[Freeciv-Dev] [bug #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Update of bug #16305 (project freeciv):

  Status:  Ready For Test => Fixed  
 Open/Closed:Open => Closed 


___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Follow-up Comment #9, bug #16305 (project freeciv):

Rats. Updated patch for trunk attached.

(file #9947)
___

Additional Item Attachment:

File name: trunk-diplomat-city-sabotage-bis.diff Size:7 KB


___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[pepeto]

Follow-up Comment #8, bug #16305 (project freeciv):

Diplomat help text is not changed in the new multiplayer ruleset.


___

Reply to this item at:

  

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev

[Freeciv-Dev] [bug #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Update of bug #16305 (project freeciv):

 Planned Release: => 2.2.3,2.3.0

___

Follow-up Comment #7:

I'll commit these patches to S2_2 and trunk soon if there are no objections.

___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Update of bug #16305 (project freeciv):

  Status:None => Ready For Test 
 Assigned to:None => jtn

___

Follow-up Comment #6:

Well, here are some patches which increase Diplomats' chances of sabotaging
city improvements/production. Do we want to apply them?

> Let's come up for a few testcase scenarios to make sure the proposed
changes are somewhat balanced. :)

What sort of thing did you have in mind? I can't think of any way of testing
the balance of this other than play-testing.

(Personally I haven't used Diplomats much, so I don't have a feeling for how
much this might upset things.)

(file #9696, file #9697, file #9698)
___

Additional Item Attachment:

File name: S2_1-diplomat-city-sabotage.diff Size:0 KB
File name: S2_2-diplomat-city-sabotage.diff Size:3 KB
File name: trunk-diplomat-city-sabotage.diff Size:6 KB


___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Follow-up Comment #5, bug #16305 (project freeciv):

The Civ2 manual at civfanatics.com
 doesn't
indicate that Spies are more likely to succeed at anything than Diplomats,
except in the case of diplomat defence (which we already have). (Some of the
details in that manual sound different to our implementation, but there's not
enough detail in the manual to be absolutely sure.)

I don't think the Apolyton Great Library thread
 shows evidence
for such a difference either.

(I don't have Civ2 to try it myself.)

___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Marko Lindqvist]

Follow-up Comment #4, bug #16305 (project freeciv):

> I cannot remember a specific decision regarding diplomat vs spy
> success rates.

I'm quite certain that this is meant to follow civ2 rules (as civ2
compatibility was freeciv goal for a long time). Does someone know if
diplomat and spy have same success probability in civ2?
Maybe this should be defined in ruleset - especially if we want default to be
different from civ2 and thus from our civ2 ruleset.


___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Daniel Markstedt]

Follow-up Comment #3, bug #16305 (project freeciv):

I cannot remember a specific decision regarding diplomat vs spy success
rates. Tried to look around a bit but couldn't find any reference.

It is a tricky business though, balancing diplomats/spies. There was a time a
few years ago when they were so powerful that they essentially replaced
military units as the main tool for warfare. Let's come up for a few testcase
scenarios to make sure the proposed changes are somewhat balanced. :)

___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[Jacob Nevins]

Follow-up Comment #2, bug #16305 (project freeciv):

Well spotted.

I suspect the difference in success probability between Diplomats and Spies
is unintentional. I think it's been there at least since this area was last
substantially modified in r9000, RT#11058
, in 2004. There's no
indication in that ticket that it was an intentional change.

Also, the help for the server option "diplchance" says "Base chance for
diplomats and spies to succeed." And all the other actions using diplchance
don't (obviously) favour the Spy.

On the other hand, the current help for the Spy does say "She can perform all
the functions of the Diplomat, but with a higher rate of success." This text
came in in r13997, RT#39541
. No indication in RT
of where that notion came from. (Daniel, if you're listening, do you happen
to remember?)

If this is fixed, I don't think the statement in the help will really be true
any more -- the only way spies have a higher rate of success is that they're
relatively stronger when in contention with other diplomatic units, I think.

Fixing this will make diplomats more effective at sabotaging improvements,
which could affect the balance of the early game. However, we don't seem to
have shrunk from rectifying similar unintentional but long-standing gameplay
issues in the past -- see for instance bug #14181 and bug #16080.

So, on balance, I think we should restore equality for diplomats and spies.

I don't think we should fix S2_2 clients to send B_LAST, as that'll give new
clients on old servers an unfair advantage. We could do so on trunk (but in
general sending foo_LAST over the network is a bit unpleasant -- see for
example bug #15743).

___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[anonymous]

Follow-up Comment #1, bug #16305 (project freeciv):

Since I see nowhere in the rules/help that diplomats have half the success
chance of spies, I figure this is a bug. Then again, it's kind of an obvious
bug when you notice that sabotaging by diplomats fail very often, so maybe
it's the security flaw after all.

If it's a bug, then either the gui's must be fixed to send B_LAST as the last
parameter, or better: put line 848 in server/diplomats.c before line 835.

If it's a security flaw, then I would still suggest putting line 848 first,
and adding an extra check, halving the chance for diplomats. E.g.:

  int success_prob = (improvement >= B_LAST && unit_has_type_flag(pdiplomat,
F_SPY) ? game.info.diplchance : game.info.diplchance / 2);


___

Reply to this item at:

  

___
  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 #16305] Incorrect sabotage handling for diplomats

[anonymous]

URL:
  

 Summary: Incorrect sabotage handling for diplomats
 Project: Freeciv
Submitted by: None
Submitted on: Monday 07/26/2010 at 19:31 CEST
Category: None
Severity: 3 - Normal
Priority: 5 - Normal
  Status: None
 Assigned to: None
Originator Email: 
 Open/Closed: Open
 Release: 2.2.1
 Discussion Lock: Any
Operating System: None
 Planned Release: 

___

Details:

The base sabotage chance for a diplomat is half that of the spy, yet could be
the same by changing the client.

This is either a bug (when the base sabotage chance should be the same for
both) or a security flaw (when it should not be possible to upgrade the
sabotage chance on the client). Since I don't know the intended behavior, I
cannot asses this.

In server/diplomats.c:835, function diplomat_sabotage(), the base success
probability is chosen depending on whether a specific target is chosen, or
random:


  int success_prob = (improvement >= B_LAST ? game.info.diplchance
  : game.info.diplchance / 2);


A few lines further down (line 848) the improvement is set to B_LAST if the
unit is not a spy:


  if (!unit_has_type_flag(pdiplomat, F_SPY))
improvement = B_LAST;


So on line 835, the 'improvement' comes straight from the client unmodified,
and based on that, the base success chance is chosen.

This value is set in the gui, for example in
client/gui-gtk-2.0/diplomat_dialog.c:120, function
diplomat_sabotage_callback()


request_diplomat_action(DIPLOMAT_SABOTAGE, diplomat_id,
diplomat_target_id, -1);


The last parameter (-1) is the improvement which will be the parameter
'improvement' in diplomat_sabotage() in server/diplomats.c:835. Since -1 <
B_LAST, this means that a diplomat will sabotage with half the base success
chance. This can be changed on the client by simply sending B_LAST as the
last parameter on request_diplomat_action()




___

Reply to this item at:

  

___
  Message sent via/by Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev