SORCERY Realm

[zitro1987]

Common Notes:
_Nagas with +1 melee is important to keep sorcery competitive early on. Maybe even a slight cost reduction or a +1 movement (a water-walking war bear may be superior anyways)
_Guardian Wind feels expensive, especially having 1 upkeep. I find it inferior in scope to resist magic (as arrows irrelevant to summons) and inferior strategically against warp wood (whole army needs it to be useful). It also costs much more. Suggest blocking also 'rocks' instead of a traditional missile immunity. Otherwise, consider 0 upkeep.
_Counter Magic - I suggest changing formula from 100% to 150% of mana spent and reducing the cost range down to 10-50. The minus 10 power reserve per success, combined with a weak 100% formula makes this spell feel desperate, not 'powerful'. As a 'common' when skill points are low, the spell is pathetic.
Uncommon Notes:
_Dispelling Wave - we seem to have a bunch of ideas. Maybe a new idea is to keep it being 2X powerful with aether binding, but reducing the combat unit dispel from 1/2 to 1/3 and maybe the overland dispel from 100% to 2/3 ... IMO its strongest effect is how it impacts up to dozens of enchantments.
_Not sad with losing conjure roads. Good decision.
_Psionic Blast - Damage increasing to 20 points (from 16) would make it a formidable uncommon spell.
_Spell Blast - I like the new concept but seems hard to implement, hard to teach AI, and hard to explain to a player, or hard to keep track of the massive '50' turns duration. I'm optimistic that you'd make it work, maybe with a warning message when the 50 turns expire. I might like it more as a less 'SP' cost but much lower duration. 50 turns feels like forever while 2X research cost for skill is brutal.
_Flight - maybe 20/100 cost is better, even if it results in magicians giving themselves flight.
_Vertigo - might be cool if hard to resist or movement-impacting but costing 20.
_Flying Fortress may be powerful for uncommon but doable with 800 research points.
_Spell Ward may be better as a slightly weaker 'rare', comparable in usefulness as cloud of shadows rare from death. I'm a little wary of uber-powerful defense enchantments for fun factor. If so, It'd be -1 to hit, -3 attack, -3 armor, -3 resistance, and the spell prevention.
_Ideas of new rare/very rare spells:
*A teleport unit enchantment, maybe limited to combat only (teleporting sky drakes!)
*bringing back word of recall, but as a very rare
*Altar of Wizardry - a city enchantment that gives 'focus magic' to all recruited units
*Global Variation of above - all normal units you control have 'focus magic'
*Tornado - a variation of the chaos vortex, no random movement, but you get 1-2 moves only to deal the damage to anything it crosses
*a combat enchantment that boosts movement or flight to all
*Morph Artifact - Select an item of yours and makes a random generated artifact of similar cost/power.
*Steal Artifact - An ultra-expensive spell like spell-binding, but relating to items.

[Seravy]

_Spell Blast -

Don't worry, I can implement it and it's not that hard to teach the AI either (although random use should most likely be disabled entirely). Spells can have states 0 (unknown) 1 (can research) 2 (known) and 3 (on research list). It's stored as a byte value so we can have 10-110 mean there are than many-10 turns left until the spell is reverted to "known" status again. Will be a lot of work but I rather have that than lose a game to Spell Blast which wasted 5-30 hours of playtime.
I'm considering a higher duration actually - as it's fairly difficulty to "catch" some spells the player might want to stop, having to do it regularly is somewhat unrealistic.
(we need to discuss how many spells should be possible to seal, and if Magic Spirit is allowed or not.)

_Vertigo - might be cool if hard to resist or movement-impacting but costing 20.

Stasis does both of those things (albeit it's rare).

Flying magicians to go along with uncommon flying fortress is a good idea.
Although...maybe I should reconsider.  Only Sorcery can grant Flight to units. Nature can Web them but that doesn't help unless the defenders are already outside. (or not even then, I don't remember if FF or web takes priority) Other realms need direct damage spells (which Sorcery can counter) or ranged units, but ranged is effectively countered by Sorcery through Invisibility and similar spells.

Added Counter Magic to the list.

[Nelphine]

Dispelling Wave: Yes, buffs have an order. Can you make it pick in reverse order? Much harder to abuse that way. But also, what I honestly expect to happen is that the AI will spend.. 3 turns dispelling wave, and everything will be gone anyway. So I don't expect it to be much different, except it will cost the AI a more reasonable amount of mana (and importantly, time). And overland, it's much harder to abuse the order, because, again, it's about time. I'd really like to work to find some sort of compromise to make this work.

Spell Blast:
Example AI only gets 1 rare and 1 uncommon summon. I can spell blast both for a paltry.. 4k? Skill points or something? (Again, remember spell blast is uncommon. Very similar to your obvious, and valid, concern about magic spirits.) Now my enemy has no strong overland presence. Given that most people will only have 3-6 spells of any given rarity, I think completely blocking them out for any more than 1-3 turns would be far too abuseable. (Imagine having your own rare summon, say storm giant with resist magic; and ALL of the AI are forced to use city troops and common summons to fight them.)

So I think the cost for spell blast would have to be based on very rare spells REGARDLESS of what kind of spell was blocked; so, for example, any time you used spell blast, it would cost say 16000 skill points, no matter what spell you hit. (Except that makes spell blast completely useless until very late in the game, which seems bizarre for an uncommon spell.)

But lets even look at that: You're sorcery, you're playing spellweaver, you have amp towers. Lets say you're using it as expected, and it's already very rare state of the game. So you have 20 amp towers. You also have 250 casting skill 'naturally' from power production expenditure. One use of spell blast will drop you to 215 casting skill - this is significant right? (35/250 means losing 14% of your casting skill) Except, your 'actual' casting skill is 580. So you're only losing 6% of your casting skill. Not such a big deal. Worse, I would assume that spell blast can't make you go negative (the game wouldn't handle that at all is my guess), so you could just repeatedly cast spell blast on EVERYTHING, and you'd still never go under 330 casting skill. Amplifying towers would just become even more important than they already are to sorcery. (And knowing this, spending anything on casting skill after a certain minimum would actually be a waste, and so sorcery wizards would simply stop doing so, with the knowledge that with spell blast they would get rid of it all anyway.)


Lastly: While the idea for spell blast is interesting, doesn't it strike you as extremely similar to spell binding.. except.. you aren't limited to globals, and you don't get the spell yourself? (The only significant difference being SoM since once it's finished casting, you don't have the opportunity to cast spell binding again. So why is an uncommon overcoming the major weakness.. of one of the strongest very rares?)

Can you list every example of scenarios where you think it is good for spell blast to exist? (I'm inclined to just say spell blast should be scrapped, EVEN knowing how much work and effort went into making it playable. I know that's probably never going to happen, which is why I'm trying to understand the scenarios you think it's good for it to exist.)

[Seravy]

An idea for Counter Magic.
It's countering spells indiscriminately, so it's a perfect example of the type of unfun effect we want to get rid of, right? On top of that it manages to be both too weak to be really useful, and absolutely useless as a common (who can afford 70 MP in the early game? On the other hand in the late game it has a...14% chance to counter  Flame Strike, oh joy. So this is kinda only useful in the endgame where 70MP is worth it to counter 2-3 medium cost spells at random from the enemy (such as Prayer and two Lionhearts or something like that) and you still have plenty of skill to cast spells afterwards.

So, we mainly use it because to protect troops from getting hurt by enemy spells, right? At least I usually want it when I expect some nasty direct damage coming towards my important unit. Even the AI is somewhat like that, it will use it only when it has a large army in the combat. If there is nothing to protect, why bother?

So what if Counter Magic was a different spell :

"Spell Blur"
Whenever one of your units would be affected by a harmful enemy spell, there is a X% chance that spell misses and your unit is unaffected by it.

X could be the same 25% as Blur, or maybe 33%. So it's like a very limited Magic Immunity for your entire army.
Would be a nice pair with Blur especially if that is moved down to common.

(coding the part to dodge special spells like Massacre which are neither curses nor direct damage would be a pain though.)

A more straightforward approach would be "X% chance to counter any enemy spell that can hurt the caster's units". So we just make a list of spell IDs or add a flag to spells and what gets listed can be countered. It's even possible to make the X% a variable - either depending on slider, or the cost of the spell, or have Runemaster/Specialist/AEther Binding affect it.

3 turns dispelling wave,

And that's one problem. A single casting of this costs 125. The AI can reasonably do it once in the early-mid game, twice late, maybe 3 times very very late but by then they dispelled all of it through overland use anyway.


Spell Blast
Oh well, I'm out of ideas then. (Amplifying Towers do mean using SP for the cost is not going to work.)

Why is it necessary?
Well, Sorcery is about "Control", just like in MtG. There are spells in the game that have immediate effects - if you can't stop them, you fail to control the game and lose.
Armageddon : yes, you can dispel and you can steal it but the volcanoes raised while doing that stay there.
Doomsday : same, plus the units that disband due to lack of gold (+5 rebels might mean your gold production drops into heavy negatives) and people that starve to death...
Meteor Storm : By the time you deal with it, you already lost a few dozen buildings and a few hundred units outside cities, for example all your node garrisons.
Spell of Mastery : Obviously, after losing the game there is no way to fix that :D
Final Wave : Yes, you can protect your units in cities by warding all of them but you need to do something about it while researching the ward and casting it. Warding 30 cities take a while. Also, you need to replace all nongarrison and node troops with 10+resistance units or buff them - again, takes time.

There are probably a lot more, these are the most obvious examples that come to my mind.

Edit : The current Spell Blast fails at this goal btw, as the AI will often be able to instant cast these spells except the SoM.

[Nelphine]

Ok, so spell blast is intended to be a way to avoid .. the very spells that already trigger maximum diplomatic penalties. I'm not sure I actually like that intended goal, because it steps on the toes of those very diplomatic penalties. Those spells are already extremely difficult to use because of the diplomatic consequences; and one of the reasons I don't even bother playing chaos anymore is because I can't afford to cast any of the listed chaos very rares; it just makes everyone mad at me.

I guess another way of saying it - i feel like it's overkill to have spell blast as a specific counter to those spells, when every ai already has a counter in terms of diplomacy. Obviously diplomacy isn't AS effective, but they're still countering the exact same things.


(Dispelling wave: I'm probably biased due to playing on lunatic, but generally by the time an AI gets dispelling wave, they're already close to getting 200 casting skill. Multiple casts is a real thing. Especially when combined with heroes. And even spellcasting creatures can cast weak versions of it, which, with runemaster or aether binding, still dispels a massive amount. Also note, I really want the diminishing returns to work overland as well.)

[Seravy]

ehhh....no I was talking about it from the viewpoint of USING Spell Blast, not being on the receiving end.
Diplomatic penalties have no effect when the spell is cast against the human player, and have reduced effect when the spell is cast by other AI. So they aren't relevant.

Bleh. I did tell Hadriex "if we teach the AI to use Spell Blast correctly we'll regret it" when he complained about the AI not stopping another AI's Spell of Mastery and Armageddon.

I don't care about diplomacy when I can cast the Chaos rares as a human player either. So what, they'll declare war? Ok, I burn all their units away with my combat spells because I'm the defender. And I can pay for it thanks to the power from Armageddon...not that much of the units even make it there through the Meteor Storm or manage to get produced despite every city being 60% rebels and randomly losing buildings while at it.

...it might be relevant on Lunatic but up to Extreme I had no problems with that. I had problems with Disjunctions but we've already made the chance not stack and even get divided by the number of wizard so...that is not the case anymore.

(also, no diplomacy or war is going to stop your Spell of Mastery while you have 9 regenerating or just plain absurdly powerful units in your capital)


If we can't come up with a better solution, we can remove Spell Blast from the list of curses, and have the AI only use it through Detect Magic which the human player can dispel..but that doesn't fix the problem that the AI can instant cast stuff you'd want to stop now that there is an overland casting bonus.

[zitro1987]

Spell Blur - I like the idea of a combat global that attempts to counter direct damage spells and remove the unfun aspect of counter magic, but as a 'common' the spell countering 25-33% of spells would be mostly useless until later in game when AI has lots of skill and spams these spells against your precious heroes and advanced summons. Maybe we can combine this situational benefit with the mechanic of counter magic and have its usefulness peak early on.

Spell Blur - Common - 25 cost - Direct Damage spells from AI (including flame strike/wave of despair) have a cost/(40+cost) chance of succeeding. At first, this is a near-guaranteed counter to early spells like star fires, fire bolt, even ice bolt/lightning bolt. More advanced spells have half a chance of succeeding or more.
*a weaker 20 cost version could be cost/(25+cost) chance.

Dispelling Wave - I still stand by my latest suggestion to weaken the unit enchantment effect into both combat and overland by a 1/3 ... meaning overland power is 67% and combat is 33%. It would still be a very powerful spell with aether binding and much stronger than dispel magic under most circumstances.

Spell Blast is still tricky. I can live with the current version, though AI generally benefits from its usage.
*Different Idea #1 - Spell Blast as is, but the user casting spell blast cannot cast spell blast for the next X turns ... requiring user to cast it at the right time. - Maybe lower cost to just 25.
*Different Idea #2 - a weaker spell blast that briefly de-activates ability to cast spell for 2-3 turns. Maybe again with 50% chance of successful counter.
*Different Idea #3 (might be impossible game mechanics) - cost 80 - attempts to learn an AI spell (disable spell if not possible due to different realm, book limitation, etc). Chance of success (which also counters the spell) is 80/spell research.
*Different Idea #3a - somewhere between original and idea #1 - cost 50 - attempts to counter spell at 50% chance. If successful, you lose the mana like in original. Meanwhile, you have a very small chance of successfully copying enemy spell at probably 50/spell research.

[Nelphine]

Idea #27, spell blast becomes an arcane spell, only AI can cast, and can only target AI, and so you can specifically use it for spells that the AI is horrible at dealing with.

[Seravy]

Spell Blast. We have 3 separate problems so we can try fixing them each separately.

1. To solve the part of the AI crushing the player with it by preventing most or all of their spells, we can reduce/remove/alter random use. Or limit it to not happen until the game reached the phase of "rares". The main problem with random spell blasts is, at the uncommon stage of the game the player can't cast anything relevant faster than 2-5 turns which is more than enough time for a Maniacal wizard to roll "I want to use a curse", so the player ends up not casting anything or at least losing half their overland casting attempts. Spell Blast doesn't need to be rare, it's enough if the AI isn't using it until turn 150-180, alternately, until the intended target has more than 150 casting skill.

2.  Intentional use is already taken care of through low recast rate of Detect Magic. The player also has the option to instant-cast spells, or time them to when the AI is busy casting an expensive spell...except with the casting advantage they can't. A good solution to this would be to make sorcery wizards use Create Artifact more often - those are fairly expensive and put the wizard out of action for about 6 turns (we can readjust the target artifact cost if not, not sure how many was set up but the AI does pick a cost based on how many turns it takes to cast)
Not a fan of the AI making more artifacts the player can steal and use against them, especially as Sorcery's main weakness is heroes with items unfortunately. Sorcery items ARE good but not if you give them away to the enemy. Unfortunately with the 50% AI casting cost, I don't think it's realistic to raise the cost of any Sorcery spell high enough that the player can get a very rare global though - but a 500 cost spell (most summons) can slip through while the AI is casting Spell Binding, Time Stop or such.
I did like how Detect Magic told me when I can safely cast a spell, and I miss that - would be nice if it could happen more. (Although this is kinda pointless as casting Disjunction is an easier and more effective solution - on the upside, timing your spells means you can avoid random blasts too.)

3. To solve the part of the player being unable to use Spell Blast on the intended targets due to the AI casting too quickly, a simple solution is to increase the cost of the intended target spells.

Spells over 1000 cost I'd consider "safe". Yes, Lunatic AI on huge landmass might be able to instant cast those in the endgame but by then they probably already won the game. Assuming we do change the casting advantage to 50%, it means they have 500 overland skill. (Spellweavers will still do it but I can live with Spell Blast not working well on Spellweavers. The name kinda implies their magic is hard to "break".)

So let's look at what I want to stop. (then I'll also list what I don't) (this is different from the AI's list)

STOP
Final Wave - this costs 500. Why? It kills hundreds of enemy units. When Hadriex used it, it HALVED the historian graph of the top wizard. Admittedly that person was playing barbarians but still, I see no real reason why this can't cost 1000.
Clairvoyance - Once the AI saw all my cities, it's too late. Cost 900, safe. By the time it's instant, I will have finished warding my cities against curses.
Spell of Mastery - expensive enough
Time Stop - expensive enough
Spell Binding - expensive enough or I can steal the spell back anyway
Armageddon - 1250, expensive enough
Doomsday - 1000, expensive enough
Meteor Storm - 800, too cheap (is costed that low to make it easier to dispel though - not sure about raising cost here is a good thing to do)
Evil Omens - 1250, safe. Once in effect, dispelling it costs 50% more so not good to let it get cast.

IGNORE
Nature's Wrath - Spell Binding offers a 100% dispel chance on this, even if Specialist, so I don't need to take a hit from it for a failed dispel. (on the other hand, at 800, the AI can instant-cast it so I can't cancel  my own spell in time and get hit by it. So a higher cost might actually be better, even if that means more expensive dispelling, but for that other reason. )
Great Unsummoning - I can protect my creatures with Spell Lock and Resist Magic if necessary. The latter is a good idea to have on more valuable units anyway.
Suppress Magic - steal or or while expensive, can be dispelled.
Summons - Spell Ward deals with that for self defense, and blasts shouldn't be for offense so denying the enemy to have creatures is wrong. I admit I mainly used it for these in the past but doing so isn't fair at all.
Call the Void - Spell Ward!
Create Artifact - I like it when the AI makes artifacts for me, yay!

[Nelphine]

I'm really confused by your ignore list (aside from create artifact); spell blast is something you could have and easily not have spell binding; similarly, you might not have nearly enough available casting skill to resist magic/spell lock a lot of your creatures. Suppress Magic specifically stops you from.. spell blasting. It's exactly the kind of spell spell blast should be for. Spell ward, again, very good chance you don't have it, even if you have spell blast. Call the Void - ok, i can accept this isn't really for spell blast, same thing with summons.

So to me, nature's wrath, great unsummoning, suppress magic (and what about death spells? Even eternal night should be on their, and some of the life ones; they may not be offensive, but life force or charm of life is absolutely a nightmare to successfully remove from the AI, especially if they're a specialist, and you don't have aether binding.)