HostedRedmine by Planio

2023-10-10

• 
  Lexxie L Freeciv

  09:20 AM · Created issue

  Task #970885 (New): Disembark to tile occupied by allies

  Apologies if this is not the right place or if divergence in platforms makes this wrong. But just in case, it could be good info on a bug we had at FCW
  Ally attempting to disembark from a ship to a tile occupied by his ally, is told t...

2022-10-22

• 
  Lexxie L Freeciv

  12:44 PM · Commented on issue

  It's easy to reproduce. Just set move_fragments really high. Experiment on a FCW server. Why? Because with FCW, goto-pathing is server-side, so you will see in the client with your own eyes what kind of int-overflow data ALL freeciv serv... Task #951985: Using granularity for move_fragments causes short int overflow in pathfinding functions.

  Recently I'm challenged by issues in the movement system which, if my problems are tl;dr, go to section 2 to read about the actual bug.
  h1. Section 1. Granularity issues in Freeciv movement system.
  If your ruleset has some move_fragm...

2022-10-21

• 
  Lexxie L Freeciv

  01:18 PM · Commented on issue

  in our MP2 rulesets we have canals, so for ocean connectivity I don't know the answer, probably some kind of extra_flag or something? Task #879656: Implement connectivity

  Let the game check if a city is connected by some mean by some route to a tile (particulary, a location of another city).
  In Civilization II, you can get +50% trade route output from road connection to the destination or +100% for rai...

2022-10-14

• 
  Lexxie L Freeciv

  09:25 AM · Commented on issue

  
  > Again, CivII looks only "straight go-to path from dst to src disregarding terrain" roads (and erroneously believes that any city of src owner met while following it is already src; that all I don't want to reproduce). Any foreign (no... Task #879656: Implement connectivity

  Let the game check if a city is connected by some mean by some route to a tile (particulary, a location of another city).
  In Civilization II, you can get +50% trade route output from road connection to the destination or +100% for rai...
• 
  Lexxie L Freeciv

  08:57 AM · Created issue

  Task #951985 (New): Using granularity for move_fragments causes short int overflow in pathfinding functions.

  Recently I'm challenged by issues in the movement system which, if my problems are tl;dr, go to section 2 to read about the actual bug.
  h1. Section 1. Granularity issues in Freeciv movement system.
  If your ruleset has some move_fragm...
• 
  Lexxie L Freeciv

  08:17 AM · Created issue

  Bug #951984 (New): Issues with ZOC over non-native terrain

  This is not a bug in legacy rulesets and works fine in them, because one of the two terrain classes is a NoZOC flag.
  However, in MP2 development, I am experimenting with bringing some limited ZOC mechanics to the water, as consensus r...

2022-01-29

• 
  Lexxie L Freeciv

  02:31 AM · Commented on issue

  Alina L. wrote:
  > A different suggestion thread just reminded me: Obviously, all the local-ranged requirement types that don't currently refer to the tile, but could be given such a meaning (e.g. UnitType) should also be included in Adj... Feature #695469: Add requirement vector "Tile" range

  The "Local" range can currently be pretty confusing, since it contains some information on the current tile (e.g. terrain type and extras), but not all (e.g. different units on the tile). Since, depending on effect type, it's not viable ...

2022-01-13

• 
  Lexxie L Freeciv

  08:38 AM · Commented on issue

  Marko Lindqvist wrote:
  > I think also the recently introduced combat_odds_scaled_veterancy helps a bit with this. I'm inclined to say the situation is good enough for 3.1 now. We may postpone this ticket to 3.2 to re-evaluate the situat... Bug #809590: Combat Rounds exploit may need server settings or flags to balance it.

  For non-victorious combat, I'm quite concerned that it assigns veteran levels. In theory it sounds good: you became more experienced in combat and survived.
  But the ability to know that certain units in certain situations would never...

2021-06-07

• 
  Lexxie L Freeciv

  05:07 AM · Commented on issue

  FYI, I don't think this patch fixed the second bug, where extreme edge cases made negative bribe costs Bug #929034: Units costing less than 10 shields have 0 bribe cost.

  /common/unit.c::unit_bribe_cost()
  blamed lines of code:
  @ /* Consider the build cost. */
  cost *= (unit_build_shield_cost_base(punit) / 10;@
  FIX: Forcing those to (float) fixed it at FCW, but note, FCW version of this function ...

2021-06-01

• 
  Lexxie L Freeciv

  06:34 PM · Commented on issue

  I forgot to mention, fcw.org version of the code has previously fixed other similar maths bugs involving rounding and etc., such as a bug that made negative bribe costs where you got PAID to bribe units. Bug #929034: Units costing less than 10 shields have 0 bribe cost.

  /common/unit.c::unit_bribe_cost()
  blamed lines of code:
  @ /* Consider the build cost. */
  cost *= (unit_build_shield_cost_base(punit) / 10;@
  FIX: Forcing those to (float) fixed it at FCW, but note, FCW version of this function ...