Really annoying spell scripting issue - Bug?

[vfb]

If the opposing army is not considered strong enough, spells that use gems will get scored down. If you've got Wind Guide as spell 1, and it gets scored down, your choices for spells 2 to 5 are not considered. It will just use the highest scoring spell that it found when it computed the spellscores for your caster in round 1. Most unfortunately for your archers, that was "Touch of Madness".

[Networkingguru]

vfb said:
If the opposing army is not considered strong enough, spells that use gems will get scored down. If you've got Wind Guide as spell 1, and it gets scored down, your choices for spells 2 to 5 are not considered. It will just use the highest scoring spell that it found when it computed the spellscores for your caster in round 1. Most unfortunately for your archers, that was "Touch of Madness".

I guess that makes some kind of warped sense, but is there any particular reason ToM is scored so high? I can't imagine a single situation in which that would be my first pick for a spell.

[Folket]

How come I have never had this problem?

[Sombre]

Playing as Man and Pangaea it seems if you have touch of madness researched and they at any point deviate from the script (you left them unscripted or they didn't want to cast the spells you scripted) it will be the spell they cast. Not something nice and safe like storm of thorns or protection or something. No, it's touch of madness every time. Especially if there are archers nearby.

It was modded to cost 1 gem in cbm, which makes it much less likely to get cast.

[Edi]

I haven't seen this particular problem in my own games, but that doesn't mean much because I use such difficult research settings that I need to prioritize carefully, therefore this often comes late.

I can add it to the shortlist with a question mark.

[Networkingguru]

Edi said:
I haven't seen this particular problem in my own games, but that doesn't mean much because I use such difficult research settings that I need to prioritize carefully, therefore this often comes late.

I can add it to the shortlist with a question mark.

Thanks Edi! I have screenshots if you want them.

[Edi]

Networkingguru said:

Thanks Edi! I have screenshots if you want them.

I don't think they are necessary. It's by no means a unique problem, ToM is one of the issues that has caused more than its fair share of angry smilies on this forum. A savegame where such a battle is taking place that turn would be more valuable from the analysis point of view.

[vfb]

I ran a simulation with all N spells researched, with an N3 caster and a small army, and the N3 caster scripted just to "Cast Spells", and holding two N gems.

Here's how the spells scored:

Eval: Sleep Touch score 0 (fat 2)
Eval: Vine Arrow score 14 (fat 5)
Eval: Sleep Cloud score 0 (fat 6)
Eval: Breath of the Dragon score 0 (fat 6)
Eval: Poison Cloud score 0 (fat 10)
Eval: Stream of Life score 35 (fat 10)
Eval: Eagle Eyes score 118 (fat 1)
Eval: Poison Touch score 0 (fat 1)
Eval: Resist Poison score 0 (fat 1)
Eval: Barkskin score 59 (fat 1)
Eval: Protection score 244 (fat 5)
Eval: Elemental Fortitude score 4 (fat 2)
Eval: Swarm score 1526 (fat 25)
Eval: Wooden Warriors score 1110 (fat 16)
Eval: Creeping Doom score 495 (fat 100)
Eval: Mass Protection score 1778 (fat 50)
Eval: Polymorph score 0 (fat 100)
Eval: Tangle Vines score 0 (fat 5)
Eval: Maggots score 0 (fat 25)
Eval: Howl score 530 (fat 66)
Eval: Wild Growth score 0 (fat 40)
Eval: Healing Touch score 0 (fat 5)
Eval: Personal Regeneration score 772 (fat 13)
Eval: Regeneration score 245 (fat 20)
Eval: Heal score 28 (fat 5)
Eval: Haste score 605 (fat 50)
Eval: Poison Ward score 17 (fat 33)
Eval: Serpent's Blessing score 6 (fat 100)
Eval: Berserkers score 115 (fat 6)
Eval: Sleep score 5 (fat 6)
Eval: Panic score 0 (fat 6)
Eval: Touch of Madness score 1580 (fat 15)
Eval: Growing Fury score 128 (fat 100)
Eval: Charm score 0 (fat 15)
Eval: Storm of Thorns score 150 (fat 6)

"Touch of Madness" gets scored really high because for some reason buffs are weighted much higher than attack spells. I can't find any way to mod the score lower. Nice stuff like "Storm of Thorns" is stuck at 150 points.

But there are 2 bright spots of hope here:

You'll probably cast "Mass Protection" instead of "Touch of Madness". But there's a long way between Thau 4 and Alt 7. So this is actually a dim spot.

Here's the actual bright spot: "Swarm" just barely lost out.

So here's a potential Touch of Madness workaround: If you mod "Touch of Madness" from fatigue 30 to fatigue 50, and your N3 caster always has N gems and has researched Alt 4, then he should cast "Swarm" instead of "Touch of Madness", when he doesn't know what to do. And "Swarm" is actually a nice spell to cast.

An easier workaround is just to mod Touch of Madness to fatigue 300, and don't carry 3 N gems.

As a side note, "Legions of Steel" is totally hosed as far as score is concerned ... the "Number of effects=5" effectively multiplies its score by 5, to the 6000 range. There's nothing even close. But at least legions is occasionally just a waste of casting time, not actually damaging like madness.

[lch]

vfb said:
"Touch of Madness" gets scored really high because for some reason buffs are weighted much higher than attack spells. I can't find any way to mod the score lower. Nice stuff like "Storm of Thorns" is stuck at 150 points.

Low fatigue, some kind of buff, what's not to love??

vfb said:
As a side note, "Legions of Steel" is totally hosed as far as score is concerned ... the "Number of effects=5" effectively multiplies its score by 5, to the 6000 range. There's nothing even close. But at least legions is occasionally just a waste of casting time, not actually damaging like madness.

Ah, that explains why unscripted mages like to spam that spell. Even though it's quite useless with the number of effects limited by 5, of course.

[vfb]

It's area 25, number of effects 5. But the effects all strike together. In each "effect", the aoe will not overlap. But the aoe from multiple effects can overlap.

Like this (the log is in "typescript", one cast of Legions in the log):

# grep "aoe1 dmg64 eff10" typescript
blastsqr: unr2034 x21 y15 aoe1 dmg64 eff10 spc1086324736 as10193 al9
.... etc ...

# grep "aoe1 dmg64 eff10" typescript | wc -l
125

# grep "aoe1 dmg64 eff10" typescript | head -25 | sort -u | wc -l
25

# grep "aoe1 dmg64 eff10" typescript | sort -u | wc -l
48

This is with five clusters of army units; the same single cluster was chosen for a "hit" by all 5 "effects". All five clusters were in range, centered about 6 squares away, but max distance from each other. Only the units in one cluster got affected.

Legions of Steel could be just the equivalent of about an aoe-50 buff, for scoring-calculating purposes.

And Legions does rank below some self-buffs, depending on who's casting. A Marverni druid preferred to spam Legions.

A A3D3S3F3E3N3W3 pretender preferred to self-buff first:

spellscore, Invulnerability score 41579 (boost 99 scorat 0)
spellscore, Astral Shield score 36663 (boost 90 scorat 0)
spellscore, Fire Shield score 18540 (boost 91 scorat 0)
spellscore, Legions of Steel score 15044 (boost 116 scorat 0)

And the best part is, none of this analysis help in the slightest for stopping ordinary E2+ casters from spamming it.

The other "best" part is during testing I finally ran into Satima and her fancy robe, first time ever ... grrr...