Return from the Abyss [SP campaign for 1.12/1.14]

[sergey]

I found one more bug in that scenario.
I closed all the tunnels, but the level was not ended.
The corresponding event (Dwarvish Miner moveto, in the current version it is line 1337) should have first_time_only=no. Also I am concerned if that will work properly, even after adding the first_time_only=no. What if the event which checks the number of opened tunnels will be triggered before the event which closes the last tunnel? I didn't find information regarding the order of events triggering of the same type. You may change the type of the "victory conditions" event from the moveto to some custom type, and trigger it at the end of each "close tunnel" event.

[skeptical_troll]

You are right, it probably never occurred to me that I manage to seal the tunnels, although I thought I checked it
I changed the [if] statement to a [filter condition] and now it works properly.

I didn't find information regarding the order of events triggering of the same type.

I empirically found that events written first in the .cfg file are triggered first. However, using custom events is probably the safest thing to do anyway, and makes the code more readable. Since this was a more serious bug I already uploaded the corrected version.

Thanks again!

[sergey]

Hello, skeptical_troll. I have finished the first chapter of the campaign.
It is a masterpiece. Intentions and behavior of the characters are very realistic from my point of view, I liked that very much. Also I liked the idea of the artificial illumination.

Regarding the Khalifate units limitation. I played on the normal level and have 61 available recruits at the end of the first chapter and 18 Khalifate units on the recall list. I lost many of the recruits in "The Old Castle" scenario, since I used them as a cannon fodder. Perhaps, dwarfs are better choice for this purpose, but I decided that some of the recruits may survive, get experience and become useful in the next scenarios.

Minor idea:
In the "Path To Light" scenario, when the friendly orcs are spotted, there is no filter on which of the orcs is speaking. When I played, the wolf rider was participating in the dialog, but it was under the shroud. Maybe remove the shroud before the dialog?

Thanks for creating this outstanding campaign!

[skeptical_troll]

Hi Sergey, thanks! I'm glad that you liked the campaign so far.

Minor idea: In the "Path To Light" scenario, when the friendly orcs are spotted, there is no filter on which of the orcs is speaking. When I played, the wolf rider was participating in the dialog, but it was under the shroud. Maybe remove the shroud before the dialog?

This is actually a good idea, I inverted the order of the actions.

[taptap]

The orcs strike back: Taking out the small leaders does not really seem viable with no excess force to spare (I had less gold than inky). Going for red first, however, works. I was very worried about the steady reinforcements red got and the green caravan guard piling up over time, but thankfully not intervening. Positive luck with the main party (somewhat mitigated by bad luck for Gnarg) and more importantly several barely surviving units. A less orc heavy recruit might have helped at times, but I wanted to give at least some of my orcish veterans a final battle (I had 8 orcish L3 on recall list *sniff*). The AI not blindly rushing out, probably reduced my losses, but also made storming the castle much harder.

[skeptical_troll]

Hi Taptap, thanks for the replay. It seems reasonably balanced, although it's probably on the hard end. I will however reduce the caravans escort and the gold of the ford guard to make secondary leaders less protected.

All those loyal units piling up and not intervening look a bit silly. It could be a good idea to release them from their guarding duties once the assigned caravan arrives (or is destroyed), and balancing this with reduced gold for each caravan/boat.

[taptap]

Iron and Mud: Cool scenario. I admittedly watched Inky's replay before playing in earnest. I played less nimble (stayed at day against thinned out merfolk) and with higher losses. The uprising contributed little to the win, but wrecked my income, which was about 0 instead of +15 or so the turn it could have been (I guess it is fine). Since I stayed to liberate everyone (and finished with 3rd kill) this did cost me somewhat. I went straight to the main prison, misplayed there at the cost of two units (accidentally lured the chief gaoler into a bottleneck), defended mainly in that area while prisoner rescue was ongoing and then spread out from there. (The amount of allies I levelled all the way sometimes depresses me, hope they do good work for their respective homelands.)

There are merfolk brawlers in the campaign unit list, if this was for this scenario, then it did not work as I did not see a single one. Could be a very cost effective unit for the merfolk.

Spoiler:

The dwarven trackers are a lovely touch in part 2. Hope it never comes to blood between them and us.

[skeptical_troll]

The uprising contributed little to the win, but wrecked my income

I considered making them loyal, but then thought it was too much. It's not a big problem, as the next scenarios do not require lot of gold, however I reduced the probability of spawning them.

There are merfolk brawlers in the campaign unit list, if this was for this scenario, then it did not work as I did not see a single one. Could be a very cost effective unit for the merfolk.

yes, I considered them for this scenario but then haven't fully convinced myself about it. I could allow one of the merfolk leaders to recruit them, it will add variety but may also change the balance.

The amount of allies I levelled all the way sometimes depresses me, hope they do good work for their respective homelands.

Yeah, I understand that the way this campaign works may generate this feeling anyway,

Spoiler:

an idea I have in the back of my mind is to make a 'bonus level' on the same map where saurians defend themselves from the Alliance after Khafir is gone. It would be triggered if some conditions are met (e.g. king and queen are still alive) and would have no victory/defeat condition, just resist as much as possible up to some time limit. This is however low priority now as I'm working on the third part.

[taptap]

Misty Valley: Misty mystery scenario, I expected more of an assault so blew up all my gold reserves. Learned that naffats have a quite good movement type. Once you get into the rhythm there isn't that much fighting anymore. The mechanism is good, although it was quite the calm scenario in the end, which I did not expect.

(I did try the other route as well, but failed since so many fire spirits spawned. These completely random spawns can mess up difficulty balancing badly due to normal variance.)

The Plague: This looks very unstable balance-wise. It can be a walkthrough with few units and close to no fighting or the village can fall early w/ little infections and intra-fanatic fighting. Then it can become hard very fast, since there are spawning bandits from 3 edges. I played through in a rather unsatisfactory way (loss of falcon and only Khafir really across the bridge, the rest stuck in north) and will replay in time.

[skeptical_troll]

The mechanism is good, although it was quite the calm scenario in the end, which I did not expect.

Damn! Now I see that there was a bug, I wonder why I didn't realize looking at Inky's replay. There should be fire sprites/dusk faeries being spawned occasionally, but they weren't due to a wrong variable stored. Now it has been fixed, I hope it doesn't make it too hard.

These completely random spawns can mess up difficulty balancing badly due to normal variance

It might be that I am allowing too much randomness in spawning in the campaign. I could maybe put some thresholds to restrain variability within some limits. In the 'occult ridge' scenario, the number of fire spirits increases drastically with time, because there is a fixed spawning probability for each lava tiles, so more tiles, more spirits.

The Plague: This looks very unstable balance-wise.

I agree, it came out a bit differently from expectations so I'll probably make significant changes on it.
EDIT: I spent some time working on 'The Plague' to make it more playable. I changed and enlarged the map so that an easy walk-through should not be an option, and modified the number and type of enemies spawned such that the villagers should be able to resist until Khafir arrives. It is still supposed to be an easy one, but hopefully less subject to cheap tactics. Suggestions for improvements are still very welcome.
I also increased enemy's gold in the following one, to make it more likely to work as intended.

[taptap]

The Plague: Just added a replay. (Thanks for the new loyal unit.)

The Forest of the Crossing: This is a large roleplay-y map (I remembered Hunter of the East large maps a bit) with a lot of hidden surprises. Only one was activated (and then not really used), but since I replayed Plague so I still have the falcon I had another means of distraction. What should work with the time limit and settings is building up a significant force in the bottleneck and then retreat into the town to get more units to bear at suitable time of day. I distracted a few with falcon and broke through directly instead. When I continue I would be desperate to bring one or two of the mercs with me (the potential leadership elf, the healer), but it won't happen and it is likely better for it. (I did saveload a turn when I forgot to cover the blue leader once and he was wiped by lancer.) Some losses especially against wolves and on the plain after crossing.

(It would be easy to exploit the cave agility related bugs here, it is removing health from some, but healing the L4 or only the L4 loyal, right? I was quite tired while playing, but it did heal I think. You know best whether caves will feature, but if not removing the "quest item" after part 1 might be best.)

[skeptical_troll]

Had a brief look at the replays. As I wrote above, 'the plague' has changed substantially, it should not be possible to just walk past now, although it won't be hard anyway.

The forest: I should somehow highlight better the Goblins' tunnels, basically they will be able to move very fast to the other side on the river, providing another effective means of distraction for the enemy troops. You probably did not need them anyway as you already assembled a force strong enough to break through the siege.

It would be easy to exploit the cave agility related bugs here, it is removing health from some, but healing the L4 or only the L4 loyal, right? I was quite tired while playing, but it did heal I think.

I re-coded the 'cave agility' ring a while ago to circumvent the bug, so it shouldn't be a problem any more. Unfortunately, that happened after you played scenario 2 so it's still affecting your game.

[sergey]

taptap, you can modify your savegame to fix the "cave agility" ability.
Event is a block of code that starts with [event] and ends with [/event]
Replace ALL events that have

Code: Select all

id="ability_cave_event_khafir"

with the next code:

Spoiler:

Code: Select all

[event]
	first_time_only=no
	id="ability_cave_event_khafir"
	name="turn refresh"
	[modify_unit]
		[filter]
			[filter_wml]
				[variables]
					cave_speeded=yes
				[/variables]
			[/filter_wml]
		[/filter]
		[movement_costs]
			cave="$this_unit.variables.old_cave_speed"
		[/movement_costs]
		[variables]
			cave_speeded=no
		[/variables]
	[/modify_unit]
	[modify_unit]
		[filter]
			[filter_adjacent]
				ability="cave_agility"
				id="khafir"
				is_enemy=no
			[/filter_adjacent]
			[not]
				[filter_wml]
					[variables]
						cave_speeded=yes
					[/variables]
				[/filter_wml]
			[/not]
		[/filter]
		[variables]
			cave_speeded=yes
			old_cave_speed="$this_unit.movement_costs.cave"
		[/variables]
		[movement_costs]
			cave=1
		[/movement_costs]
	[/modify_unit]
[/event]

If you not sure how to do that, you can send me your savegame and I will modify it for you. If you will decide to modify the file, make a backup copy just in case. This works in case if Khafir has the ability (and your replay shows that Khafir has it). Fix will be present in all subsequent scenarios after the modified savegame.

[Inky]

Hey, I have been following along in this thread even if I haven't posted in a while. (Also, big thanks to sergey for fixing my save file for me!!)

I replayed The Plague, and I still found it quite easy to just run by the village before the battle there really begins, and also outrun the eastern bandits (I only recalled one keep of fast units like jundi). Maybe a possible change would be to have the eastern bandits spawn to the northeast of the village instead? That way you'd have to fight your way through them and also would reach the village later. (Maybe this would be too hard )

Another possible change is to raise the chance of adjacent units being infected and/or have the infected bandits stay on the same side as the other bandits, so you might have to fight your way through some infected bandits; right now there is almost no chance of your own units being infected because the villagers won't move next to your units for no reason, and any infected bandits are quickly killed by the other bandits.

[skeptical_troll]

Alright, you convinced me that a more drastic change was needed. I elaborated on the intra-fanatic fight mechanic (although it's unilateral) and gave the infected ones a brand new leader guarding the bridge. Now infection should be more of a danger, but hopefully not at the level of becoming annoying. I also took the chance to characterize a bit more the fanatics.

By tuning the new leader's gold I can potentially make this one very hard, but I think I prefer to keep it as a simple (but non trivial) one.