Considering purchasing it, as it's on currently on sale on Steam and I've been in mood for another quality CRPG following Pillars of Eternity: Deadfire, but the reviews I've seen have been mixed.


It sounds like the game may have improved from a rough launch state; I'm interested in the opinions of anyone here who has played it.

Neither of them is a big enough factor. Having adamantium is more powerful than both combined.

(Note what I just said: adamantium is stronger than 3 of your starting picks. This is a huge issue for variance and trying to balance city units.)

While I can't guarantee I can do any particular improvement in this department, it's worth collecting ideas anyway so this thread is for that.

Here is mine : 
Cities screen should somehow show total rebels in city. (maybe instead of total gold? Or as an additional column if we can somehow make space for that?)
Main screen should show total amount of rebels in the empire above gold.

These two will likely remain dreams though - unrest is not stored on cities, it's calculated by a function.

maybe we need a new sticky thread for UI changes like queue, new hotkeys and new options?

Because we can.

We have 8 signups from the last thread, which is still good for 4v4 (If anyone else wants in, speak up now!)

RNG gives us:

Red:
sunrise
JR4
Fenn
Azoth
Commodore

Blue:
WK
Mardoc
Lewwyn
DaveV
T-hawk

Last thing we need is Round 1 spymasters. I can't do the first round, but I can take Round 2.

Now, for something completly different.

Would 1-2 brave souls be interested in trying out to play those scenarios with me? 

Poland 3

Viking 3

Australia 3

Gifts of the Nile 2

I do not think those scenarios are half bad and they are a nice challenge run without direct competition.

I would love to try out the Viking one first.

The building queue feature is done, download CasterQX.zip.
As it's a major new feature, I would be surprised if it had no bugs, so this is for testing purpose only. It's not compatible with older save files (although if every player had absolutely no destroyed buildings in the turn you saved, then it should be safe to use that save file. This should be the case for most saves created right at the beginning of your new turn.).

Two details are not yet implemented - no click sound on the buttons, and the background image hasn't been replaced to have three "slots" for the buttons. Also the background does go black when using the queue function, that's unfortunate but can't be helped.

How to use

There are 3 buttons on your production screen.

"Clear" clears your entire building queue.

"OK" will build whatever you selected as normal, IF THE QUEUE IS EMPTY. So you can still build things one at a time with the same amount of button clicks. Clicking on the already selected building is a shortcut to the "ok" button functionality, whatever its effects are.
If the queue wasn't empty and a building was selected, the OK button starts to process your queue. It will NOT build the building currently selected and it WILL override your existing production order.
If the queue wasn't empty but a unit, trade goods, or housing was selected, those will be built. Those don't use the queue - while building them, your queue remains unchanged and will not progress until you manually change back to building production.

"Queue" will add the building currently selected to the end of the building queue. If a unit was selected, nothing happens. If Housing or Trade Goods were selected, it'll start processing your queue and will NOT build housing or trade goods. To do that, you need to use the "OK button.

You cannot sell buildings required for anything that's in your queue. You have to clear the queue first, sell the building, then re-queue anything else you still want to build.

If your city loses a building to a destruction effect then the following process is executed :
1. If the building was required for another building in your queue, the building will be added to the FRONT of your queue, meaning you will not need to do anything to fix your queue - it'll rebuild the requirement and continue as normal. However, it'll immediately rebuild the requirement even if the item is further up in the queue, for example losing a Granary with a queue of Library,Magic Market, Farmer's Market, will still build the Granary first before moving to the Library, as it is required for the Farmer's Market.
2. If the building wasn't required, but the "REBUILD ALL" optional setting was enabled, the destroyed building will be put back to the front of your queue the same way. This operates on the assumption that anything you already built was more important for you than anything you are building in the future, which is true in general but not always - if you want to maximize your strategy, you probably want to turn this setting off. However, it is an easy way to ensure you never end up losing a building permanently to Meteor Storm or similar effects without noticing and rebuilding it.
3. If a building was currently being produced, then the current production switches to whichever thing is the first on your queue after the additions in the previous steps. This ensures production will always remain consistent with the queue order.
4. If a unit was currently being produced, and its requirement was destroyed, production will be set to housing (even if there were elements in the queue, or added now), and you'll be prompted to change that with a "buildings have been destroyed in this city" message.
5. If the setting "REBUILD PROMPT" is enabled, you will be directed to the city screen the same way anyway, no matter what happened. You'll still be able to see any destroyed buildings that weren't automatically requeued as rubble with the usual "destroyed marketplace" text.

I haven't tested the Grand Vizier, but it should override the completion and destruction messages, but should not override the queues, if one exists for the city. I'm not sure anyone will care to use the Vizier now that queueing is available.

Cities changing owners have their queues cleared, to ensure the AI isn't affected - this is also untested currently.

So, I have a new idea on how to implement build queues for the game.
We had 3 problems preventing it before :

1. No space to store the queue data.
This can be solved by a workaround.
Currently buildings are stored in form of an array with each element containing -1 for not built, 0 for replaced, 1 for built, and 2 for destroyed (rubble) buildings. If I were to say numbers 10+X means that building is in the Xth place in the queue, then we managed to store the queue. This however requires going through every single building related procedure, making sure they check "building status =0 or 1" for the building existing. Those that instead check for "building status not equal to -1 or 2" would need to be changed.
This means units can't be queued, but I think that's fine. You usually don't know in advance how many of those you'll want anyway - depends on how your wars are going. (and most often once a city starts producing units, it keeps doing that for a while)

2. Checking requirements for queuing.
The above workaround actually already solves this, if we make an exception and for the "buildings available for production" procedure only, and nowhere else, we consider buildings in the queue as already built (so replacing the "status =0 or 1" with "status isn't -1 or 2", the opposite of what we normally need to do), then building requirements would automatically work. We need to only do this for buildings, not units.

3. User interface
This one is tricky, but I think we can manage with the existing two buttons on the production screen.

"cancel" would no longer be a thing, instead it would have the function "clear queue". 
"Ok" would have two functions. If a building is selected, it adds the building to the queue.
If nothing is selected, then it leaves the production screen, so to build a building without queueing, you'd need to press it twice.
And finally if a unit, housing, or trade goods are selected, it would produce those immediately, replacing whatever is currently produced, or pushing it into the building queue (if it was a building). The building queue would keep existing, but won't have an effect until production is switched away from these options.
We would however need a way to have "nothing" available for selection, maybe by clicking anywhere else that's not a building, unit or button.

Two other changes will also be necessary - conquering a city will need to clear the queue to ensure AI players never meet the queued buildings since we won't make the AI used the queues obviously but it might confuse them (since it's a building with an effectively invalid status code from the AI's perspective), and, when a building is destroyed, the entire queue needs to be scanned for anything that requires that building, and if one exists, the destroyed building has to be queued immediately to the front of the queue instead of the usual end of it. (alternately, the entire queue could be cleared but that defeats the purpose of queues being an easy way to deal with earthquakes and meteor storms.)
Oh and of course the most important, when the building completes, it'd need to start building the next in queue automatically instead of showing the usual "completed" window.

Do let me know if there are any flaws in my logic and this can't be done. If there aren't any, then space might still be an issue (we'd need to add a lot of new code, to at least 3 different places, but then it's worth a try.

You might have noticed this update has disabled the zombie+black sleep combo. This is a temporal solution until we manage to come up with something better (or decide this is the correct solution afterall).
I believe this discussion is worth its own thread.

I'll try to summarize. So the problem is, black sleep is an effective save or die spell with the drawback of the target not actually dying until you attack it with a unit. Zombies are a weak combat summon that can raise things they kill as undead. But combine the two and you get an effective "save or get raised as undead" spell where the drawback of black sleep turns into an advantage. 

So basically...

1.The cost of producing a unit this way is tremendously lower than producing it normally. A 120 hammer unit would cost you 240 gold to buy, which, if you have alchemy, costs 240 mana. Making an enemy sleep then summoning a zombie costs about 60 if sleep chance is low (30-40%), but only 30-40 if the chance is high. Even considering distance modifiers, the average cost of these units is about half of normal, or less.
Okay that's not as bad as I thought. At least until the target is a summon - you can't buy those for money and to grow your casting skill to summon more, you need to spend non-linearly, while raised creatures have a linear cost in MP instead.

2.Sleep isn't the only culprit, there are dozen ways to make the enemy unit weaker or your zombie stronger to beat them. So disabling only that one might not be the best solution.
However, Sleep is by far the most powerful, and cost effective. While other methods work, they either only work on low value targets that aren't a threat to balance, or the cost is too high to be a major problem. I mean ok, my high prayer zombies can convert even groups of paladins, but high prayer costs 70 plus distance modifier.

3. Werewolves, now, are a particularly bad case of this problem - they have low resistance (which I can't and don't want to raise) and no longer have death immunity, so they are extra easy to convert - but they are also a unit the AI spams early, and one the AI spamming it is actually weak against if turned to undead (giving them Death Immunity), plus it regenerates, ensuring the advantage to stay and snowball as you can reuse all the stolen wolves in every battle.

4. I generally like combos that require using some brainpower...
...but this one isn't rocket science to be honest so not sure there is merit here.

5. The combo is annoying to use. You have to manually move your zombie at one tile/turn while you hit done on all other units, plus you might need to waste turns waiting for the opponent to run out of casting power (which might kill the zombie).

6. It promotes bad play. Using a full stack of 9 is usually the correct tactic to maximize your chances, but to summon a zombie you need a slot so you either need to use 8, or be absolutely sure the enemy will manage to kill something (in fact you might end up letting a unit die on purpose just to have room). It also promotes being lazy - as making undead with zombies is extremely tedious, you might prefer to skip doing it which actually hides the full power of the problem as there is a lot of untapped potential remaining for more advantage (more undead units).

7. It's a human only tactic. Let's face it, the AI can't do this, although they do get some undead to simulate using combat magic to make undead, it's a more balanced quantity, and doesn't work in battles against the human.

And these are some already discussed solutions :

A. More resistance on werewolves. We can't do this one because we actually balanced werewolves around the fact they have low resistance. This is not an option (and would require a too large increase anyway).

B. Nerf Black Sleep save modifier. Again not an option, the spell is good as is, and changing an already balanced and good spell for an uninteded interaction is a bad move. (Note how zombies can be a problem even without sleep, too)

C. Zombies losing "create undead". This is more viable but honestly I think it would be a major loss to the realm, and would leave the realm with too few options on making undead overall. Especially so as Ghouls are still untested and might need a nerf. So I'm against this one.

D. Zombies becoming an overland summon.
That means you are paying overland skill for them, bring them to combat, and they have to survive the battle to turn anything undead. That balances it much better. However, losing the ability to summon them in combat isn't acceptable - it's the main AI invisibility counter, and the only combat summon for Death before very rare, and one of the few spells Death can use against other Death wizards (albeit they are much less effective now as Weakness can disable them). Which brings us to :

E. Zombies becoming an overland or combat summon, where only the overland version has Create Undead.
Basically a similar hack as Demons, Centaurs, etc - if summoned in combat, change the ability on the unit.
This might work, but it's a solution that effectively turns zombies into much inferior Ghouls as an overland spell - which has no use, summoning Ghouls will always be better - while taking away the combat undead creation, ultimately meaning the realm still loses one common source of making undead.
Furthermore, I most definitely don't want to do this until Ghouls have been tested - if they prove too powerful, overland zombies would have the same problem.

F. G. H. ...your ideas here?

Anything I missed?

Some seemingly easy proposals

With selected units:
- W for wait
- R for Patrol (P, A, T are already used... I think that P for plane is a bit useless, you're not changing plane all the time, I'd rather use P for patrol and L for plane actually)
(D is already linked to Done, B to build)

Others:
- Link Esc with closing the X windows - info and spells. The different screens (cities, etc) are already linked.


... I'll add any that come to mind