colossus-tracker Mailing List for Colossus (Page 11)

Bugs item #3133960, was opened at 2010-12-10 03:56
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 7
Private: No
Submitted By: BrentNZ (salient)
Assigned to: Clemens Katzer (cleka)
Summary: Conceding battle leads to hung game

Initial Comment:
I conceded a battle with one creature left to try to speed things up since the AI takes so long to decide which of its troops to send against me.  Nothing seemed to happen.  The opponent then moved and rolled his attacks, but I was not offered strike-backs.  Trying to close the window resulted in it telling me that I had "alrady" conceded, but it was waiting for the opponent to finish his turn.  Conceding again did not do anything.  The game had to be abandoned because there appeared to be nothing else possible.

You should be allowed to concede at any time - not just on your turn.  You may wish to concede before strike-backs so as to not kill a weak creature to make room in the stack for an Angel or new recruit.  I am sure that the rules state that you may concede at any time during the battle.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2011-01-18 17:04

Message:

Might or might be similar case as 

https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160873&group_id=1939

(perhaps the AI got illegal strike, no target messages forever...?)


At least in the other case there is a log which gives some clue what
happened.


----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-10 19:24

Message:
"Any time" was very poor wording on AH's part. Some actions are considered
by most people to be indivisible (e.g. rolling the dice for a strike) and
in the end the ruling was made that you can't interrupt a single strike.
Was you start rolling dice, you need to complete that strike before anyone
gets to concede. Another issue is that checking with your opponent to see
if they want to concede before doing every action in a battle gets very
tiresome. In face to face you can mostly avoid that, but remote play you'd
need to wait to make sure you opponent had a chance to act. In practice
there are only a few places where you need to concede right then as the
nonphasing player (since you don't typically kill your opponent's pieces
when it isn't your action). So the ruling was made that the nonphasing
player only gets to concede at these points not before every action the
phasing player does. And the last issue is that there can races or
anti-races to concede. It isn't particularly fair in remote games to let
the person who gets to see the updated information first get to concede
first. In the pathological cases where this happens, the more natural
person to get priority to concede is the nonphasing player. So the ruling
is that the phasing player can typically only attempt to concede. The
nonphasing player has the option to concede instead in response. The timing
of all of this is described at: http://wolff.to/titan/errata.html in the
complete sequence of play section near the end.

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-10 09:55

Message:
yes, conceding should be possible at "almost" any moment; however, I think,
there are some constraints; namely, if you concede during other player's
turn, HE might opt to concede instead/as well and then his wish has
priority. (Or something like that).

Also I am not sure, whether conceding "to prevent killing a weak creature"
is allowed: it *might* indeed be, that the ongoing strike+strikeback round
must be completed.

Bruno might be able to tell more precisely.

I agree that the current behavior (=hang) is a bug. Last time I checked
it, however it worked - inactive player conceded, and when active player
completed his strikes, battle was over and cleaned up properly. At least it
did not hang.

Needs investigation.

Raising priority since this has really bad impact when it happens.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Bugs item #3160873, was opened at 2011-01-18 16:55
Message generated for change (Tracker Item Submitted) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160873&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 7
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: Concede hangs game

Initial Comment:

One player conceded, and game was stuck because the other player got then "no target in hex B2" messages;
   SEVERE: No target in hex B3 in Server.strike()
   
Probably could not click done either [would Forced done have helped?].

So eventually player's only chance was to withdraw totally.


>From 10812:

INFO: Battle phase advances to Move
...
Jan 14, 2011 11:49:21 PM net.sf.colossus.server.BattleServerSide concede
INFO: Rd11 concedes the battle
Jan 14, 2011 11:49:50 PM net.sf.colossus.server.Server allRequestPingIfNeeded
INFO: Last ping round is 30 secs ago - doing another.
...
Jan 14, 2011 11:49:57 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Ogre moves from C2 to C3
Jan 14, 2011 11:49:59 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Cyclops moves from D3 to C4
Jan 14, 2011 11:50:03 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X3 to D4
Jan 14, 2011 11:50:07 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Titan moves from A1 to D3
Jan 14, 2011 11:50:09 PM net.sf.colossus.server.BattleServerSide$BattlePhaseAdvancer advancePhaseInternal
INFO: Battle phase advances to Fight
Jan 14, 2011 11:50:10 PM net.sf.colossus.server.Server strike
SEVERE: No target in hex B3 in Server.strike()
Jan 14, 2011 11:50:10 PM net.sf.colossus.server.Server strike
SEVERE: No target in hex B3 in Server.strike()

<repeated hundreds of time>

Jan 14, 2011 11:50:50 PM net.sf.colossus.server.Server strike
SEVERE: No target in hex B3 in Server.strike()
Jan 14, 2011 11:50:50 PM net.sf.colossus.server.Server readFromChannel
INFO: EOF on channel for client Forthright setting isGone true
Jan 14, 2011 11:50:50 PM net.sf.colossus.server.GameServerSide handlePlayerWithdrawal
FINE: Player Forthright withdraws from the game.
Jan 14, 2011 11:50:50 PM net.sf.colossus.server.PlayerServerSide die
INFO: Player 'Forthright' is dying, killed by Juliard


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160873&group_id=1939

Bugs item #3160863, was opened at 2011-01-18 16:50
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160863&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
>Priority: 3
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: illegal call to engage() message due to slow server response

Initial Comment:
>From 10809

Jan 14, 2011 10:57:57 PM net.sf.colossus.game.Engagement <init>
INFO: A new engagement: hex Hills hex 18 attacker Bk07 defender Gr10
Jan 14, 2011 10:57:57 PM net.sf.colossus.server.GameServerSide engage
WARNING: illegal call to Game.engage(): engagement ongoing: A new engagement: hex Hills hex 18 attacker Bk07 defender Gr10
Jan 14, 2011 10:58:00 PM net.sf.colossus.server.GameServerSide handleConcession
INFO: Legion Gr10 flees from legion Bk07

Similar like in previous case (with move), the update of display for the engagement takes so long that in meanwhile player clicked it once more.

So perhaps should set a flag that "engagement selection still pending" and give visual feedback and do not send same (or any) request to server before response for pending one has arrived.



----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2011-01-18 16:50

Message:
Lowering priority

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160863&group_id=1939

Bugs item #3160863, was opened at 2011-01-18 16:50
Message generated for change (Tracker Item Submitted) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160863&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: illegal call to engage() message due to slow server response

Initial Comment:
>From 10809

Jan 14, 2011 10:57:57 PM net.sf.colossus.game.Engagement <init>
INFO: A new engagement: hex Hills hex 18 attacker Bk07 defender Gr10
Jan 14, 2011 10:57:57 PM net.sf.colossus.server.GameServerSide engage
WARNING: illegal call to Game.engage(): engagement ongoing: A new engagement: hex Hills hex 18 attacker Bk07 defender Gr10
Jan 14, 2011 10:58:00 PM net.sf.colossus.server.GameServerSide handleConcession
INFO: Legion Gr10 flees from legion Bk07

Similar like in previous case (with move), the update of display for the engagement takes so long that in meanwhile player clicked it once more.

So perhaps should set a flag that "engagement selection still pending" and give visual feedback and do not send same (or any) request to server before response for pending one has arrived.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160863&group_id=1939

Bugs item #3160858, was opened at 2011-01-18 16:44
Message generated for change (Tracker Item Submitted) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160858&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 3
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: Move failed messages due to delayed server response

Initial Comment:


In game 10808:

INFO: Legion Rd09 (Spiral) in Brush hex 102 moves to Desert hex 139 entering on RIGHT
Jan 14, 2011 8:41:59 PM net.sf.colossus.server.Server doMove
SEVERE: Move failed, reason List for normal moves [] + of Rd09 Spiral from Desert hex 139 does not contain 'Desert hex 139'
Jan 14, 2011 8:41:59 PM net.sf.colossus.server.Server doMove
SEVERE: Move failed, reason List for normal moves [] + of Rd09 Spiral from Desert hex 139 does not contain 'Desert hex 139'

When a legion moves, client side gui is updated only after server response has arrived (and says its ok).
In this case, I guess, the player has attempted within that delay twice more to move it.

So probably should have some state/flag on client side that "move is still in progress", and some visual indicator for that,
(hourglass cursor? white box in target hex, ...) and if flag set, instead of just forwarding it to server, give a beep
and some explanative message (e.g. in status bar?).

Arrival of response updates display and clears flag. 

Should there be a timeout, if that response would never come, what then?




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3160858&group_id=1939

Support Requests item #3160220, was opened at 2011-01-17 12:54
Message generated for change (Comment added) made by cnorborg
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Craig Norborg (cnorborg)
Assigned to: Nobody/Anonymous (nobody)
Summary: Registration failed (NULL reply from server)

Initial Comment:
Trying to register on "play-colossus.net", when I enter my info I get the error message "Registration failed (NULL reply from server)".   I do get an email with a confirmation code, but never get a box to put it in?  Tried several times.

----------------------------------------------------------------------

>Comment By: Craig Norborg (cnorborg)
Date: 2011-01-17 14:13

Message:
Tried again, this time it gave me the box to put in the registration code,
which I did.  Then I got the same error again and then prompted me for the
code again...

However, I just tried to log in and it appears to have succeeded as I can
log in.   Still getting the error, but at least it works... 

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2011-01-17 13:57

Message:
Uh. I thought I had fixed that. But sure enough, it happened to me now as
well several times.

This is a "timing" related problem; and as I see now from code I fixed it
for login but not for registration case. After I fixed it (and installed
new jar file on server), registration now works for me.

Can you please retry?

BR,
Clemens


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Support Requests item #3160220, was opened at 2011-01-17 21:54
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Craig Norborg (cnorborg)
Assigned to: Nobody/Anonymous (nobody)
Summary: Registration failed (NULL reply from server)

Initial Comment:
Trying to register on "play-colossus.net", when I enter my info I get the error message "Registration failed (NULL reply from server)".   I do get an email with a confirmation code, but never get a box to put it in?  Tried several times.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2011-01-17 22:57

Message:
Uh. I thought I had fixed that. But sure enough, it happened to me now as
well several times.

This is a "timing" related problem; and as I see now from code I fixed it
for login but not for registration case. After I fixed it (and installed
new jar file on server), registration now works for me.

Can you please retry?

BR,
Clemens


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Support Requests item #3160220, was opened at 2011-01-17 12:54
Message generated for change (Tracker Item Submitted) made by cnorborg
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Craig Norborg (cnorborg)
Assigned to: Nobody/Anonymous (nobody)
Summary: Registration failed (NULL reply from server)

Initial Comment:
Trying to register on "play-colossus.net", when I enter my info I get the error message "Registration failed (NULL reply from server)".   I do get an email with a confirmation code, but never get a box to put it in?  Tried several times.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3160220&group_id=1939

Feature Requests item #2943039, was opened at 2010-01-31 05:59
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=2943039&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Fixes to strengthen AI during battles.  Tundra.

Initial Comment:
First, although it is sometimes (rarely) tactically useful to suffer the 2-points of damage to stand inside a Drift in the Tundra, it is useful so rarely that the AI would be stronger if they were programmed NEVER to stand in Drifts.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2011-01-16 12:05

Message:
John David Galt reported on the mailing list an especially bad case:

Playing the Badlands-JDG variant against three "Random AIs" (I can't tell
which AI the other player was in this case):
An AI player attacked me from the right on the Badlands Tundra map, then
put his titan in F2 (a drift space) and kept it there until it died.  I
didn't even have to move (after entry).



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=2943039&group_id=1939

Bugs item #3147905, was opened at 2010-12-30 03:02
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3147905&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
>Category: General
Group: None
Status: Open
Resolution: None
>Priority: 7
Private: No
Submitted By: Nobody/Anonymous (nobody)
>Assigned to: Clemens Katzer (cleka)
>Summary: Battle phase - lose creatures 

Initial Comment:
I seems when there is a lag, there is a bug when you battle. as defender, I was placing legions on the board, and all of a sudden the ones not on the board where crossed out and dropped, I was not asked to verify leaving legions off the map, they were just gone and my movement roll was completed. This has happened three times in three different matches in the past week.

example game 11316

Gardner

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-30 12:13

Message:
That's really odd. I can't imagine how that can happen. Will need deeper
investigation. 
Good that you provide a game nr.

Raising priority as this is a serious bug.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3147905&group_id=1939

Bugs item #3147905, was opened at 2010-12-30 01:02
Message generated for change (Tracker Item Submitted) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3147905&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Battle phase - lose legions 

Initial Comment:
I seems when there is a lag, there is a bug when you battle. as defender, I was placing legions on the board, and all of a sudden the ones not on the board where crossed out and dropped, I was not asked to verify leaving legions off the map, they were just gone and my movement roll was completed. This has happened three times in three different matches in the past week.

example game 11316

Gardner

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3147905&group_id=1939

Feature Requests item #3146651, was opened at 2010-12-27 23:53
Message generated for change (Tracker Item Submitted) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3146651&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Speed Game Option

Initial Comment:
I have an idea, maybe not fully baked, but how about an option for speed match. Choose to play and you have a max of 5 minutes to split and move. If you do not complete, then autoplay takes over that turn for unmoved legions and muster. Obviously battles stop the clock until completed.

You may, at each minute mark from 3 on, throw up a popup reminding players they are timing out.

- Gardner


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3146651&group_id=1939

Support Requests item #3143392, was opened at 2010-12-24 08:27
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Closed
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Hung process after "new game"

Initial Comment:
I am running the latest variant from http://colossus.sourceforge.net/ using Java version 1.6.0_22-b04 on Windows 7 32 bit. After I get the splash screen and I can select players, I hit the "new game" button. The screen goes away, but the game screen never appears. I never see an error message or a reason for the game to not launch, and when I check the Task Manager I can see the java process still running (and consuming lots of memory), but it never terminates without me doing an end process on it. Is there a debug mode or error log that I can check to see why the game is getting hung?

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 14:12

Message:
Closing this tracker item

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 14:12

Message:
Ah. Interesting effect. From commandline for automated testing, we have an
option "force one screen", if set at least one of the AIs will bring up a
masterboard.

Also, when GetPlayers dialog comes up, perhaps first player should be
pre-initialized as Human.  I guess AI-only playing is not the major use
case :)

Have you discovered Colossus recently? Check out the public server, there
you can play against others via the internet. (In GetPlayers, click on
"Start WebClient").

http://colossus.sourceforge.net/docs/Colossus-Public-Game-Server.html

-Clemens


----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-24 09:18

Message:
Thanks for the super-fast response, Clemens. I figured out the problem. In
the player select window, I had entered my name under the color selection,
but it didn't automatically change my player type to human instead of AI.
Thus, when I hit "new game" it tried to generate a game with all A.I.'s
selected instead of the one human player. It was simply a missed option on
my part. I don't know if a future version of the game can change the player
type to human automatically when someone enters a name, but that is just
something for the developers to think about. Thanks for the help and
letting me enjoy playing Titan/Colossus again!

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 09:09

Message:

Hm, Windows 7...

Can you download the zip from the download area, unpack it. Then open a
cmdline (DOS-box) in that directory and type 
    run  <ENTER>

Now warnings etc. would be written to this DOS box window.

Another possibility is (if you start it by clicking the pink icon, i.e.
using Java Webstart), to enable the java console and/or logging for it.
However I do not know how that looks like on Windows 7.

Colossus should write a log file anyway in a temp directory; under XP,
that would be under C:\Documents and Settings\<username>\Local
Settings\Temp . Not sure how Win7 does that... perhaps you could search
your whole hard disk for a file called Colossus0.log ?


Also, for cross-checking, instead of downloading newest (0.11.0) download
one of the much older ones (like 0.9.0, or even one with just a date in the
OLD_RELEASES folder => just to check whether recent changes how to handle
sockets, ports, threads etc. have broken anything.

-Clemens



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Support Requests item #3143392, was opened at 2010-12-24 08:27
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Hung process after "new game"

Initial Comment:
I am running the latest variant from http://colossus.sourceforge.net/ using Java version 1.6.0_22-b04 on Windows 7 32 bit. After I get the splash screen and I can select players, I hit the "new game" button. The screen goes away, but the game screen never appears. I never see an error message or a reason for the game to not launch, and when I check the Task Manager I can see the java process still running (and consuming lots of memory), but it never terminates without me doing an end process on it. Is there a debug mode or error log that I can check to see why the game is getting hung?

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 14:12

Message:
Ah. Interesting effect. From commandline for automated testing, we have an
option "force one screen", if set at least one of the AIs will bring up a
masterboard.

Also, when GetPlayers dialog comes up, perhaps first player should be
pre-initialized as Human.  I guess AI-only playing is not the major use
case :)

Have you discovered Colossus recently? Check out the public server, there
you can play against others via the internet. (In GetPlayers, click on
"Start WebClient").

http://colossus.sourceforge.net/docs/Colossus-Public-Game-Server.html

-Clemens


----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-24 09:18

Message:
Thanks for the super-fast response, Clemens. I figured out the problem. In
the player select window, I had entered my name under the color selection,
but it didn't automatically change my player type to human instead of AI.
Thus, when I hit "new game" it tried to generate a game with all A.I.'s
selected instead of the one human player. It was simply a missed option on
my part. I don't know if a future version of the game can change the player
type to human automatically when someone enters a name, but that is just
something for the developers to think about. Thanks for the help and
letting me enjoy playing Titan/Colossus again!

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 09:09

Message:

Hm, Windows 7...

Can you download the zip from the download area, unpack it. Then open a
cmdline (DOS-box) in that directory and type 
    run  <ENTER>

Now warnings etc. would be written to this DOS box window.

Another possibility is (if you start it by clicking the pink icon, i.e.
using Java Webstart), to enable the java console and/or logging for it.
However I do not know how that looks like on Windows 7.

Colossus should write a log file anyway in a temp directory; under XP,
that would be under C:\Documents and Settings\<username>\Local
Settings\Temp . Not sure how Win7 does that... perhaps you could search
your whole hard disk for a file called Colossus0.log ?


Also, for cross-checking, instead of downloading newest (0.11.0) download
one of the much older ones (like 0.9.0, or even one with just a date in the
OLD_RELEASES folder => just to check whether recent changes how to handle
sockets, ports, threads etc. have broken anything.

-Clemens



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Support Requests item #3143392, was opened at 2010-12-24 06:27
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Hung process after "new game"

Initial Comment:
I am running the latest variant from http://colossus.sourceforge.net/ using Java version 1.6.0_22-b04 on Windows 7 32 bit. After I get the splash screen and I can select players, I hit the "new game" button. The screen goes away, but the game screen never appears. I never see an error message or a reason for the game to not launch, and when I check the Task Manager I can see the java process still running (and consuming lots of memory), but it never terminates without me doing an end process on it. Is there a debug mode or error log that I can check to see why the game is getting hung?

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-24 07:18

Message:
Thanks for the super-fast response, Clemens. I figured out the problem. In
the player select window, I had entered my name under the color selection,
but it didn't automatically change my player type to human instead of AI.
Thus, when I hit "new game" it tried to generate a game with all A.I.'s
selected instead of the one human player. It was simply a missed option on
my part. I don't know if a future version of the game can change the player
type to human automatically when someone enters a name, but that is just
something for the developers to think about. Thanks for the help and
letting me enjoy playing Titan/Colossus again!

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 07:09

Message:

Hm, Windows 7...

Can you download the zip from the download area, unpack it. Then open a
cmdline (DOS-box) in that directory and type 
    run  <ENTER>

Now warnings etc. would be written to this DOS box window.

Another possibility is (if you start it by clicking the pink icon, i.e.
using Java Webstart), to enable the java console and/or logging for it.
However I do not know how that looks like on Windows 7.

Colossus should write a log file anyway in a temp directory; under XP,
that would be under C:\Documents and Settings\<username>\Local
Settings\Temp . Not sure how Win7 does that... perhaps you could search
your whole hard disk for a file called Colossus0.log ?


Also, for cross-checking, instead of downloading newest (0.11.0) download
one of the much older ones (like 0.9.0, or even one with just a date in the
OLD_RELEASES folder => just to check whether recent changes how to handle
sockets, ports, threads etc. have broken anything.

-Clemens



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Support Requests item #3143392, was opened at 2010-12-24 08:27
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Hung process after "new game"

Initial Comment:
I am running the latest variant from http://colossus.sourceforge.net/ using Java version 1.6.0_22-b04 on Windows 7 32 bit. After I get the splash screen and I can select players, I hit the "new game" button. The screen goes away, but the game screen never appears. I never see an error message or a reason for the game to not launch, and when I check the Task Manager I can see the java process still running (and consuming lots of memory), but it never terminates without me doing an end process on it. Is there a debug mode or error log that I can check to see why the game is getting hung?

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-24 09:09

Message:

Hm, Windows 7...

Can you download the zip from the download area, unpack it. Then open a
cmdline (DOS-box) in that directory and type 
    run  <ENTER>

Now warnings etc. would be written to this DOS box window.

Another possibility is (if you start it by clicking the pink icon, i.e.
using Java Webstart), to enable the java console and/or logging for it.
However I do not know how that looks like on Windows 7.

Colossus should write a log file anyway in a temp directory; under XP,
that would be under C:\Documents and Settings\<username>\Local
Settings\Temp . Not sure how Win7 does that... perhaps you could search
your whole hard disk for a file called Colossus0.log ?


Also, for cross-checking, instead of downloading newest (0.11.0) download
one of the much older ones (like 0.9.0, or even one with just a date in the
OLD_RELEASES folder => just to check whether recent changes how to handle
sockets, ports, threads etc. have broken anything.

-Clemens



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Support Requests item #3143392, was opened at 2010-12-24 06:27
Message generated for change (Tracker Item Submitted) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Hung process after "new game"

Initial Comment:
I am running the latest variant from http://colossus.sourceforge.net/ using Java version 1.6.0_22-b04 on Windows 7 32 bit. After I get the splash screen and I can select players, I hit the "new game" button. The screen goes away, but the game screen never appears. I never see an error message or a reason for the game to not launch, and when I check the Task Manager I can see the java process still running (and consuming lots of memory), but it never terminates without me doing an end process on it. Is there a debug mode or error log that I can check to see why the game is getting hung?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3143392&group_id=1939

Bugs item #3133960, was opened at 2010-12-10 01:56
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 7
Private: No
Submitted By: BrentNZ (salient)
Assigned to: Clemens Katzer (cleka)
Summary: Conceding battle leads to hung game

Initial Comment:
I conceded a battle with one creature left to try to speed things up since the AI takes so long to decide which of its troops to send against me.  Nothing seemed to happen.  The opponent then moved and rolled his attacks, but I was not offered strike-backs.  Trying to close the window resulted in it telling me that I had "alrady" conceded, but it was waiting for the opponent to finish his turn.  Conceding again did not do anything.  The game had to be abandoned because there appeared to be nothing else possible.

You should be allowed to concede at any time - not just on your turn.  You may wish to concede before strike-backs so as to not kill a weak creature to make room in the stack for an Angel or new recruit.  I am sure that the rules state that you may concede at any time during the battle.

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-10 17:24

Message:
"Any time" was very poor wording on AH's part. Some actions are considered
by most people to be indivisible (e.g. rolling the dice for a strike) and
in the end the ruling was made that you can't interrupt a single strike.
Was you start rolling dice, you need to complete that strike before anyone
gets to concede. Another issue is that checking with your opponent to see
if they want to concede before doing every action in a battle gets very
tiresome. In face to face you can mostly avoid that, but remote play you'd
need to wait to make sure you opponent had a chance to act. In practice
there are only a few places where you need to concede right then as the
nonphasing player (since you don't typically kill your opponent's pieces
when it isn't your action). So the ruling was made that the nonphasing
player only gets to concede at these points not before every action the
phasing player does. And the last issue is that there can races or
anti-races to concede. It isn't particularly fair in remote games to let
the person who gets to see the updated information first get to concede
first. In the pathological cases where this happens, the more natural
person to get priority to concede is the nonphasing player. So the ruling
is that the phasing player can typically only attempt to concede. The
nonphasing player has the option to concede instead in response. The timing
of all of this is described at: http://wolff.to/titan/errata.html in the
complete sequence of play section near the end.

----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-12-10 07:55

Message:
yes, conceding should be possible at "almost" any moment; however, I think,
there are some constraints; namely, if you concede during other player's
turn, HE might opt to concede instead/as well and then his wish has
priority. (Or something like that).

Also I am not sure, whether conceding "to prevent killing a weak creature"
is allowed: it *might* indeed be, that the ongoing strike+strikeback round
must be completed.

Bruno might be able to tell more precisely.

I agree that the current behavior (=hang) is a bug. Last time I checked
it, however it worked - inactive player conceded, and when active player
completed his strikes, battle was over and cleaned up properly. At least it
did not hang.

Needs investigation.

Raising priority since this has really bad impact when it happens.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Bugs item #3133960, was opened at 2010-12-10 03:56
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
>Priority: 7
Private: No
Submitted By: BrentNZ (salient)
Assigned to: Clemens Katzer (cleka)
Summary: Conceding battle leads to hung game

Initial Comment:
I conceded a battle with one creature left to try to speed things up since the AI takes so long to decide which of its troops to send against me.  Nothing seemed to happen.  The opponent then moved and rolled his attacks, but I was not offered strike-backs.  Trying to close the window resulted in it telling me that I had "alrady" conceded, but it was waiting for the opponent to finish his turn.  Conceding again did not do anything.  The game had to be abandoned because there appeared to be nothing else possible.

You should be allowed to concede at any time - not just on your turn.  You may wish to concede before strike-backs so as to not kill a weak creature to make room in the stack for an Angel or new recruit.  I am sure that the rules state that you may concede at any time during the battle.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-10 09:55

Message:
yes, conceding should be possible at "almost" any moment; however, I think,
there are some constraints; namely, if you concede during other player's
turn, HE might opt to concede instead/as well and then his wish has
priority. (Or something like that).

Also I am not sure, whether conceding "to prevent killing a weak creature"
is allowed: it *might* indeed be, that the ongoing strike+strikeback round
must be completed.

Bruno might be able to tell more precisely.

I agree that the current behavior (=hang) is a bug. Last time I checked
it, however it worked - inactive player conceded, and when active player
completed his strikes, battle was over and cleaned up properly. At least it
did not hang.

Needs investigation.

Raising priority since this has really bad impact when it happens.



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Bugs item #3133960, was opened at 2010-12-10 14:56
Message generated for change (Tracker Item Submitted) made by salient
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: BrentNZ (salient)
Assigned to: Clemens Katzer (cleka)
Summary: Conceding battle leads to hung game

Initial Comment:
I conceded a battle with one creature left to try to speed things up since the AI takes so long to decide which of its troops to send against me.  Nothing seemed to happen.  The opponent then moved and rolled his attacks, but I was not offered strike-backs.  Trying to close the window resulted in it telling me that I had "alrady" conceded, but it was waiting for the opponent to finish his turn.  Conceding again did not do anything.  The game had to be abandoned because there appeared to be nothing else possible.

You should be allowed to concede at any time - not just on your turn.  You may wish to concede before strike-backs so as to not kill a weak creature to make room in the stack for an Angel or new recruit.  I am sure that the rules state that you may concede at any time during the battle.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3133960&group_id=1939

Feature Requests item #3059850, was opened at 2010-09-05 18:05
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3059850&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Augustin Ku (augustinku)
Assigned to: Clemens Katzer (cleka)
Summary: balanced towers in 5 player games

Initial Comment:
The most balanced start in a 5 player game is to have no players in tower 100.
Right now a player in tower 600 can roll a 4 and take a creature next to tower 100.
So not only do you move last, you can't move out of the tower in all 3 directions.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-08 19:31

Message:

At the moment (r4932), I have undone (reverted) the changes done earlier,
because in some cases the result was not balanced any more at all, and in
fact  broke some unit test case.

Need to come up with a different idea how to avoid usage of either tower
100 or 600 in 5 player game (actually also 4 player should avoid to use
both 100 and 600 together).



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2010-09-09 20:05

Message:

Resolved in r4923:

If only 5 (or 4) of 6 towers are used, we would want to make sure that at
least either first or last (100 or 600) stay unused, because player in
tower 100 would even be further in disadvantage since player in 600 might
block one of his 3 possible pathways (hexes 41 or 42).

For 12, I think it's like the following:
If there is free towers, but more than half of towers are used ( => there
is a case where two neighbored towers will be used).

Generalized, as it happens, it seems this rule applies also to 6  ( 4 > 3,
and 3 = 6/2 ).

For Abyssal9, the tower placement and their numbers are so odd, I don't
even attempt to find a logic there...

So, for 6 or 12 towers, and less than <nr of towers> players, but more
than half of towers used, randomly (50/50) keep either first or last tower
unoccupied.

As usual, keeping it open until it's visible to users in a release/public
build.


----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-09-05 19:18

Message:
This would also apply in a 4 player balanced tower game with players in
both towers 6 and 1.
Maybe there should be a don't use both both towers 6 and 1 when less than
6 players button. (I am not sure how that would translate to other
variants, maybe don't use highest and lowest unless all slots are full?)

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3059850&group_id=1939

Bugs item #3130844, was opened at 2010-12-07 01:53
Message generated for change (Settings changed) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Rules
Group: None
>Status: Closed
>Resolution: Invalid
Priority: 5
Private: No
Submitted By: driehm ()
Assigned to: Clemens Katzer (cleka)
Summary: 2nd try: Cannot Muster Guardian in Tower

Initial Comment:
Colossus 0.11.0
1on1 vs ExperimentalAI
Legion containing 3 Cyclops and 1 previously mustered Guardian entered tower hex 600.
Cannot muster another Guardian.
Added the saved .xml
Did "post" it before but it doesn't show up (neither here in the list nor in my account history). If double post: sorry.
You did an awesome Job there, love Colossus.
Cheers

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2010-12-07 12:25

Message:
Closing as invalid since it was an user misperception.

----------------------------------------------------------------------

Comment By: driehm ()
Date: 2010-12-07 04:10

Message:
Urrgl. I checked before. I think.. I thought.
Now i loaded the file again and now all 6 are in play.. Yuk.
Jeez. Ok, where do i delete this :)

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-07 02:05

Message:
Are you sure all 6 weren't already in play?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Bugs item #3130844, was opened at 2010-12-06 23:53
Message generated for change (Comment added) made by 
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Rules
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: driehm ()
Assigned to: Clemens Katzer (cleka)
Summary: 2nd try: Cannot Muster Guardian in Tower

Initial Comment:
Colossus 0.11.0
1on1 vs ExperimentalAI
Legion containing 3 Cyclops and 1 previously mustered Guardian entered tower hex 600.
Cannot muster another Guardian.
Added the saved .xml
Did "post" it before but it doesn't show up (neither here in the list nor in my account history). If double post: sorry.
You did an awesome Job there, love Colossus.
Cheers

----------------------------------------------------------------------

Comment By: driehm ()
Date: 2010-12-07 02:10

Message:
Urrgl. I checked before. I think.. I thought.
Now i loaded the file again and now all 6 are in play.. Yuk.
Jeez. Ok, where do i delete this :)

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-07 00:05

Message:
Are you sure all 6 weren't already in play?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Bugs item #3130844, was opened at 2010-12-06 23:53
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Rules
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: driehm ()
Assigned to: Clemens Katzer (cleka)
Summary: 2nd try: Cannot Muster Guardian in Tower

Initial Comment:
Colossus 0.11.0
1on1 vs ExperimentalAI
Legion containing 3 Cyclops and 1 previously mustered Guardian entered tower hex 600.
Cannot muster another Guardian.
Added the saved .xml
Did "post" it before but it doesn't show up (neither here in the list nor in my account history). If double post: sorry.
You did an awesome Job there, love Colossus.
Cheers

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-12-07 00:05

Message:
Are you sure all 6 weren't already in play?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Bugs item #3130844, was opened at 2010-12-06 23:53
Message generated for change (Tracker Item Submitted) made by 
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Rules
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: driehm ()
Assigned to: Clemens Katzer (cleka)
Summary: 2nd try: Cannot Muster Guardian in Tower

Initial Comment:
Colossus 0.11.0
1on1 vs ExperimentalAI
Legion containing 3 Cyclops and 1 previously mustered Guardian entered tower hex 600.
Cannot muster another Guardian.
Added the saved .xml
Did "post" it before but it doesn't show up (neither here in the list nor in my account history). If double post: sorry.
You did an awesome Job there, love Colossus.
Cheers

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3130844&group_id=1939