Canceling movement/support bug?

Post a reply

Confirmation code
Enter the code exactly as it appears. All letters are case insensitive.
Smilies
:points: :-D :eyeroll: :neutral: :nmr: :razz: :raging: :-) ;) :( :sick: :o :? 8-) :x :shock: :lol: :cry: :evil: :?: :smirk: :!:
View more smilies
BBCode is ON
[img] is ON
[flash] is OFF
[url] is OFF
Smilies are ON
Topic review
   

If you wish to attach one or more files enter the details below.

Expand view Topic review: Canceling movement/support bug?

Re: Canceling movement/support bug?

by Jamiet99uk » Fri Apr 24, 2020 5:47 pm

Oh, and finally, while we're on the subject of "things you might not realise", here's one more:

You can capture Supply Centres by retreating.

This detail often catches new players out.

Here's a quick example:

It is Fall 1903.
France has Armies at Paris and Gascony and a Fleet at Marseilles.
Germany has an Army at Burgundy.

France orders:
A Par - Bur
A Gas support move to Bur from Par
F Mar - Spain(sc)

Germany orders:
A Bur - Par

Result? Germany's attempted move to Paris fails, of course. France's attack succeeds and Paris moves to Burgundy with Gascony's support. Meanwhile the French fleet moves to the south coast of Spain.

But now Germany must retreat.

So he retreats to Marseilles, because it is now vacant. And because it is the Fall retreat phase, Germany captures Marseilles and gains a Supply Center!

You would be surprised how many relatively inexperienced players would be caught out by that.

Re: Canceling movement/support bug?

by Jamiet99uk » Fri Apr 24, 2020 5:35 pm

ohmann wrote: ↑
Fri Apr 24, 2020 4:04 pm
 Thanks everyone for the explanations. In particular, the part I did not expect was:
English Channel tapping Picardy does not stop Picardy from tapping Brest.
Regarding backstabbr, apologies for the inaccuracy, but it appears that when I ensure all units actually belong to the correct country, then backstabbr's sandbox behaves in the same way as webDiplomacy did (nobody moves). I think I rushed it last time and didn't want to go through the extra turns until I could set up all the units correctly (one country owns all four that were England in the original case, and one country owns all three of the others), so I'm guessing that must have changed the behavior.

I think this thread is resolved, both my original confusion and the supposed (but not actual) contradiction with backstabbr. Thanks!
Happy we could help.

To clarify the above rule even further, just because it might help in future:

A unit can *always* cut the support of an enemy unit simply by moving against it, *irrespective* of what else happens.

So, for another example:

Turkey has Fleets in Ankara and Constantinople.

Russia has Fleets in Black Sea and Sevastopol, and Armies in Armenia and Smyrna.

Austria has an Army in Rumania.

The orders are as follows:

Turkey:
F Ankara - Black Sea
F Constantinople - Hold

Russia
F Sevastopol - Rumania
F Black Sea support move to Rumania from Sevastopol
A Smyrna - Ankara
A Armenia support move to Ankara from Smyrna

Austria:
A Rumania - Hold

When these moves are resolved, Russia will capture Ankara with the army from Smyrna, but will NOT capture Rumania. This is because even though the Turkish fleet in Ankara is successfully attacked, this does not prevent it from "tapping" the Russian fleet in the Black Sea and thus breaking Black Sea's support for the move to Rumania from Sevastopol.

Re: Canceling movement/support bug?

by ohmann » Fri Apr 24, 2020 4:04 pm

Thanks everyone for the explanations. In particular, the part I did not expect was:
English Channel tapping Picardy does not stop Picardy from tapping Brest.
Regarding backstabbr, apologies for the inaccuracy, but it appears that when I ensure all units actually belong to the correct country, then backstabbr's sandbox behaves in the same way as webDiplomacy did (nobody moves). I think I rushed it last time and didn't want to go through the extra turns until I could set up all the units correctly (one country owns all four that were England in the original case, and one country owns all three of the others), so I'm guessing that must have changed the behavior.

I think this thread is resolved, both my original confusion and the supposed (but not actual) contradiction with backstabbr. Thanks!

Re: Canceling movement/support bug?

by jmo1121109 » Wed Apr 22, 2020 11:58 pm

backstabber does not follow the DATC rules entirely, meaning they coded in their logic in a custom way that might not adjudicate as you'd expect. They also have issues with allowing incorrect moves around coasts. I would love to see the sandbox though to see if this is one of the cases where their logic fails or not.

Re: Canceling movement/support bug?

by Squigs44 » Wed Apr 22, 2020 11:05 pm

Now I'm interested in backstabbr's adjudication of this. Is your sandbox saved and shareable? I've heard backstabbr's adjudicator has bugs, but their dev team insists it passes the DATC, and usually when someone says there is a bug it is human error, not the software.

Re: Canceling movement/support bug?

by Doug7878 » Wed Apr 22, 2020 10:39 pm

http://www.diplomacy-archive.com/resour ... 2AH3rd.pdf

Rulebook, page 4, example 10 "In effect, any attack upon a unit supporting another cuts that support."

Re: Canceling movement/support bug?

by Jamiet99uk » Wed Apr 22, 2020 10:33 pm

ohmann wrote: ↑
Wed Apr 22, 2020 9:47 pm
 We ran into this situation, which didn't resolve as I expected: https://i.imgur.com/82Swpxz.png

England (pink)'s orders:
  1. The army at Gascony move to Paris. (fail)
  2. The fleet at English Channel move to Picardy. (fail)
  3. The army at Marseilles move to Burgundy. (fail)
  4. The army at Brest support move to Paris from Gascony. (fail)
Germany (brown)'s orders:
  1. The army at Picardy move to Brest. (fail)
  2. The army at Burgundy move to Gascony. (fail)
  3. The army at Paris support move to Brest from Picardy. (fail)
I expected English Channel's move to cancel Picardy's move such that Picardy would *not* cancel Brest's support and, finally, such that Gascony would successfully move to Paris.

I did find a site that let me set up a sandbox for this (https://www.backstabbr.com/), and they resolved it as Gascony moving to Paris, Burgundy moving to Gascony, and Marseilles moving to Burgundy. Most importantly, that resolution has Gascony's move succeeding, from which I think the rest follows.

Is this a bug, or an I misunderstanding something (and therefore backstabbr's resolution is also wrong)?

Thanks!
Gascony's move to Paris fails because Picardy taps Brest, thus cutting Brest's support for Gascony > Paris.

English Channel tapping Picardy does not stop Picardy from tapping Brest.

Re: Canceling movement/support bug?

by cdngooner » Wed Apr 22, 2020 10:25 pm

Brest didn't move, it supported. Therefore, the only army that moved to Brest was from Picardy (Paris's support failed because Gascony moved there). Therefore Picardy's moved failed and it bounced back home. Therefore English Channel's move to Picardy failed (no support).

Canceling movement/support bug?

by ohmann » Wed Apr 22, 2020 9:47 pm

We ran into this situation, which didn't resolve as I expected: https://i.imgur.com/82Swpxz.png

England (pink)'s orders:
  1. The army at Gascony move to Paris. (fail)
  2. The fleet at English Channel move to Picardy. (fail)
  3. The army at Marseilles move to Burgundy. (fail)
  4. The army at Brest support move to Paris from Gascony. (fail)
Germany (brown)'s orders:
  1. The army at Picardy move to Brest. (fail)
  2. The army at Burgundy move to Gascony. (fail)
  3. The army at Paris support move to Brest from Picardy. (fail)
I expected English Channel's move to cancel Picardy's move such that Picardy would *not* cancel Brest's support and, finally, such that Gascony would successfully move to Paris.

I did find a site that let me set up a sandbox for this (https://www.backstabbr.com/), and they resolved it as Gascony moving to Paris, Burgundy moving to Gascony, and Marseilles moving to Burgundy. Most importantly, that resolution has Gascony's move succeeding, from which I think the rest follows.

Is this a bug, or an I misunderstanding something (and therefore backstabbr's resolution is also wrong)?

Thanks!

Top