Spells in treasure

[Seravy]

There are a few problems with spells found in treasure, and it might be a good idea to rewrite the entire procedure handling this.

1. If a spell of X rarity is to be found, and no such spell can be given in one of the realms the wizard has due to low amount of books, a lower rarity spell is given. Example : A very rare spell should be found. The wizard has 4 Sorcery 1 Life and 2 Chaos books. Actual spell found is 33% Sorcery very rare, 33% Life common, 33% Chaos uncommon.
The correct result in this case should be 100% Sorcery very rare. However, fixing this will make rare and very rare spells more frequent, especially for the AI, and will result in early very rare AI spells more often which might hurt game balance.

2. If a spell of X rarity is to be found and all spells of that rarity is known in that realm, the spell granted is of a higher rarity. Example : Common spell should be found, wizard has 8 Chaos and 2 Sorcery books, and knows all Chaos commons, uncommons and rares already, and all Sorcery commons as well as all Arcane spells. Actual spell found will be 50% Chaos Very Rare, 50% Sorcery Uncommon.
The problem with this one, it influences gameplay too much. It's significantly more beneficial to research everything first, then crack the lairs and get very rares out of common treasure, than doing it immediately. (maybe I'm the only one playing for this tactic? idk but when playing 6-9 books of a realm I always try to push research early to make sure my treasure gets as many rank-ups as possible.) On the other hand, if this mechanic is removed, spell treasure will be discarded if no valid spell can be found in the intended rarity, resulting in the loss of a few hundred treasure points. But as long as it's present, it encourages slow play which is most likely undesired.

3. Spells the player can research later (or even those currently being researched) can be found. It's questionable whether this is bad or good.
Not sure about the preferred solution. As long as spells can rank up, finding potential research is necessary, otherwise it makes "rank up" happen a lot more frequently, allowing easy access of all very rares even with 6-7 books. However if that is removed, the effect will be less drastic - treasure will still guarantee getting every common and most uncommon spells, but all rares and very rares won't happen unless the book count allows close to the maximal. On the positive side, this removes the other motivation for pushing early research - ensuring you find new spells within the same rank (a smaller benefit than leveling up the spell but still present)

[Nelphine]

I think this is very well laid out as to what the various issues are. I'll try to think of something.

[Domon]

Can't the treasure points value of the spell go back to treasure determination?

[Seravy]

Can't the treasure points value of the spell go back to treasure determination?

As is, they can't. I might be able to convert them into gold or mana as a workaround but it'll always need to be the same thing. There is no treasure generation during the game, it happens at map creation only.

[Nelphine]

So I think this definitely should be about how many very rates you want in the game. It nay be significantly influenced by Catwalks algorithm for guaranteed spells based on books.

But if we go under the assumption that we leave the spells per book as it is, then we shouldn't want to have a lot more very rares in the game.

Ideally, what I would love to have, but probably can't because its a new mechanic, is scrolls. Instead of finding spells, you find a scroll of a spell. You can either copy the scroll out into your spellbook (if its a rarity and realm your books grant access to) in which case you learn it just as finding s spell currently does, or you can use the scroll to cast the spell once - if used to cast the spell, then you don't expend the mana required, the scroll does so for you. However you must still provide the casting skill - so if you had 50 mana and 30 casting skill, and had a scroll of phantom beast, you wouldn't be able to use the scroll in combat until you had 35 casting skill. Similarly, if you had a scoll of summon naga it would take you a little over 3 turns to cast it, but (assuming no mama income) you would still have your 50 mana.

This system would specifically allow you to find spells you already know; and spells of realms you had no books in. If you do have any books or retorts that would affect casting a spell (such as conjuror, or 8 sorcery books in the example above) they WOULD apply to scrolls, by doing things like reducing required casting skill.

A scroll would only be used up once successfully cast. If you started to cast summon naga, and changed spells or canceled it after 1 turn, you would not lose the scroll.


If doing this system, I would increase the occurrence of found scrolls compared to the current amount of found spells. However, I would also increase the likelihood of finding common and uncommon spells, while leaving rare and very rare spells the same.

Super ideally, you could steal scrolls from wizards when you banish them. (1-4 per banish, randomly chosen from scrolls the banished wizard owns).

Additionally, when banishing a wizard, instead of stealing spells he knows, you would get to inscribe 1-4 scrolls - these would have to be spells he knows, of realms and rarity you have access to via your own spellbooks. 1 of these scrolls (if possible) would be of the highest rarity spell he knows that is also a spell you don't know. Any others would be randomly picked from spells he knows, but would be most likely to be of the rarity matching the highest rarity spell you have researched (say, 40% to match your highest rarity, 30% to be one rarity lower, 20% to be one rarity higher, and 10% to be the fourth rarity. In the case the case that your highest rarity is either common or very common, then 40% to be same, 30% to be 1 difference in rarity, 20% to be 2 different, 10% to be 3 different).

I'm fully aware this is probably impossible to implement as it would require a new tab in your spell book (both combat and overland) but whatever, I had to suggest it.

I'd also probably give a penalty to casting skill if you don't gave access to the right rarity or realm of a scroll. Like it takes 10% more to cast a spell of one rarity higher than you normally can in a realm (common if 0 books, very rare if 3 books); 25% more if two rarities beyond your access (uncommon if 0 books, very rare if 2 books), 50% more if 3 rarities (rare if 0 books, very rare if 1 book), and 100% more if very rare with 0 books.

I'll keep thinking more about more realistic suggestions.

[Seravy]

Interesting idea but the last thing I want is to see is a Life wizard casting Armageddon from a scroll while I get a scroll of, say, Disintegrate instead (which is a great spell of course but a single shot combat spell and a global enchantment are on a way different scale).

It's also redundant with the Spell Charges in items as far as combat spells are concerned.

[Nelphine]

It doesn't require heroes, and yes it would have lots of RNG. *shrug*

I just edited in a casting skill penalty for spells you don't normally have access to - you could add further penalties for overland, and/or global.

But that Armageddon would be super expensive with my edit, whereas the disintegrate would at least still probably yseable, though not terribly helpful.

And, you could sell scrolls for mana, as per selling items, so you could always just sell that disintegrate. Not sure how much they would be worth. Probably based on research value.

[MrBiscuits]

What would be the issue with just learning a spell from a realm you don't have enough books in? You are finding the spell rather than researching it from books after all.

It bugs me that you earn a reward but it is random luck as to whether you actually get it or not.

[Seravy]

What would be the issue with just learning a spell from a realm you don't have enough books in?

You make the largest scale decisions based on enemy books and retorts - this wizard is a threat, that one isn't, this I can defeat, that other I can not, etc., and you declare war or make treaties according to that. If treasure ignores books, those decisions cannot be made as books no longer tell you what spells that player might have right now, or in the future. That undermines the goals of playing a strategy game.

Meanwhile, completed the new system, although I won't release it until we decide on the book/spell system changes as well.

The new system works as the following :
-Only spells of the exact same rarity as the treasure can be found. Arcane spells are commons or uncommons, depending on research cost.
-If one exists, a spell not available for future research is chosen. If that's impossible, a spell not available in the current research options is chosen. If it's also impossible, a spell that is not currently being researched is chosen. If even that fails, you get what you are researching currently.
-The realm is not "locked in" before selecting the spell - having 1 Chaos and 4 Sorcery books and getting a Very Rare spell treasure will always result in a Sorcery Very Rare if one is possible. If all of them are already researched, no spell is given - no replacement Chaos common, or Sorcery Rare/Common/Uncommon.

As the entire procedure was rewritten, it's easy to modify it in the future if needed.

[zitro1987]

Is there a way to have 'very rares' cost much more treasure points compared to other tiers? Currently, the modifiers for the 4 spell tiers are 1X,4X,9X,16X ... Is there a way to have v.rare at 25X?

I know this seems conceptually like a minor number tweak, but v. rare spells are too important when you get them as reward, especially when you have fewer than 8 books or even 9 books. At a mere 16X, you only need less than 2500 treasure points I think, which is quite easy in mid to late part of the game. Doing a 25X (3750 research cost), it is more limiting, challenging, and balanced.