/?16863 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #2076] Crazy caravan patch
Follow-up Comment #2, patch #2076 (project freeciv): Could this be extended to show the homecity of any unit? ___ Reply to this item at: http://gna.org/patch/?2076 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16905] Goto and do XXX for caravans
URL: http://gna.org/bugs/?16905 Summary: Goto and do XXX for caravans Project: Freeciv Submitted by: mrvn Submitted on: Wednesday 10/20/2010 at 14:37 Category: general Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: None Planned Release: ___ Details: It would be nice to have a goto and do - help build wonders - enter marketplace - establish trade route - keep moving for caravans. When one has multiple caravans moving around it can be hard to know what to do with a popup. ___ Reply to this item at: http://gna.org/bugs/?16905 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16863] faster goto
Follow-up Comment #6, bug #16863 (project freeciv): This could be still improved by only using the fastest and slowest unit with comparable capabilities on each tile. I am not sure to understand this sentence, but I doubt this is a good idea. Say you have a warrior A (1/3 movement points), a warrior B (3/3 movement points) and a horsemen (6/3 movement points) on a single tile. Then you only need to compute the goto for the warrior A (to get the maximum number of turns) and horsemen (minimum number of turns). Strictly speaking the path for warrior B might differ from those taken by warrior A and the horsemen so maybe it isn't worth it. Actualy it might be good for multiple unit gotos to take paths so that the least number of units are on any one tile at any one time. I.e. if you have 2 path with the same cost and 2 units have each unit take one path. But that would be a huge change in client and server. Not part of this bug. Worth a new bug? ___ Reply to this item at: http://gna.org/bugs/?16863 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16874] shield upkeep for units is ignored
URL: http://gna.org/bugs/?16874 Summary: shield upkeep for units is ignored Project: Freeciv Submitted by: mrvn Submitted on: Saturday 10/16/2010 at 06:54 Category: general Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: svn 18227 Discussion Lock: Any Operating System: None Planned Release: ___ Details: When I start a game with default settings, build a city and then build 4 warriors then the GUI shows the 4th warrior needing one shield upkeep but the cities shield surplus == shield production so the warrior remains free. ___ File Attachments: --- Date: Saturday 10/16/2010 at 06:54 Name: freecv.png Size: 0B By: mrvn http://gna.org/bugs/download.php?file_id=10805 ___ Reply to this item at: http://gna.org/bugs/?16874 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16863] faster goto
Follow-up Comment #2, bug #16863 (project freeciv): Also when actualy doing the goto maybe the client could skip showing each single unit move. Esspecially when moving 100 identical units from the same tile. ___ Reply to this item at: http://gna.org/bugs/?16863 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16874] shield upkeep for units is ignored
Additional Item Attachment, bug #16874 (project freeciv): File name: gna_bug_16874-fix_unit_upkeep.patch Size:1 KB ___ Reply to this item at: http://gna.org/bugs/?16874 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16874] shield upkeep for units is ignored
Follow-up Comment #1, bug #16874 (project freeciv): Seems like all upkeep costs are ignored. Settlers don't consume any food either. Attached a patch that adds the units upkeep costs to pcity-usage[]. ___ Reply to this item at: http://gna.org/bugs/?16874 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16867] Fails to build: struct packet_server_setting_bitwise not defined
Follow-up Comment #4, bug #16867 (project freeciv): Marko seems indeed to be right. The generated files weren't removed in clean so an older version ended up in the debian patch and that causes timestamp problems. I've made sure they are removed now on debian/rules clean and then it builds fine. Please close the bug. ___ Reply to this item at: http://gna.org/bugs/?16867 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16871] Having more AIs than nations segfaults at start
URL: http://gna.org/bugs/?16871 Summary: Having more AIs than nations segfaults at start Project: Freeciv Submitted by: mrvn Submitted on: Friday 10/15/2010 at 16:13 Category: None Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: svn 18227 Discussion Lock: Any Operating System: None Planned Release: ___ Details: I tried playing a game with 120 AIs (one has to test out the new limit, right :). But I used an older ruleset that plain doesn't have enough nations. At start the server picks nations for the AIs and after a few it segfaults, when it runs out of nations I believe. Unfortunately I replaced the old nations file with the current default one without keeping a backup to fix the problem. But it should be easy to reproduce by commenting out enough nations. ___ Reply to this item at: http://gna.org/bugs/?16871 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16867] Fails to build: struct packet_server_setting_bitwise not defined
URL: http://gna.org/bugs/?16867 Summary: Fails to build: struct packet_server_setting_bitwise not defined Project: Freeciv Submitted by: mrvn Submitted on: Thursday 10/14/2010 at 12:00 Category: client Severity: 5 - Blocker Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99+svn18226 Discussion Lock: Any Operating System: GNU/Linux Planned Release: ___ Details: Compiling revision 18226 fails with: make[3]: Entering directory `/home/mrvn/src/freeciv/trunk/build-xaw3d/client' CC options.lo ../../client/options.c:3451: warning: 'struct packet_server_setting_bitwise' declared inside parameter list ../../client/options.c:3451: warning: its scope is only this definition or declaration, which is probably not what you want ../../client/options.c:3451: warning: no previous prototype for 'handle_server_setting_bitwise' ../../client/options.c: In function 'handle_server_setting_bitwise': ../../client/options.c:3453: error: dereferencing pointer to incomplete type ... The struct packet_server_setting_bitwise is not declared in any header and was introduced in revision 18214. Seems like some of the patch was lost. ___ Reply to this item at: http://gna.org/bugs/?16867 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16809] Barbarian/Pirates should have skill levels
Follow-up Comment #4, bug #16809 (project freeciv): Any new barbarian should be easy and any new pirate should be hard. As player I won't know when barbarians or pirates are created unless they are near my cities, right? It would also be anoying having to set them every time. ___ Reply to this item at: http://gna.org/bugs/?16809 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16808] Civil war should clone the AI level
Follow-up Comment #2, bug #16808 (project freeciv): Maybe a server setting is appropriate then switching between using the same difficulty and the default difficulty. ___ Reply to this item at: http://gna.org/bugs/?16808 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16809] Barbarian/Pirates should have skill levels
Follow-up Comment #6, bug #16809 (project freeciv): Just for fun. Some incentive against building coastal towns. Inland save, coast unsafe. ___ Reply to this item at: http://gna.org/bugs/?16809 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16809] Barbarian/Pirates should have skill levels
Follow-up Comment #2, bug #16809 (project freeciv): But what if I want weak barbarians and mean pirates? ___ Reply to this item at: http://gna.org/bugs/?16809 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16812] Cities report should show revenue from trade routes
Follow-up Comment #1, bug #16812 (project freeciv): To be clear currently it shows just 4 when a city has 4 trade routes. I want that to show 4 (+10) when the 4 trade routes produce 10 extra tade. So only a minor change. ___ Reply to this item at: http://gna.org/bugs/?16812 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16809] Barbarian/Pirates should have skill levels
URL: http://gna.org/bugs/?16809 Summary: Barbarian/Pirates should have skill levels Project: Freeciv Submitted by: mrvn Submitted on: Sunday 10/03/2010 at 00:34 Category: ai Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99+svn18101 Discussion Lock: Any Operating System: None Planned Release: ___ Details: I have both barabrians and pirates in my game and the AI is listed as skill level Easy. There should be a setting what level they get when they are created and that should have an effect on how dangerous they are. ___ Reply to this item at: http://gna.org/bugs/?16809 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16778] Client segfaults when buying while the cities report is open
Additional Item Attachment, bug #16778 (project freeciv): File name: freeciv-T0002-Y-3998-manual.sav.bz2 Size:19 KB ___ Reply to this item at: http://gna.org/bugs/?16778 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16778] Client segfaults when buying while the cities report is open
Follow-up Comment #2, bug #16778 (project freeciv): Seems to be hapening in just any game. Attached a game anyway. Load game, press f4, sort by Currently Building, select Michelangelos Chapel and press buy. ___ Reply to this item at: http://gna.org/bugs/?16778 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16793] Diagonal/sperical multiple selection box (for hex tiles)
URL: http://gna.org/bugs/?16793 Summary: Diagonal/sperical multiple selection box (for hex tiles) Project: Freeciv Submitted by: mrvn Submitted on: Thursday 09/30/2010 at 18:04 Category: client Severity: 1 - Wish Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: None Planned Release: ___ Details: I love using the multiple selection box to manage a line of road builders (late in the game I just build road everywhere). In a game with square tiles you can arange your workers horizontally and vertically and select a whole line by right clicking+draging a selection box over them. In a game with hex tiles though you have vertically NW-SE and NE-SW facing lines, 3 axis instead of 2. It would be nice if one could select along any of the three axis. It would also be nice to have a selection mode with center and radius instead. Instead of a box it should have a hexagon when using hex tiles. ___ Reply to this item at: http://gna.org/bugs/?16793 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16776] Research report isn't updated when changing governor in cities report
Follow-up Comment #1, bug #16776 (project freeciv): Same when changing tax rates. ___ Reply to this item at: http://gna.org/bugs/?16776 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16777] Having multiple units on a field resizes the large map view
Follow-up Comment #2, bug #16777 (project freeciv): I've been seeing this for a long time now. Can't say if it was fixed in some versions I didn't use but it is still present in a recent svn checkout (revision 18101). ___ Reply to this item at: http://gna.org/bugs/?16777 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16777] Having multiple units on a field resizes the large map view
Follow-up Comment #3, bug #16777 (project freeciv): Oh, and yes, that is the same behaviour as #15504. Just a different cause. Probably a different solution too since it should just display less units before adding the '' icon. ___ Reply to this item at: http://gna.org/bugs/?16777 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16776] Research report isn't updated when changing governor in cities report
URL: http://gna.org/bugs/?16776 Summary: Research report isn't updated when changing governor in cities report Project: Freeciv Submitted by: mrvn Submitted on: Tuesday 09/28/2010 at 16:24 Category: client-gtk-2.0 Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99+svn18101 Discussion Lock: Any Operating System: GNU/Linux Planned Release: ___ Details: I wanted to see the effect of different governors on my research. So I opened the cities report (F4) and the research report (F6). Then I selected all cities and changed the governor. The amount of bulbs in the research report does not change. Closing and reopening the research report displays the new value. ___ Reply to this item at: http://gna.org/bugs/?16776 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16777] Having multiple units on a field resizes the large map view
URL: http://gna.org/bugs/?16777 Summary: Having multiple units on a field resizes the large map view Project: Freeciv Submitted by: mrvn Submitted on: Tuesday 09/28/2010 at 16:40 Category: client-gtk-2.0 Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: GNU/Linux Planned Release: ___ Details: When you have multiple units on a square the gtk client shows the ative unit and below that other units on the same square. Now when there are too many units to show all it shows a '' to indicate more units. Unfortunately the '' makes width greater than it normaly is and resizes the large map view. When one changes to a different unit that is alone on a tile (or only one or two other units) the '' disapears and the large map view is grown again. Overall this causes the large map to shift and flicker during gameplay and if the window is just the wrong size it causes a lengthy map redraw every time it grows (when the larger size has more tiles than the small). ___ Reply to this item at: http://gna.org/bugs/?16777 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #16778] Client segfaults when buying while the cities report is open
URL: http://gna.org/bugs/?16778 Summary: Client segfaults when buying while the cities report is open Project: Freeciv Submitted by: mrvn Submitted on: Tuesday 09/28/2010 at 19:51 Category: client-gtk-2.0 Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.2.99+svn18101 Discussion Lock: Any Operating System: GNU/Linux Planned Release: ___ Details: Opening the cities report (F4), sorting by Currently Building and then buying something causes a segfault. The problem seems to be that the client stores NULL entries in the gtk tree in merge_list_stores(). This causes the compare function to be called with a NULL string which segfaults when it tries to split it. As a workaround I patched int cityrepfield_compare(const char *str1, const char *str2) to handle NULL strings. This avoids the segfault but then GTK complains about (freeciv-gtk2:6001): Gtk-CRITICAL **: gtk_list_store_move_after: assertion `!GTK_LIST_STORE_IS_SORTED (store)' failed and the list of cities shown in the report becomes incomplete. Closing and reopening the report fixes that. Patch attached. MfG Goswin ___ File Attachments: --- Date: Tuesday 09/28/2010 at 19:51 Name: freeciv-cities-segfault.patch Size: 525B By: mrvn Prevent segfault when buying things. Does not fix the real problem. http://gna.org/bugs/download.php?file_id=10494 ___ Reply to this item at: http://gna.org/bugs/?16778 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Longturn Freeciv
Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Am Tuesday 29 June 2010 22:59:03 schrieb Andreas Røsdal: * Server architecture: The current single-threaded architecture of the Freeciv server will also be a limitation for large Longturn games. The server should probably be multithreaded to be more scalable. Would this be realistic? No idea here. Is the server actually taking so much time? I think the worst time hog is the AI. Split that out into seperate threads and things should improve greatly. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] AI Documentation Tips
Bernd Jendrissek bernd.jendris...@gmail.com writes: On Mon, May 24, 2010 at 6:15 AM, Goswin von Brederlow goswin-...@web.de wrote: I would also like the AI to run as seperate client (that way it can't use privileged information from the server at all) or at least seperate thread. In large maps the AI takes a majority of the time and nowadays multi-core systems are more and more common. Freeciv would run 4 times as fast here with the AIs spread over 4 cores. Just for the record, I'm making a bit of progress again on my client-side AI experiment. I'm back to sending settlers off to found cities, this time in a less arbitrary and hardcoded way. I'm somewhat surprised by how well it picks out city sites, given the crudeness of its calculation. I'd be delighted to share patches with anyone who's curious, but this is really just a hacking-for-fun effort: expect no academic papers on it. Url? I think I've chatted to you about some stuff that flew right over my head, some stuff about Voronoi diagrams and convex hulls. Frankly, I don't understand any of it! I like your proposal for a wear algorithm (read: I can at least understand how to implement that). Well, a Voronoi diagramm is quite simple. For every city associate every tile with the city nearest to it. Those tiles that have multiple nearest cities form the lines in the Voronoi diagramm. But you probably mean a Delaunay-Triangulation, which you can create from a Voronoi diagramm. A triangulation means you cover the whole map with triangles with the cities in the corners so that there are no gaps or overlaps. The special property of a Delaunay-Triangulation is that the inner angles of the triangles are maximised. So no thin wedges. You could build roads along the edges of such a triangulation. That probably gives you a good way to go from any city A to any city B by road wich is no more than twice (pulled completly out of the air but there is probably such a limit. Just a feeling) as long as the direct way. And that with probably a minimum amount of road building. But I'm wondering if that is really such a good way to build roads. The map resolution is too small and going diagonal costs the same as straight. The shortest path between two cities is not a line. It is a series of left, right, up, down and across steps. And in no way is there just ONE shortest path. Maybe the following is more usefull and still simple to understand. Every tile gets a want road count. Each time a unit wants to go from A to B it finds shortest paths imagining there were roads everywhere. It adds up max(0, limit - want road), call it penalty, for each such tile that doesn't have roads and picks the shortest path with least penalty (or possibly all such paths). Then along the path each want road count is increased (increased by a fraction if there are multiple paths). Building road then becomes a matter of finding tiles with high want road count. As you can see the algorithm has a feedback loop. Tiles that other units would have used are prefered. My hope is that thereby paths would share tiles resulting in less roads being needed. But only when it doesn't delay the journey. The one thing that may (or may not, I'm not familiar with the current AI code) be new is that my struct strategy understands the time value of money. Founding a mediocre city in 2 turns is, in some cases, better than founding a good one in 20. Also, struct strategy knows about purposes and dependencies: it has a way of encoding that to move a (nonexistent) unit, you have to build it first. Intuitively, both these features feel very right. Time will tell if they prove to be a lasting asset. That is highly dependend on the level of your civilization. At the start you want to build many small cities near each other. As time goes on you want to have large cities with enough room inbetween. But if you don't build cities where migration will eat them up anyway then you already do this. But there is another factor. At the start you have to build where a starting city will do great. As time goes on you learn how to transform terrain and have the workforce to do so. At that point city placement becomes more and more about space. Who cares if there is a desert around the town? You just change it. At that point you have to calculate what the city can become andhow much it is going to cost to transform the tiles for it. It is also highly likely that space becomes limited by either ocean or enemies. At what point is it better not to build a city somewhere because it would just limit the surrounding cities? Or do we build a city in a bad spot so it can migrate into another and help it grow beyond 8? Keeping a settler around costs every turn. Sometimes it is better to waste it than to keep it. I'm not sure if the AI really needs long-range strategy. At the moment everything seems okay with a stateless model, where each new turn starts a fresh cycle of planning. In a sense
Re: [Freeciv-Dev] New AI
David Lowe doctorjl...@verizon.net writes: On 23 May, 2010, at 10:15 PM, Goswin von Brederlow wrote: An argument for seperate AI clients would also be that it makes developing the AI more fun. I like this idea. Battle For Wesnoth has an API so that people can try their hand at making a new AI. AIs used to be python scripts, but that was removed for security reasons. One of the goals now is rewriting the API for Lua. Or you can easily play different versions of the AI against each other. This is true genius. Being able to use a different AI is good, but having multiple AIs available in the same game is great. If i'm not mistaken, Civ3 nations have unique 'personalities'. Some are more defensive while others are more aggressive. Also, pitting them against each other would be the best way to test them out. Sent from my MacBookPro I wasted time, and now doth time waste me. William Shakespeare Always play as Zulu or they will slaugter you. :) MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] AI Documentation Tips
Israel israel_...@yahoo.com.br writes: Hello, I am interested in working on FreeCiv AI, and would like to know if there is any documentation or begin reading this doc to help me. It's my first contact with the freeciv code, so maybe I should start from something more general than the AI itself. I have already downloaded the source code (altough not compiled it yet) and read these wiki articles: http://freeciv.wikia.com/wiki/AI_Documentation http://freeciv.wikia.com/wiki/AI_Plans I'm a brazilian civ fanatic (but no relation with civfanatics.com) and actually studying in a Master in Computer Science course. I have some ideas that may improve freecivs' AI global decisions, more specifically what each city will build in a turn. But first I would like to understand better how it is done today, so I can evaluate if my idea makes sense or not. If the results are good I plan to use them on my final work and, of course, make it part of the main freeciv code (if it is good to the project too). Thanks for any help Israel Personally I think it would be a good idea to scrap the AI and start again. The code has so much bitrot in it and the rest of freeciv has changed so much that bringing it up to date might just be more work. I would also like the AI to run as seperate client (that way it can't use privileged information from the server at all) or at least seperate thread. In large maps the AI takes a majority of the time and nowadays multi-core systems are more and more common. Freeciv would run 4 times as fast here with the AIs spread over 4 cores. An argument for seperate AI clients would also be that it makes developing the AI more fun. You can start your game and play it a bit and when you notice something off you kill the AI, fix the issue and restart the AI all without stopping the game. Or when the AI crashes the game continous. Or you can easily play different versions of the AI against each other. As to what is wrong with the AI here are a few things: - road/rail building Road/raid is build by looking at a tile and its neighbours. Building a road/rail when two neighbours have one connects them so it gives that a big bonus. Apart from that it looks where roads would be beneficial to a city. More or less. Here one could do 3 thinks to improve this: 1) plan to connect 2 cities with a road/rail, find a good route for the road/rail and then send worker to the affected tiles. Also consider road/rails to future cities. 2) keep track of where units move. Tiles that units cross a lot need road/rail first. 3) If you are a worker and you are on a tile without road/rail then build one before doing whatever you were doing. - Explore near your land Explorers move in a way that discovers the most hidden tiles. But what use it discovering a hidden tile at the other end of the earth if there is one right next to your city? The score for tiles should be weight by the distance from the border. Also, with fog of war, the time since a tile was visible last should be considered. Have the exporers revisit tiles to see if maybe someone moved in. In effect, after an initial time of discovering hidden tiles, the explorers should become a border patrol. - Pick a job and stick with it Often you see units move towards A and the next turn they all move back again. That easily repeats. That is just a collosal waste of time. If you decide to attack with an unit then stick with it for a while. Rember that you were going to attack something across turns. - Pick a target and coordinate attacks This would massively improve the AI, possible to the point where it becomes anoyingly hard to beat. Currently afaik each unit on its own decides wether to atack something and if so then what. Instead it should find something to attack and then gather a force by finding idle units near it. If a sufficient force is ready then units should be moved so they all end up attacking the target at the same turn with a full move left. Since stacking units too deep is dangerous they could also come in waves. And don't forget to put a good defensive unit in each stack that won't attak (so it keeps up its strength). Building a road/rail towards the target can also help greatly. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [patch #1349] Rework (+fix) auto-settler code
Pepeto pepet...@gmail.com writes: Le dimanche 01 novembre 2009 à 16:16 +0100, Goswin von Brederlow a écrit : pepeto no-reply.invalid-addr...@gna.org writes: Update of patch #1349 (project freeciv): Status: In Progress = Ready For Test ___ Follow-up Comment #2: Updated version of the patch against current svn. Also: * Remove potential crash on tile_index(NULL). * Cleanup in build city stuff: * All settlers use the same pf_parameter. * Units dying are not considered as a failure. If you have too many auto settler some will remain idle. If one dies then it should retask one of the idle ones. Yes, it doesn't change that behaviour. * Units which don't reach the position at the turns are not considered as a failure. If a unit can't reach its destination it should pick some other task. The failure makes the outer loop repeat assigning tasks. Yes, I agree, and it does that. I am talking about the settlers able to reach their destination but the path is too long to be reached in one turn. In your patch for building city, they were considered as failed mission. Should have read the diff between the two patches. The patch currently never releases auto-worker when there is nothing to do. The old code made them not auto-workers so human players could assign them new roles. I would like to use something inbetween. Have them keep their auto flag but bring them into focus in the client. If the client assigns no new activity to them they stay auto-worker and do more work whenever some comes along. Any idea how to do that? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [bug #14621] Cities not refreshed when crossing Empire_Size_Step
pepeto no-reply.invalid-addr...@gna.org writes: Follow-up Comment #3, bug #14621 (project freeciv): Only every 10/11/12/14 cities added. This also needs to happen when a city is lost. Were you planning to post a patch? Not me. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [patch #1234] more than one build slot for units
Matthias Pfafferodt no-reply.invalid-addr...@gna.org writes: Follow-up Comment #4, patch #1234 (project freeciv): The problem I see is that this makes it rather unpredictable how many units will be build. Even if you have 2 or 3 build slots that does not mean you will build 2 or 3 units of each type. That depends on the surplus of shield. This was planned as a possibility to use the production of big cities. Without this, these cities would accumulate shields with no possibility to use them. I restricted the possibility to units without population cost so that you can expect that will happen. Would a checkbox, which is only active if such a unit is build be OK? Using the checkbox you could choose if 1,2,3...x units should be (tried to) build next turn. This would also be the visual indicator for this feature. What is wrong with building the next unit in the list and possibly showing the number of turns required as 0 for all but the first of a set? And in your mode what would happen if I set it to 3 and it only completes 2? Would one remain on the list? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [patch #1234] more than one build slot for units
Matthias Pfafferodt no-reply.invalid-addr...@gna.org writes: Follow-up Comment #5, patch #1234 (project freeciv): What is wrong with building the next unit in the list and possibly showing the number of turns required as 0 for all but the first of a set? I think this would be to powerfull (space ship?). Not a unit and putting 2 space ship components into the list would be the same as clicking the 2 box in your system. I'm not arguing about changing what gets build. Just how to account for/coordinat that with the build list. I'm perfectly fine with building just one Diplomat if you have a Diplomat and an Armor in the build list. But if you have 2 Diplomats and then an Armor and enough shields it should build 2 and remove 2 from the build list. Lets explain it this way: The unit type build can only be changed once a turn but it will produce as many units as are queued up and shields allow. Diplomat, Armor, Diplomat, Armor takes always = 4 turns. Diplomat, Diplomat, Armor, Armor takes = 2. Armor, Coinage builds exactly one armor and coinage the next turn. Freight will build as many Freights as possible each turn. And in your mode what would happen if I set it to 3 and it only completes 2? Would one remain on the list? As one can estimate the amount of shields available (stock + production) only the max. number you could reach with the shields available would be active. This number would further be restricted by the number of build slots. The build list can be long, the city can grow or shrink many times during that time. You can easily gain or loose build slots as well as change shield production and shields stored. Even in the same turn you can disband units and gain more shields that way and suddenly more slots are able to be filled. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [bug #14585] Wish: timeout progress bar
Michael Kaufman kauf...@physics.wisc.edu writes: interesting idea, and easily doable. In fact a nice little project for someone wanting to get their feet wet. Add a timing struct for each AI. Each turn for each AI, calculate the time it takes for various major actions (city management, unit management) and store that time. Use number of cities and units and time previous turn took as a predictor. Add all AI times together for interturn prediction. I imagine that the prediction would be pretty good. M What I ment was not the timeout of a turn (which gets counted down) but the time the server needs between turns. I.e. the time from pressing turn-done till one can move again. In large games I have seen it take from 1 to 20 minutes. Might be overkill to even split this up. Plainly measure the time each turn takes and average the last 10 turns. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Build tests with FullMoon
Marko Lindqvist cazf...@gmail.com writes: For a couple of years now I have been testing freeciv building on several environments. Results of these builds are at http://www.cazfi.net/freeciv/testing/compilation.php These builds are not fully automatic nightly builds but have to be initiated manually. I'm also developing (nightly) build tool called fullmoon ( http://www.cazfi.net/fullmoon/ ). I'm using freeciv building as test case for fullmoon. This should benefit both projects. Now I'm nearing the point where fullmoon can start sending emails to this mailing list about failed nightly builds. Does that include running a testgame with just AIs and an observer for every client? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] Migration and Pyramids: Was: Freeciv patch #1235: [Metaticket] variable city radii
Goswin von Brederlow goswin-...@web.de writes: Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: My biggest city is size 17 so far. I will have to see if the plague allows the city to grow to size 30 at all. Did you try the experimental ruleset? There, some buildings with a 'health' effect are added. This means, that the possibility of an illness is reduced. Furthermore you could check the corresponding wiki page (http://freeciv.wikia.com/wiki/Math_of_Freeciv). Matthias Hospital, genlab and cure for cancer help a lot. Biggest city is now size 54. 270+ production in that city. I didn't merge changing unit upkeep from shield to gold in my ruleset. I have tons of shield but never enough gold. Given the amount of pollution, even with recycling I will have to destroy the Hover Dam and build solar plants instead. MfG Goswin I actualy went and fixed the ruleset so that solarfarm still reduces pollution to 0 even with Hover Dam and Recycling Center. That really helps. Biggest city now 125. I have to say that the Pyramids combined with a Granary are made even more powerfull by migration now. Initially with Pyramids and Granary in a size 2 city one can buy a settler every turn and the city just pops back up to size 2. This allows for a rapid expansion of ones empire giving one a huge headstart. Now, with migration same thing happens. A citizen migrates and the city just pops back to size 2, thus giving a steady stream of migrants to grow a city. So even when expansion stops you still get a huge grows benefit throughout the rest of the game. I also believe the migration is done every 5 turns with a 50% change for the source city (default). That means surrounding a large city by 10 small cities will (on average) grow the large city by 1 each turn. Maybe there should be at most one migration to the same city per 5 turns. I.e. people can only migrate to a city that didn't already recieve migrants in the last 5 turns. Currently migration is more powerfull than rapture. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Freeciv patch #1235: [Metaticket] variable city radii
Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: [...] city_radius_sq = 5 = city_radius = 2 mgr_dist = 2 max distance for migration (tiles) = 2 + 2 =4 city_radius_sq = 10 = city_radius = 3(.1) = 3 mgr_dist = 2 max distance for migration (tiles) = 3 + 2 =5 Yes, that is along the lines of my though. a new patch is available implementing radius_sq and this definition of the migration distance Defining max distance for migration (tiles) = radius1 + radius2 + X would make things more difficult I guess. radius2 won't be known so you would have to assume the maximum, scan for cities and then check if their radius is big enough for migration. But just radius1 + X should be easy to code. [...] Actualy I thing something like that is already needed. The road/rail building AI currently has no notion of connecting cities, only adjacent tiles. With city_min_dist raised that performes really bad. If a city knew the surrounding cities the auto-worker could build a road on a line between two cities. But a lot of AI code needs a fresh look and redesign I feel. This is known but nobody takes this job. I still have one thing open: bug #13599 [RFC; patch] pollution on lakes / ocean (https://gna.org/bugs/?13599). But it is really hard to understand the ai code ... But is it too much work? If I understood migration right then it is done only once every 5 turns. That's right. Migration is checked every fifth turn after a city was founded (default settings). Does the variable city radii work for you? Seems to work great so far. I'm not sure the plague settings are fine tuned yet though. I have a large food box and cities always get plague before hitting size 12 an requiring a Sewer System. The plague effectively obsoletes the size 12 limit except when migration happens a lot and the city shoots up. Not sure if that should be that way. My biggest city is size 17 so far. I will have to see if the plague allows the city to grow to size 30 at all. Did you try the experimental ruleset? There, some buildings with a 'health' effect are added. This means, that the possibility of an illness is reduced. Furthermore you could check the corresponding wiki page (http://freeciv.wikia.com/wiki/Math_of_Freeciv). Matthias Hospital, genlab and cure for cancer help a lot. Biggest city is now size 54. 270+ production in that city. I didn't merge changing unit upkeep from shield to gold in my ruleset. I have tons of shield but never enough gold. Given the amount of pollution, even with recycling I will have to destroy the Hover Dam and build solar plants instead. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Freeciv patch #1235: [Metaticket] variable city radii
Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Am Saturday 10 October 2009 09:54:26 schrieb Goswin von Brederlow: Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Am Thursday 08 October 2009 01:43:23 schrieb Goswin von Brederlow: On a related matter, how does migration work with the patch? The patch says the city radius is considered. Does the other city need to lie within the city raidus? Do the city radius of both cities need to overlap? Can there be a gap of x tiles between the city radii of both cities? I think the migration distance config option should select this, e.g. -99: city must lie within city radius -x: city radii must overlap 0: city radii must touch x: there may be x tiles between city radii I think a default of -1 would make sense. At the moment migration is defined by a migration distance (1 to 7). The patch #1232 adds the possibility to set the migration distance to 0. In this case, the migration distance depends on the city radius (migration distance = city radius + 1; your case 0). I will definetly want to have more choices there. The migration distance setting could also be defined as the value added to the current city radius. This would require a change of the current definition (name). Example: city_radius_sq = 5 = city_radius = 2 mgr_dist = 2 max distance for migration (tiles) = 2 + 2 =4 city_radius_sq = 10 = city_radius = 3(.1) = 3 mgr_dist = 2 max distance for migration (tiles) = 3 + 2 =5 Yes, that is along the lines of my though. Defining max distance for migration (tiles) = radius1 + radius2 + X would make things more difficult I guess. radius2 won't be known so you would have to assume the maximum, scan for cities and then check if their radius is big enough for migration. But just radius1 + X should be easy to code. As for the migration one has to check all tiles within migration radius of the receiving city for giver cities the radius should be as small as possible. If checking the tiles is too expensive then cache the result: When creating a city check if it is within the migration radius of another city and link it into a migration list for that city. When the city radius of a city changes (and therefor migration radius) then scan the area once to rebuild the list. This list has to be updated every time a city is created or destroyed. Furthermore it would spread the code needed for migration over a lot of files. At the moment it is mainly in one file. Actualy I thing something like that is already needed. The road/rail building AI currently has no notion of connecting cities, only adjacent tiles. With city_min_dist raised that performes really bad. If a city knew the surrounding cities the auto-worker could build a road on a line between two cities. But a lot of AI code needs a fresh look and redesign I feel. But is it too much work? If I understood migration right then it is done only once every 5 turns. That's right. Migration is checked every fifth turn after a city was founded (default settings). Does the variable city radii work for you? Seems to work great so far. I'm not sure the plague settings are fine tuned yet though. I have a large food box and cities always get plague before hitting size 12 an requiring a Sewer System. The plague effectively obsoletes the size 12 limit except when migration happens a lot and the city shoots up. Not sure if that should be that way. My biggest city is size 17 so far. I will have to see if the plague allows the city to grow to size 30 at all. Matthias MfG Goswin MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Freeciv patch #1235: [Metaticket] variable city radii
Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Hello Goswin, Thanks for testing the patch. Here only a short answer; more later ... At the moment the freeciv development is really fast. So there seems to be some changes which require an update of the patch. I tested the entire patch set (see patch #1235; https://gna.org/patch/?1235; file #6879: cityradius.diff) before I updated it. I will check for compile errors and update it today in the evening. Have you ever used quilt? It might be usefull to provide the colective diff as quilt series. It is easy enough to generate a quilt series from git preserving the individual patches and comments from the individual bug reports you filed. To use variable city radii you need some changes to the ruleset. You could try my experimental ruleset (patch #1236; https://gna.org/patch/?1236). Within this ruleset the following city radii are defined (see effects.ruleset and also ./doc/README.ruleset_experimental): - Activate variable city radii: radius 1: start radius 2: city size 3 radius 3: city size 10 + Railroad radius 4: city size 30 + Railroad + Mass Transit + Super Highway The vision range of the city is adapted accordingly. To get all the new things listed in this file some additional patches are needed (settings patch series for game settings in the game.ruleset file). What else besides #1235 and #1236? Changing the definition to use radius_sq should be possible and would also be consistend with the use of radius_sq for borders and vision range. I will check this. The 'lot of black' in the city window is a known problem. I don't know how to code for the (gtk) client. Ideally there would be a 5x5 map (default size) and scrolling bars. After opening, this window should be centered on the city. Please use the bug / patch tracker for comments. It helps to follow the discussion(s). Sure. Just didn't want to clutter it up with compile errors. I should probably create a login for it too instead of anonymous. Matthias MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [bug #14452] Wonders binding other buildings effects
Pepeto pepet...@gmail.com writes: Currently, this can be done with the nreqs array. But I agree with your solution which looks better for future. Maybe the value of the effects could be relative or absolute: * a absolute value (then the maximal one would be used). * a relative value (+=bonus, -=bonus, *=bonus or /=bonus). Le mercredi 07 octobre 2009 à 06:14 +0200, Goswin von Brederlow a écrit : anonymous no-reply.invalid-addr...@gna.org writes: Follow-up Comment #1, bug #14452 (project freeciv): Possible guidelines: 1) Never allow building of buildings replaced by active wonder 2) Always allow building of buildings replaced by active wonder 3) Allow building buildings replaced by active wonder that will obsolete at some point We should go by 2. Wonder might be in a city that player knows (s)he will eventually be unable to defend. What about when you can already see you are going to loose some city to the enemy and it contains the wonder? One might want to build replacement buildings before the city falls. Another issue is with wonders and buildings that don't quite do the same: - a power plant is worse than the Hoover Dam. - a solar plant (galactic ruleset) is better than the Hoover Dam. In general a wonder and a building might have overlapping effects but not identical effects. So having local buildings override the wonder or the wonder override the building are both unsatisfactory. It would be nice if the effects of the wonder and building would be merged, e.g. take the maximum of each effect the wonder and bulding have in common. I also like the way the Pyramids/Granary work. Both have a big effect but combined you get a little bit extra. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] Freeciv patch #1235: [Metaticket] variable city radii
Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Am Wednesday 07 October 2009 13:22:13 schrieb Goswin von Brederlow: Matthias Pfafferodt matthias.pfaffer...@mapfa.de writes: Hello Goswin, Thanks for testing the patch. Here only a short answer; more later ... At the moment the freeciv development is really fast. So there seems to be some changes which require an update of the patch. I tested the entire patch set (see patch #1235; https://gna.org/patch/?1235; file #6879: cityradius.diff) before I updated it. I will check for compile errors and update it today in the evening. Have you ever used quilt? It might be usefull to provide the colective diff as quilt series. It is easy enough to generate a quilt series from git preserving the individual patches and comments from the individual bug reports you filed. I'm using git (format-patch) and as user interface qgit2. I have not tested quilt. I only compile tested the patch for gtk2. Thanks for catching this errors on the other clients! To use variable city radii you need some changes to the ruleset. You could try my experimental ruleset (patch #1236; https://gna.org/patch/?1236). Within this ruleset the following city radii are defined (see effects.ruleset and also ./doc/README.ruleset_experimental): - Activate variable city radii: radius 1: start radius 2: city size 3 radius 3: city size 10 + Railroad radius 4: city size 30 + Railroad + Mass Transit + Super Highway The vision range of the city is adapted accordingly. To get all the new things listed in this file some additional patches are needed (settings patch series for game settings in the game.ruleset file). What else besides #1235 and #1236? Nothing! #1235 includes all the patches listed in this metaticket and #1236 adds a ruleset which activates the variable city radii. You have to execute 'rulesetdir experimental'. I tested it and got a city with a radius of 5 ;-) Changing the definition to use radius_sq should be possible and would also be consistend with the use of radius_sq for borders and vision range. I will check this. It would be possible to use radius_sq. But this would require a complete rewrite of most code of the patch due to the changes needed to the city tiles iterator. All references to city_radius (or the function city_map_radius()) have to be replaced by city_map_radius_sq() ... At the moment I don't have much time, but this seems something which should be done. I think 2.3.0 (or 2.2.1?) will have this ;-) Matthias On a related matter, how does migration work with the patch? The patch says the city radius is considered. Does the other city need to lie within the city raidus? Do the city radius of both cities need to overlap? Can there be a gap of x tiles between the city radii of both cities? I think the migration distance config option should select this, e.g. -99: city must lie within city radius -x: city radii must overlap 0: city radii must touch x: there may be x tiles between city radii I think a default of -1 would make sense. Another option would be that only used tiles of one city that lie within the city radius of another may cause migration. MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
Re: [Freeciv-Dev] [bug #14452] Wonders binding other buildings effects
anonymous no-reply.invalid-addr...@gna.org writes: Follow-up Comment #1, bug #14452 (project freeciv): Possible guidelines: 1) Never allow building of buildings replaced by active wonder 2) Always allow building of buildings replaced by active wonder 3) Allow building buildings replaced by active wonder that will obsolete at some point We should go by 2. Wonder might be in a city that player knows (s)he will eventually be unable to defend. What about when you can already see you are going to loose some city to the enemy and it contains the wonder? One might want to build replacement buildings before the city falls. Another issue is with wonders and buildings that don't quite do the same: - a power plant is worse than the Hoover Dam. - a solar plant (galactic ruleset) is better than the Hoover Dam. In general a wonder and a building might have overlapping effects but not identical effects. So having local buildings override the wonder or the wonder override the building are both unsatisfactory. It would be nice if the effects of the wonder and building would be merged, e.g. take the maximum of each effect the wonder and bulding have in common. ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] multiple use of arguments in MACROS and not inlined functions
Hi, after getting fed up with the slow speed of freeciv in a large game I did some profiling. One thing I noticed is that there are MACROS that use one of their arguments twice. With the change to make the player_index a function call they are not called with a function call as argument instead of an interger. Using the argument twice means the function call is executed twice. In the server 3% of the total time where spend in map_is_known: [56] 3.05.43 12.74 152585010 map_is_known [56] 8.890.00 305170020/565345748 player_number [62] 3.850.00 305170020/541877534 player_index [103] Notice how player_number and player_index is called twice for every map_is_known? ---[ server/maphand.c ]-- / Return whether the player knows the tile. Knowing a tile means you've seen it once (as opposed to seeing a tile which means you can see it now). / bool map_is_known(const struct tile *ptile, const struct player *pplayer) { return BV_ISSET(ptile-tile_known, player_index(pplayer)); } ---[ utils/shared.h ]-- #define BV_ISSET(bv, bit) \ (_BV_ASSERT(bv, bit), \ ((bv).vec[_BV_BYTE_INDEX(bit)] _BV_BITMASK(bit)) != 0) ---[ common/player.c ]- /** Return the player index. Currently same as player_number(), paired with player_count() indicates use as an array index. **/ int player_index(const struct player *pplayer) { return player_number(pplayer); } /** Return the player index/number/id. **/ int player_number(const struct player *pplayer) { assert(pplayer); return pplayer - game.players; } - Is there any reason BV_ISSET (and similar ones) isn't written as a static inline function? That one macro alone would save 1.5%. Inlining the player_number() and player_index() functions (and other similary trivial and often called functions) would save a lot of time too. Any reason such frequent one-liners aren't declared static inline in the header file? MfG Goswin ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev