The Deceiver's Gambit - New Mainline Candidate!

[revolting_peasant]

I'm attaching a save from the end of (my) Turn 7 of Galcadar. I note particularly how Delfador's Counterspell aura stops and then resumes while the game is "thinking". This might not be a full 30 seconds, but somewhere in the vicinity.

[FunkLord]

I've encountered a small bug in "Ruins of Saurgrath":

1) Have familiar with you
2) Attack the Matriarch Fliiss with the familiar
3) The dialogue of being spotted appears
4) The attack is being done, but not completely (e.g. only twice if you chose melee)
5) You can once again chose your spells
6a) Select "choose later" or deselect the familiar

Then the error message appears (see screenshot). If I select "wiederholen" (repeat) or "abbrechen" (cancel), the game crashes. If I select "ignorieren" (ignore), my familiar disappears and the Matriarch stays alive with negative health (see screenshot).

6b) Keep the familiar selected

Then there's no error message and the attack finishes normally.

Save game is attached.

Log:

Spoiler:

Code: Select all

Battle for Wesnoth v1.19.10 x86_64
Started on Sat Apr 19 10:57:30 2025

Automatically found a possible data directory at: C:\Program Files (x86)\Steam\steamapps\common\wesnoth

Game data:    C:\Program Files (x86)\Steam\steamapps\common\wesnoth
User data:    C:\Users\USER\Documents\My Games\Wesnoth1.19
Cache:        C:\Users\USER\Documents\My Games\Wesnoth1.19\cache

  error config: Invalid data in credentials file
Setting mode to 1920x1080
Checking lua scripts... ok

Game version:

Spoiler:

The Battle for Wesnoth version 1.19.10 x86_64
Running on Microsoft Windows 10 22H2 (10.0.19045) x86_64
Distribution channel: Steam

Game paths
==========

Data dir: C:\Program Files (x86)\Steam\steamapps\common\wesnoth
User data dir: C:\Users\USER\Documents\My Games\Wesnoth1.19
Saves dir: C:\Users\USER\Documents\My Games\Wesnoth1.19\sync\saves
Add-ons dir: C:\Users\USER\Documents\My Games\Wesnoth1.19\data\add-ons
Cache dir: C:\Users\USER\Documents\My Games\Wesnoth1.19\cache
Logs dir: C:\Users\USER\Documents\My Games\Wesnoth1.19\logs

Libraries
=========

Boost: 1.83
Lua: 5.4.7
OpenSSL/libcrypto: 3.1.0d-dev (runtime 3.1.0d-dev)
libcurl: 8.4.0 (runtime 8.4.0)
Cairo: 1.18.0 (runtime 1.18.0)
Pango: 1.50.14 (runtime 1.50.14)
SDL: 2.28.4 (runtime 2.28.4)
SDL_image: 2.6.3 (runtime 2.6.3)
SDL_mixer: 2.6.3 (runtime 2.6.3)

Features
========

Lua console completion: yes
Win32 notifications back end: yes

Current video settings
======================

SDL video drivers: [windows] offscreen dummy
Window size: 1920x1080
Game canvas size: 1920x1080
Final render target size: 1920x1080
Render refresh rate: 60
Screen refresh rate: 60
Screen dpi: 96.00x96.00
Renderer: direct3d11 (hw)
Maximum texture size: 16384x16384
VSync: on

Current audio settings
======================

SDL audio drivers: [wasapi] directsound winmm
Number of channels: 2
Output rate: 44100 Hz
Sample format: signed 16 bit little-endian
Sample size: 4096 bytes

Installed add-ons
=================

An_Orcish_Incursion: 1.0.4.3
Ashevieres_Shadow: 1.4.3
Gobowars_Plus: 1.1.0a
Under_the_Burning_Suns_Classic: 1.0.0

PS: Very creative campaign, I'm enjoying it a lot so far (even though I haven't really played around with all the possibilities of the different spells yet).

[revolting_peasant]

In "Revelry, revisited", you are tasked with "Defeating as many enemy leaders as possible" - but that's not a concrete objective. There is no turn limit, nor any other condition with which the scenario ends.

Spoiler:

So, it turns out the challenge is to survive for as many turns as possible.

I believe that should be spelled out explicitly. It is particularly important for deciding which veterans to send Eastwards by Turn 6.

[revolting_peasant]

So, I finally finished both parts of TDG.

I have to say - I don't think this is up to par as a mainline campaign - mostly in terms of the story.

• Delfador seems like a frivolous and shallow character throughout both parts. His behavior and demeanor were not convincing. Not for any person, and especially not for a mage. One could perhaps say the same for the heroine of, say, Secrets of the Ancients, but that campaign has a bit of a tongue-in-cheek aspect to it, while this one is supposed to be dramatic and tragic. I didn't feel like Delfador faced significant moral quandries - and certainly not the player.
• There seems to be no resolution - AFAIAC - of the mysterious visions at the beginning of Part 1; and though there is stylistically-reminiscent episode in Part 2 - that's also more mystery and allusion rather than resolution.
• ... and the Seer's portents and advice to Delfador is just ignored, as though it had never been given.
• Asheviere is at the same time a character and a non-character in both parts. She's talked about, but does nothing and says nothing. And it doesn't help that she's a passive damsel in distress at the last scenario of Part I; and that she gets talked about quite a bit.
• Who is the deceiver, and what is their gambit? In part I, one could only suspect Asheviere, but she's barely in that campaign; she can barely said to have done any deceiving, when she doesn't speak. Then in part II, she's again not in the campaign, and it ends up seeming like maybe Eldred is more of the protagonist of the story.
• Too much of characters talking about what other characters, off-screen, have said or done. Like the unrest among the nobles.
• These campaigns are each actually quite short: 7 and 8 scenarios each. Delfador's Memoirs - although not directly comparable - was 22 scenarios for a single campaign.
• Many of the scenarios either have no recruitment or no recall or no keeping of veterans for later recall. With that and other circumstances - I felt I basically had nobody to recall.
• Little opportunity for scenarios where Delfador puts his new spells to good use.
• Orcish Shamans are supposed to be rare, and yet they're used like just so much cannon fodder.
• I don't buy the ultra-strong Prince Eldred unit - neither on the battlefield nor after drinking his potion. Not to mention it feels like a rehash of Omaranth.
• Eldred says he took precautions after the previos defeat - but what defeat was that?
• I was missing closure in the ending of Part I.
• There's a third campaign mentioned - but it's not one of the mainline campaigns. Is it supposed to be? If not - why does Part II have a "to be continued" ending?
• There were essentially no major choices to be made - diverging paths of the campaign(s). At most you could choose between, say, paying a few bandits or fighting them, or whether to skip the Seer's puzzle. There also doesn't seem to be a choice regarding how to "shape" Delfadore's set of veterans, to lean towards one or ther other
• We get shadow mage recruits. While those are fun units, I don't know that that's at all justified story-wise.
• After the first scenario - there is nothing about Alduin, the place of mages in Wesnothi society and in politics, internal squabbles among mages etc. (except for a throwaway leader with a half-paragraph of text in Revelry Revisited); but Delfdaor is a mage... and mages work like a guild, with greatness and knowledge achieved through peer recognition and accredited achievement, not worldly glory.

It's far from a bad pair of campaigns, and there are several very nice scenarios in this pair of campaigns; and I did like the Saurian plot overall. But - the bottom line is that I don't feel it "comes together" well enough. Sorry for being this critical, I'm sure a lot of effort has gone into this one so far, but that's just how I feel.

[Dalas120]

I've encountered a small bug in "Ruins of Saurgrath":

Will fix, thanks!

[Dalas120]

Thanks for finishing the campaign! As much as I'd love to always receive glowing reviews, critical feedback is invaluable for improving.

I'd like to focus on feedback regarding the plot, as that's where I agree that the campaign could use the most improvement - it's gone through several iterations under two subsequent SP leads, and has suffered quite a bit for it. The gameplay/units/spells I'm quite happy with overall, your concerns notwithstanding.

I believe that should be spelled out explicitly. It is particularly important for deciding which veterans to send Eastwards by Turn 6.

I can do that. I think most players gathered this from the dialogue, but more clarity is always better.

I don't buy the ultra-strong Prince Eldred unit - neither on the battlefield nor after drinking his potion. Not to mention it feels like a rehash of Omaranth.

On the battlefield, his high hitpoints are necessary to prevent certain cheesy Delfador 1-shots, but it's fair that there's no need for his damage to be so much higher than the Grand Marshal, since they're both leadership level 4s. I'll lower that a little.

After drinking his potion I do think it's fairly essential for him to have a big power-up. Firstly I think it's a cool fight that gives Delfador a chance to flex his spells, and secondly we need a major justification for Liberty's cursed soldiers.

Eldred says he took precautions after the previos defeat - but what defeat was that?

Spoiler:

This was his failed assassination of Arand, explained at the end of S12 "The Traitor." I'll make Eldred's line more explicit.

I was missing closure in the ending of Part I.

Can you elaborate on this? What closure did you feel was missing? (aside from the stuff explained in Part II, like the saurians)

There's a third campaign mentioned - but it's not one of the mainline campaigns. Is it supposed to be? If not - why does Part II have a "to be continued" ending?

Asheviere is at the same time a character and a non-character in both parts. She's talked about, but does nothing and says nothing. And it doesn't help that she's a passive damsel in distress at the last scenario of Part I; and that she gets talked about quite a bit.

"Asheviere's Dogs" is a campaign by Mechanical, slated for mainline whenever he gets around to making the PR. It focuses on orcs under Asheviere's command, and she gets quite a bit of screentime in it.

We get shadow mage recruits. While those are fun units, I don't know that that's at all justified story-wise.

While rogue mages do make an appearance in Liberty, I don't think there's any implication that they're ONLY in the Grey Forest. Their description simply says they're failed mages operating independently, which IMO fits with Delfador's "You work for cheap. Lousy Eldred wouldn’t give me any of my veterans" line when he gains them as recruits.

After the first scenario - there is nothing about Alduin, the place of mages in Wesnothi society and in politics, internal squabbles among wizards etc. (except for a throwaway leader with a half-paragraph of text in Revely Revisited); but Delfdaor is a mage... and mages work like a guild, with greatness and knowledge achieved through peer recognition and accredited achievement, not worldly glory.

I disagree. Some mages certainly might, but that doesn't mean literally everyone does. I think Delfador is already established as a bit of an unusual mage in S02 Fort Garard, when he leaves his apprenticeship early to eventually become Garard's personal war-mage. IRL, for example, we have plenty of brilliant minds who graduate into academia, and plenty others who graduate into building weapons for military contractors. More importantly, I think there's already way too much going on with the plot to risk adding more intrigue.

[Dalas120]

As for the plot at large... I absolutely agree with your concerns about the campaign "coming together" - thanks for putting in words what I've had a sinking feeling about.

When TDG first started development, there was planned to be an "Angel of Ashes" campaign centering on Asheviere's actions during the same time as TDG. But now that AoA is cancelled and the creator (my former 'boss') has left Wesnoth, TDG really needs to stand alone.

Let's work together to improve it. I can change dialogue, especially around major events. I'm not able to wholesale add/remove scenarios, characters, etc.

Delfador seems like a frivolous and shallow character throughout both parts. His behavior and demeanor were not convincing. Not for any person, and especially not for a mage. One could perhaps say the same for the heroine of, say, Secrets of the Ancients, but that campaign has a bit of a tongue-in-cheek aspect to it, while this one is supposed to be dramatic and tragic. I didn't feel like Delfador faced significant moral quandries - and certainly not the player.

My goal with TDG is to show how Delfador starts as an immature kid, goes through trials, makes mistakes, and eventually develops into the wise mage we all know from HttT (with a very personal motivation to restore the throne)

You're right that ideally I'd like the campaign to feel somewhat tragic, where we see Delfador's unresolved flaws come back to bite him and ruin everything (in "Revelry", when he agrees to Garard's demand). I hadn't meant to give the player moral quandaries - rather, like DiD, I'd like the player to know what the right solution is even though the character does something different.

What do you feel is the best way to achieve this?

There seems to be no resolution - AFAIAC - of the mysterious visions at the beginning of Part 1; and though there is stylistically-reminiscent episode in Part 2 - that's also more mystery and allusion rather than resolution.

... and the Seer's portents and advice to Delfador is just ignored, as though it had never been given.

I am planning to include a little more resolution in HttT (i.e. "the ruby of the wesfolk restored. The nameless son and the ashen daughter."), but you're right that this needs more resolution in TDG itself.

For starters, I can tweak Eldred's final scenario dialogue to better-explain Malal, and Asheviere's role in the ending of Part I.

For the prophecy itself - I'll have Delfador mention it sometime during Part II's saurian arc. But the bulk of its resolution is probably going to have to tie closely into whatever we end up doing with Delfador's Part II personality and his decision in "Revelry" What are your thoughts?

[Dalas120]

Drafted a PR to try and improve some of the dialogue issues: https://github.com/wesnoth/wesnoth/pull/10168

[dank_knight]

Played all & completed all but the last scenario in The Deceiver's Gambit II - and made a report re: slowness between turns for that final scenario here viewtopic.php?p=698404#p698404 - but, as I say there, it may partly be that folks are not ordinarily expected to persist in fighting against literally hundreds of troops. However as I think that final scenario has 10 teams it may be a combination of number of opposing troops and number of opposing teams that is producing the slowdown.

It is possible to survive in the final scenario for a lot longer than the scenario remains playable(if "playable" is defined as it consistently taking a couple of minutes at most for the AI to make its moves). Saves attached to that report in case anyone wants to check if issue is reproducible.

Congratulations to those making this campaign for something with gameplay that's novel for Wesnoth - as can be seen in the saves on my report the Blizzard spell was well used to manage opposing troops in the final scenario.

[Dalas120]

Wow. ...wow! That is a LOT of enemies - kudos to you for having the strategies and the patience to survive that long

With this many units I'm not surprised the game slowed down massively. While there are a large number of sides, all combat units get assigned to side 2 (for performance). Probably the crazy slowness is purely from the huge numbers of units - even very large battles like NR's Eastern Flank have at most a quarter of what you impressively managed to play with here.

Since this is such an extremely unusual case and unlikely to happen in a normal playthrough, I won't worry about changing anything in The Deceiver's Gambit specifically. But perhaps a dev skilled with C++/Lua will consider global AI changes based on your (very detailed, thank you) report in the "SLOW Performance Issues: post here to report" thread.

[dank_knight]

Since this is such an extremely unusual case and unlikely to happen in a normal playthrough, I won't worry about changing anything in The Deceiver's Gambit specifically. But perhaps a dev skilled with C++/Lua will consider global AI changes based on your (very detailed, thank you) report in the "SLOW Performance Issues: post here to report" thread.

I agree that global AI changes is probably the best way - though after talking about an idea for that ("marching mode") I will suggest the issue of slowness may be more commonly relevant for the Deceiver's Gambit II final scenario than you might think. TL;DR: using another campaign's finale as the example the 85 turn mark is easy to reach in a defensive/attritional strategy so that, unless "rush at Eldred with big army straightaway" is only approved strategy for the Deceiver's Gambit II finale, anyone with a slower strategy might experience the slowdown - and "being overcome by the number of minutes it took for the enemy's turn" is a prosaic defeat of the player compared to "being overcome by the number of enemy units attacking".

Anyhow: folks can replay earlier scenarios if there is no other fix for the slowdown. Only thing I'd ask the campaign designers to explicitly do if this becomes an established mainline campaign is to express an official position re: the slowdown. As an example way to do that:- in the official walkthrough put something like "Due to the high recruiting rate of Eldred's forces the enemy turns will take a long time after several dozen turns - slowing down the game very noticeably. Unless/until future AI changes fix this behaviour it is the designer's position that players should start this scenario with sufficient gold for a rapid, sustained assault to reach and kill Eldred". That sort of statement would make clear the slowdown is known, a fix has been considered in principle at least, and the official advice is to replay earlier scenarios to ensure sufficient gold in the finale rather than endure the slowdown.

General AI change idea ("marching mode")

So "marching mode" is what I'd call my idea of a general AI change. Of course as well as being a non-trivial change it might not be compatible with existing AI - so may have no potential beyond an idea.

Idea comes from watching the AI do a load of troop swaps well behind the frontlines that don't have a clear rationale (e.g. moves aren't obviously moving the stronger troops to the front and these moves take troops away from the front, as well as towards it).

Proposal would be for marching mode to be active in area of map away from frontlines based on a "yes" to all the following:

1) do your, AI, troops occupy a substantial portion of the map e.g. over 10%? (this should be computationally easy answer based on total map squares and AI unit list)

2) do the enemy, player, positions describe a small contiguous section on of the map area? (not familiar with calculating area on hex grids but this might an easy calculation).

3) have you, the AI, flagged all the villages outside the player's portion of the map? (i.e. no more income to get).

Under those 3 conditions the area occupied by the player's troops can be expanded a certain amount (perhaps by farthest movement of a player unit plus 2 squares) to form an expanded combat zone and all outside that expanded area is a marching ground where marching mode rather than the default AI algorithms would be used. In the marching ground the object is not to optimise each unit but to move the units as blocks in the direction of the frontline/combat area. Obviously impassable mountains or water represent a challenge.

The attached graphic (below) sketches out how this might play out as troop layouts - which is obviously a lot easier on uniform traversable terrain - and impassable terrain or resulting bottlenecks make it something requiring a lot of thought and testing in practice.

Slowness may not be so niche?

I attach a ~170 turn replay of the final scenario of The Rise of Wesnoth as illustration that a lot of turns and a defensive strategy might occur in the Deceiver's Gambit II finale more often than just with someone like me being stubborn/curious to try a strategy that does not involve replaying prior scenarios for more gold. The replay shows that the Rise of Wesnoth finale can be done (with save cheating but not that much compared to simply rotating units properly) as an attritional strategy - and under such conditions over 100 turns is not unreasonable (if asking "Why so slow?": reach the Lich with 1 loss to 832 kills).

The non-speed based scenario for the Deceiver's Gambit II finale I'd sketch as maybe more broadly applicable is one where the player, like me, uses Blizzard spell to slow some enemy troops while killing other enemy leaders, then holds some villages & accumulates gold for another round of recruiting while recuperating from the initial rush, then with the extra recruits/recalls and rested veterans tries to push through to Eldred.

Unless "fast, immediate assault with a large army & sufficient veterans" is explicitly assume/designed to be the only viable solution to the Deceiver's Gambit II finale I am suggesting that anyone trying a non-speed based scenario may survive to the point the game slows down a lot! Referencing the Rise of Wesnoth finale save: half way through that save is 85 turns - certainly the point at which the slow down in the Deceiver's Gambit II finale is likely to be marked.

Hence my thinking that if this becomes a mainline campaign it would be good to have an official position on the slowdown and envisioned strategy for the finale (such as a brief statement in the official walkthrough). Because the slowdown may not really be that uncommon, so a statement could help folks reaching the finale - and is a lot easier to do some kind of statement than do the AI changes that might render the slowdown less of an issue.