Are you, in fact, a pregnant lady who lives in the apartment next door to Superdeath's parents? - Commodore

Create an account  

 
Retort balance, hopefully last time

I've been wondering about the balance of retorts as it feels it's always the same ones getting used while others are not. So i'll try to summarize my thoughts here.

Warlord and Tactician - Being the only two pure military retorts in the game, and military being the most trivial thing to go for (basically the only thing you have in all other 4X games, which do not have summoning and heroes, plus even here you always have cities that make normal units, and can rarely play a game without relying on at least some)  these definitely see a lot of use, but mutual exclusivity will improve it - at the very least you have to play one or the other instead of both, which leaves more picks for other retorts or books. (see the discussion in the Life thread for details on that.)

Famous and Artificer - These are pure hero retorts that you will only pick for hero strategies, but hero strategies are not as commonly played as normal units and even with hero strategies, you don't always pick either of these retorts - but you do pick them often enough. So these see a reasonable amount of use for their role.

Alchemy - This is likely overused. The military effect is not very powerful anymore - Alchemist Guilds only cost 100 so it's merely a +200 gold per military production center - but the convenience it offers is very tempting, as you can't know in advance whether you will have a balanced flow of resources, or need to convert regularly due to having bad luck on either gold or magic power. I don't think we can do anything about that, human psychology, fear of the unknown. I do think the cases where alchemy is picked because a large amount of conversion is part of the strategy are probably a reasonable share of the games, and not the problem, so while this retort does see above average use, it's not a balance problem and not something we can do anything about (nor do we need to - while it's a tempting retort, it's also the first one to drop if there is no room for it in a strategy)

Myrran - This is really a game mode, not a retort in my opinion, so I'll ignore it for this thread.

Guardian - Definitely underused, partly because it's 2 picks, partly because defense is usually not a desirable tactic for human players, and maybe also because it relies on having strong units in garrison which is a strain on your economy. We still haven't decided on how to improve this retort, it's been on the "open problems" list for like 2 years.
(idea : what if it also reduced maintenance costs on patrolling units? Yes that's useless for the AI (do they even patrol their garrisons? Or just not order them to move...), but the AI picks it for the defense and it's pretty good for that, it needs to be improved for humans, really. We also had that idea of somehow improving spellcasting in city defense battles, that would also help AI. Considering how expensive the retort is in picks, we might even want both effects added...)

Sage Master, Conjurer - These are used very often but reasonably so, no change needed.

Channeler - This feels underused, and I blame the range penalty changes for that, see the other thread posted today.

Archmage, Spellweaver - Used often but not always and Mutual Exclusivity helps, like the Warlord/Tactician case. I think the current Spellweaver is actually good as is, it was nerfed enough to be no longer a "must always pick" retort even for spell strategies.

Omniscient- Most versatile retort in the game that outright allows otherwise unplayable combinations to win, while it doesn't seem to be picked excessively, definitely sees enough use.

Cult Leader - A bit overlooked but definitely a strong contender for games that rely on magic power. I think it's fine as is. Only retort that requires you to pick a race that can build certain types of buildings btw.

Charismatic - Only diplomacy retort. I'd say it's underused because "kill everything that moves as soon as possible" seems to be the trend nowadays. Not really a problem with the retort itself, if we succeed on our goal to make longer games more viable and shorter games harder to pull off, this will also start seeing more use.

Specialist - This kinda feels too good as it's basically allowing you to have a 11th pick in you main realm and it also encourages single realm play (which is already strong enough and doesn't seem to need that.) - I think it's both overused and a somewhat poor game mechanic. Would be sad not to have it, though, being very good at one thing is a strong element of flavor in a game. I'm tempted to say toning down the numbers is a good idea but it's not that simple because it can't be worse than picking +1 book - it has to be viable even if you haven't maxed out the books in the realm, and books do give cost and research discounts.

Astrologer - Node Mastery was popular for a reason, not only is this a pretty good effect, but treasure hunting is also popular. While not a mainstream retort, I don't think we need to worry about it not seeing enough use either.

Runemaster - I'm definitely worried about this the most.
Faster Spell of Mastery is a big deal, but for 2 picks you could just get Sage Master and something that enhances magic power and probably be at the same position or better. This effect might need to be much more powerful to matter, like +100% instead of 50.
Double dispel power sounds great but Sorcery gets it by default, and others...don't get dispelling wave. Being resistant to city curses and globals is a big deal but city curses were recently nerfed and this doesn't even cover them all, only those 5 that stay on the city and can be dispelled, and not every enemy will have strong globals, in fact it's quite likely only the last enemy reaches them and they might end up not getting any. So this is a nice extra but no one would actually pick the retort for it.
Dispel resistance is the big selling point, but it's not always necessary, specialist also provides it, for less picks, and Spelweaver is also an overlapping effect (you get to recast them 33% faster when dispelled - far less effective but also works if there were no dispels!)

So it's overall a package of 3 not very desirable effects for a very high cost of 2 picks, and all 3 are late game effects, too. (ok dispel resistance can be useful even early, but Dispel Magic isn't that much of a threat. Dispelling Wave comes quite a bit later, in the midgame only.)

(December 4th, 2018, 03:38)Seravy Wrote: Alchemy - This is likely overused. The military effect is not very powerful anymore - Alchemist Guilds only cost 100 so it's merely a +200 gold per military production center - but the convenience it offers is very tempting, as you can't know in advance whether you will have a balanced flow of resources, or need to convert regularly due to having bad luck on either gold or magic power. I don't think we can do anything about that, human psychology, fear of the unknown. I do think the cases where alchemy is picked because a large amount of conversion is part of the strategy are probably a reasonable share of the games, and not  the problem, so while this retort does see above average use, it's not a balance problem and not something we can do anything about (nor do we need to - while it's a tempting retort, it's also the first one to drop if there is no room for it in a strategy)
 
Let's count. Most of the games i am limited by the casting skill. If i go werewolves i need casting skill. If i buff horses again i am limited by the casting skill. I can produce a horse a turn or even more if i need to, but can i fully buff a horse a turn? No, so after i pick warlord/tactitian i need somthing to improve casting skill or reduce spell costs. For spellweaver math is obvious, for archmage as i remeber that is sqrt(1.5). What is for alchemy? I want to put all my mana power into casting skill. With alchemy i can do that for sure(except for a few starting turns may be). How much of my mana power will i have to invest into mana if i dont have alchemy? Hard to answer that question... i feel like beeing able to fully invest into training casting skill costs more than even archmage or spellweaver. +1 to hit is just a nice bonus.

(December 4th, 2018, 04:18)Sapher Wrote: Let's count. Most of the games i am limited by the casting skill. If i go werewolves i need casting skill. If i buff horses again i am limited by the casting skill. I can produce a horse a turn or even more if i need to, but can i fully buff a horse a turn? No, so after i pick warlord/tactitian i need somthing to improve casting skill or reduce spell costs. For spellweaver math is obvious, for archmage as i remeber that is sqrt(1.5). What is for alchemy? I want to put all my mana power into casting skill. With alchemy i can do that for sure(except for a few starting turns may be). How much of my mana power will i have to invest into mana if i dont have alchemy? Hard to answer that question... i feel like beeing able to fully invest into training casting skill costs more than even archmage or spellweaver. +1 to hit is just a nice bonus.

Looked up the calculations we did in the research rebalancing thread last year, basically, putting all your MP into SP doubles-triples your SP production. So it's a +100% or higher compared to the 50% of Archmage - but you are actually paying this cost in gold entirely which will be missing from your economy (fewer cities built, etc) while Archmage adds the 50% for free. So it's 0 vs 50 on actual gain, but 100-200 vs 50 on the maximal output of SP. This gain is SQRT on the resulting actual casting skill though, which makes it much less than it appears at first sight. The more wars you do (and more mana you spend), the greater the skill increase is compared to other retorts.

As early win strategies use lots of mana in battles, Alchemy is very effective for them to get high skill, in fact it's probably the only way to do so, other than Amplifying Towers which are not cost-effective unless you skill is already higher than what you typically reach in these short 6 year games.

There is probably a hidden, indirect resource gain here - power is harder to produce than gold so converting in that direction at a 1/1 rate actually increases the value of your gold. I'd risk the statement that 1/2 is the fair conversion rate and 1/1 is a +100% gain in value. However you aren't using all your gold on that, about half goes on maintenance, some on buying units and buildings, so it's probably equivalent to a +25% gold retort. That's good but not that outstanding, Omniscient does more with 4 Life books.

No matter how I look at it I can't find a major balance flaw in the retort - except for one thing, it allows you to break the pacing of the game. You don't need the slow-to-build power buildings but can still cast spells at a rate you'd normally only be able to much later as you raise skill from gold indirectly - the amount isn't really the problem, the timing is. The gold you save on Alchemist Guilds adds to this as it's also a gain that happens early, and the retort also enables converting found mana crystals to gold to buy early settlers or military in the unlikely case you end up with too much of them. All of this might make it the most "strong early" retort in the game.

I like your guardian idea of free upkeep for garrison. It is something.

Channeler - increase upkeep bonus from 50% to 2/3

Rune Master needs improvement while specialist could be limited to picking 10 books of the same realm.


Charismatic has the same problem that majestic aura has: in-game events trump their effect completely. If you're stronger than an AI you don't need them, the AIs will accept all the deals after maybe a present or two, if you're weaker or have too many cities they won't accept your offers anyway. Or maybe I'm mistaken, but this is my feeling.

Was thinking a little about the Guardian maintenance and it's a bad idea - units used to expand also end their turns in cities most of the time - cities you conquered from enemies, that is.
Maybe if the retort instead provided an economic bonus to cities of your own race (like inquisitor but without forced razing and a different, more balanced bonus that isn't early game oriented), then it would work better.
Something like "max pop +5 in all cities of your starting race"?
Did we have any particular reason for not adding a max population/population growth retort in previous brainstorming threads?

...these are what I found :

Quote:One more argument against "terraformer" - it encroaches Nature realm's specialty  -raising max population through Change Terrain and Gaia's Blessing.
Quote:By the way, "Terraformer" isn't flawless either - it clearly benefits more on the "Dry" Climate setting and is worse on "Wet", albeit it's useful on both of them at least.all.

These are true but not strong enough to discard the idea for them - we discarded it for another reason : Omniscient including the ability (+X max pop for Nature books) already.

But limiting it to your home race is functionally different and would open up interesting combinations. (stacking it with Omniscient and Nature to build very good cities on horrible spots comes to mind first. Stacking with Cult Leader to pacify the higher population comes next. Have we disabled building on the poles yet? We definitely should if we add this effect.)

The ability also matches the intended role and design goal well - it makes conquest less desirable and peaceful economic expansion more effective, while also including a "and can defend myself if at war against stronger enemies" safety net.

Interesting idea about granting max pop bonus to your own race. I like it and fits the theme of a defensive player. The concern I have is that it will help AI players even more than it already does, due to their city multiplier bonuses (growth).

My wishful thinking alternative is to rethink the idea and make the bonuses tied to the range-multiplier variable (distance from fortress), as if the area around the capital benefits more than farther away, making the effect quite intuitive and applicable to a defensive player. This can help AI less as they tend to overexpand. Examples:
*Areas where 1X range multiplier - +6 max pop
*Areas where 1.5X range multiplier - +4 max pop
*Areas where 2X range multiplier - +2 max pop
*Areas where 2.5X or higher multiplier - +0 max pop


Considering it's a 2 cost retort that does absolutely nothing for the AI while they are not under attack, and their bonus on growth is very small compared to other categories, I don't think we need to worry about that.

Remaining question is, how much bonus do we want? I'd say 4 is probably the lowest we can go and about 8 is the highest?
Very bad spots tend to be in the 6-9 range (yea worse exists but they are actually quite rare), with granary they are already 10-14, so a +4 would put them at 14-18 (fairly decent), while a +8 would make them 18-22 which feels excessive.

Rune Master suffers tremendously from early game lack of benefits, and gets overshadowed by specialist for mono wizards.

Ideas:
*Start with Dispel Magic and maybe Disenchant Magic (a minor 400 rp saving that can help early in game)
and
*The research bonus being +100% instead of +50% (easier to get dispelling spells, especially disjunction, or can help speed up hero summoning)
and
*Something to protect the cheapest unit enchantments (e.g. unit enchantments when dispelled have as minimum a 25/125 mana cost in dispel calculations)


Edit: probably +4 or +5 is the most balanced and you made a good point about AI not benefiting from the +1 to all guardian bonus in general (unless directly attacked by you).


Edit: specialist is too much of an obvious choice, so my suggestion remains making it limited to wizards with at least 9 or 10 books of one kind - really a specialist.


How often do you find a pop 10-15 location? With granary that's 15-20. With this bonus that's 20-25, which seems absurdly good.

I'd personally go with +3. It's meant to be a minor extra as the retort is already somewhere between 1 and 2 picks, so the bonus should only be about half a pick worth. If we look at omniscient, a 5 nature book, dual realm player gets +3 pop. So that's half a pick. Therefore guardian should get +3 as well. Yes omniscient can work on all cities but I'd like to keep omniscient as the preferred choice for city growth, and guardian should be picked for the defense, with a minor bonus. Particularly on myrror it's very easy to conquer more cities of your own race, but even on arcanus, if your home race is high men (or nomads or high elves), you're likely able to conquer your home race as well.



Forum Jump: