[Freeciv-Dev] [bug #18672] Lock up after 10 mins

Follow-up Comment #2, bug #18672 (project freeciv): Problem persists after resetting client config to default ruleset. GUI is gtk2 with default_theme_name="oxygen-gtk" (from KDE 4.7 System Settings). Video driver is Nvidia (also tried Nouveau). Did not have this issue with 2.2. ___

[Freeciv-Dev] [bug #18633] set timeout - seconds are slightly shorter than in real life

Follow-up Comment #2, bug #18633 (project freeciv): > After /set timeout 82800 (23 hours) the client reports 22:55 Not for me; but it's differently less than 23 hours. What's happening is that the timeout is counted from the absolute time when the turn started (even if there was no timeout set at

[Freeciv-Dev] [bug #18687] Server timeouts >60 hours do not cause client update

URL: Summary: Server timeouts >60 hours do not cause client update Project: Freeciv Submitted by: jtn Submitted on: Sun Sep 18 11:58:55 2011 Category: None Severity: 3 - Normal

[Freeciv-Dev] [bug #18687] Server timeouts >60 hours do not cause client update

Follow-up Comment #1, bug #18687 (project freeciv): When receiving a game_info packet with a large value of seconds_to_phasedone, the client sees it as negative -- exactly -214748.359375. (It seems to be values larger than approximate this value positively that trigger this.) ___

[Freeciv-Dev] [bug #18633] set timeout - seconds are slightly shorter than in real life

Follow-up Comment #3, bug #18633 (project freeciv): Split out the 60h bug into bug #18687, keeping this one for discussion of whether "set timeout" should reset to the full timeout specified. ___ Reply to this item at:

[Freeciv-Dev] [bug #18687] Server timeouts >60 hours do not cause client update

Follow-up Comment #2, bug #18687 (project freeciv): Oops. The network representation of FLOATs appears to be fixed-point in a uint32 divided by 1/1. Taking into account signedness, then values greater than about 60 hours can't be represented on the wire. Fixing this will involve a change in t

[Freeciv-Dev] [bug #18687] Server timeouts >60 hours do not cause client update

Follow-up Comment #3, bug #18687 (project freeciv): To network: dio_put_uint32(&dout, (int)(real_packet->seconds_to_phasedone * 1)); >From network: int tmp; dio_get_uint32(&din, &tmp); real_packet->seconds_to_phasedone = (float)(tmp) / 1.0; where seconds_to_phasedone is a float.

[Freeciv-Dev] [bug #18687] Server timeouts >60 hours do not cause client update

Follow-up Comment #4, bug #18687 (project freeciv): (This doesn't stop larger timeouts being effective, it just means the client displays an inaccurate timer countdown.) ___ Reply to this item at: _

Re: [Freeciv-Dev] T0 timeout

Michal Mazurek writes: > On Sat, Sep 10, 2011 at 11:17:15PM +0300, Marko Lindqvist wrote: >> On 10 September 2011 22:29, Michal Mazurek wrote: >>> I remember writing a diff and sending it to gna, to allow setting the T0 >>> timeout to a different value. Was it ever commited? I cant seem to find >>

[Freeciv-Dev] [bug #18664] unitwaittime - client needs an indicator

Follow-up Comment #1, bug #18664 (project freeciv): I've thought this too. It would require a network protocol change to somehow expose the unit's action_timestamp field to clients. Also, I suspect Freeciv's somewhat sloppy time synchronisation between server and client would tend to make this an

[Freeciv-Dev] [bug #18664] unitwaittime - client needs an indicator

Follow-up Comment #2, bug #18664 (project freeciv): Longturn uses a value of 10 hours. ___ Reply to this item at: ___ Message sent via/by Gna! http://gna.org/

[Freeciv-Dev] [bug #18664] unitwaittime - client needs an indicator

Follow-up Comment #3, bug #18664 (project freeciv): Ah, well, that wouldn't be a problem :) What I'm worried about is the rush of events at turn change that means there might be a significant delay between the server starting a timer and sending a packet and the client processing that packet and

[Freeciv-Dev] [bug #18688] Window manager decides client is "not responding" during long turn change

URL: Summary: Window manager decides client is "not responding" during long turn change Project: Freeciv Submitted by: jtn Submitted on: Sun Sep 18 13:02:27 2011 Category: client-gtk-2.0

[Freeciv-Dev] [bug #18688] Window manager decides client is "not responding" during long turn change

Follow-up Comment #1, bug #18688 (project freeciv): There was a "tactic" suggestion in one of the longturn games during turnchange, that one of the allies would move a unit around on railroad in a loop using goto in a way that the enemy would see. The other ally that would not see the unit would

[Freeciv-Dev] [patch #2962] Antiguan and Barbudan nation

URL: Summary: Antiguan and Barbudan nation Project: Freeciv Submitted by: mixcoatl Submitted on: Sun 18 Sep 2011 02:55:45 PM GMT Category: rulesets Priority: 5 - Normal

[Freeciv-Dev] [bug #18689] timestamps in game chat in 2.3 are broken

URL: Summary: timestamps in game chat in 2.3 are broken Project: Freeciv Submitted by: akfaew Submitted on: Sun Sep 18 16:36:39 2011 Category: None Severity: 3 - Normal

[Freeciv-Dev] [patch #2946] Indo-European nation

Follow-up Comment #2, patch #2946 (project freeciv): The solar symbol is my remake of Ishtar's star symbol, but also inspiration was flag of Indo-European language family by Vitaly Vetash: http://images4.wikia.nocookie.net/__cb20110401105910/althistory/images/thumb/6/6e/Indo-European_flag_by_Vita

[Freeciv-Dev] [bug #18690] fix possible division by 0 for a map without goodies

URL: Summary: fix possible division by 0 for a map without goodies Project: Freeciv Submitted by: syntron Submitted on: So 18 Sep 2011 21:34:05 CEST Category: general Severity:

[Freeciv-Dev] [bug #18407] Version 2.3.0-RC1 - Delay too much to change turn

Update of bug #18407 (project freeciv): Status: Ready For Test => Fixed Assigned to:None => syntron ___ Reply to this item at:

[Freeciv-Dev] [bug #18407] Version 2.3.0-RC1 - Delay too much to change turn

Update of bug #18407 (project freeciv): Open/Closed:Open => Closed ___ Reply to this item at: ___ Nachricht g

[Freeciv-Dev] [bug #18689] timestamps in game chat in 2.3 are broken

Follow-up Comment #1, bug #18689 (project freeciv): BUt in 2.2 you only did got this information if you were logged in at that time - so it is an improvement. That's needed is a timestamt which is included in the message. Something like: 'xyz logged in at 2011/09/18 12:00', or? _

[Freeciv-Dev] [bug #18547] fix number of bulbs for global goal

Update of bug #18547 (project freeciv): Status:None => Fixed Assigned to:None => syntron Open/Closed:Open => Closed

[Freeciv-Dev] [bug #18674] Map generator on trunk has gone nuts

Update of bug #18674 (project freeciv): Status: Ready For Test => Fixed Open/Closed:Open => Closed ___ Reply to this item at:

[Freeciv-Dev] [bug #18689] timestamps in game chat in 2.3 are broken

Follow-up Comment #2, bug #18689 (project freeciv): I am really sorry, it's my bad. The client has some configuration option apparently, to show a server timestamp next to the client one: [21:56:46] (T51 - 12:36:01) (server prompt): 'save lastsave.sav.gz' I didn't set it in S2.3, thus my confusio

[Freeciv-Dev] [bug #18499] Science progress shows incorrect number of turns

Update of bug #18499 (project freeciv): Status:None => Ready For Test ___ Follow-up Comment #12: Could you test the attached patch? (file #14109)

[Freeciv-Dev] [bug #18689] timestamps in game chat in 2.3 are broken

Update of bug #18689 (project freeciv): Status:None => Invalid Assigned to:None => syntron Open/Closed:Open => Closed _

Re: [Freeciv-Dev] T0 timeout

Am Sonntag, 18. September 2011, 13:32:43 schrieb Jacob Nevins: > Michal Mazurek writes: > > On Sat, Sep 10, 2011 at 11:17:15PM +0300, Marko Lindqvist wrote: > >> On 10 September 2011 22:29, Michal Mazurek wrote: > >>> I remember writing a diff and sending it to gna, to allow setting the > >>> T0 t

[Freeciv-Dev] [bug #18675] Unit can move into city where it should not be able to exist

Follow-up Comment #1, bug #18675 (project freeciv): Do you think there are more bugs like this? I do hope to get the transporter patches (patch #2270) into 2.4 and they will force anything connected to such errors. ___ Reply to this item a

[Freeciv-Dev] [bug #18673] Savegame setting loading can reject valid setting combinations in contrived circumstances

Follow-up Comment #1, bug #18673 (project freeciv): An idea/possible solution: define different levels to set/validate the values: - basic checks (limits which should _never_ broken) - real checks (with settings for all variables) - real+ checks (with settings for all variables; if a value is l

[Freeciv-Dev] [patch #2944] Volga German nation

Follow-up Comment #3, patch #2944 (project freeciv): I get these errors for the Volga German nation: 2: Loading rulesets. 1: "People's Commissar" male ruler title for nation "Volga German" (nb 480) is not a format. It should match "%s" 2: Backtrace: [...] 1: "female:People's Commissar" female r

[Freeciv-Dev] [bug #18612] more LUA accessors for terrain

Follow-up Comment #13, bug #18612 (project freeciv): Some comments to the code: - please use a small function for terrain_transform_result(). There you could check if the given terrain is valid (!= NULL). Use fc_assert_ret() for the check. - ou do not need a macro for tile_change_terrain() and c

[Freeciv-Dev] [bug #18499] Science progress shows incorrect number of turns

Follow-up Comment #13, bug #18499 (project freeciv): Comment by Michal Mazurek via mail: > Any chance to either not round it, or provide info on how many > bulbs/turn needed for researching the tech in Y - 1 turns? I > find myself quite often calculating this manually. Any ideas / hints how it s

Re: [Freeciv-Dev] T0 timeout

On Sun, Sep 18, 2011 at 10:19:58PM +0200, Matthias Pfafferodt wrote: > Am Sonntag, 18. September 2011, 13:32:43 schrieb Jacob Nevins: > > Michal Mazurek writes: > > > On Sat, Sep 10, 2011 at 11:17:15PM +0300, Marko Lindqvist wrote: > > >> On 10 September 2011 22:29, Michal Mazurek wrote: > > >>> I

[Freeciv-Dev] [bug #18689] timestamps in game chat in 2.3 are broken

Follow-up Comment #4, bug #18689 (project freeciv): I have found no need for knowing the date, only the turn and hour, for checking if my opponent logged in yet. Can a timestamp be added to the servers stdout log? (vdo_log()) ___ Reply to

Re: [Freeciv-Dev] T0 timeout

Am Montag, 19. September 2011, 06:42:38 schrieb Michal Mazurek: > On Sun, Sep 18, 2011 at 10:19:58PM +0200, Matthias Pfafferodt wrote: > > Am Sonntag, 18. September 2011, 13:32:43 schrieb Jacob Nevins: > > > Michal Mazurek writes: > > > > On Sat, Sep 10, 2011 at 11:17:15PM +0300, Marko Lindqvist wr