Suggestion - Black/Whitelist for Battlespells

[pino]

It's a good idea

[Folket]

I hope you mean 8th Century.

[HoneyBadger]

nope...

[calmon]

Example from one of my mp games:

After scripting my pretender (4E/4F) with Summon Earthpower, Phoenix Power, Flame Bolt, Flame Bolt, Flame Bolt (i was fighting niefel giants) and 'cast spells' in battle turn 6 he started to cast the level 0 Spells Fire Flies and Flying shards. Fatigue was below 20. The spreading level 0 spells kills mainly my own troops, the enemy giants didn't notice anything. I don't know why not 1 flame bolt or something equal was cast by ai, it costs the nearly same fatigue and works a way better then the level 0 spells against fire susceptible giants.
Even the stupiest battle ai should be able to choose a spell at random. This is even better then to choose the worst possible spell!

After some Fire Flies/Flying Shard battle rounds later my army rout and my pretender flees with a very low fatigue value. Maybe i'm the 4721st person who say this but: thats just stupid!

[Endoperez]

calmon said:
Example from one of my mp games:

After scripting my pretender (4E/4F) with Summon Earthpower, Phoenix Power, Flame Bolt, Flame Bolt, Flame Bolt (i was fighting niefel giants) and 'cast spells' in battle turn 6 he started to cast the level 0 Spells Fire Flies and Flying shards. Fatigue was below 20. The spreading level 0 spells kills mainly my own troops, the enemy giants didn't notice anything. I don't know why not 1 flame bolt or something equal was cast by ai, it costs the nearly same fatigue and works a way better then the level 0 spells against fire susceptible giants.
Even the stupiest battle ai should be able to choose a spell at random. This is even better then to choose the worst possible spell!

After some Fire Flies/Flying Shard battle rounds later my army rout and my pretender flees with a very low fatigue value. Maybe i'm the 4721st person who say this but: thats just stupid!

I don't know the inner workings of the AI, apart from the fact that it simulates few "what would happen if I did this" few times and chooses the best option, but this is what I suspect happened:

The AI checked what a Flame Bolt might do:
1) It could cause 22+ (at least 25) AP (+fire vulnerability) damage to one target
2) It could miss
3) It could hit a friendly unit, although it isn't likely

The AI checked what Fire Darts would do:
1) It would fire 7 (at least; your god had Fire 5 and you get at least one more Dart per exta level) darts, each doing 10 AP (+ fire vuln.) damage if it hits.
2) Some or all could miss
3) Some or all could hurt own troops

After checking both spells (and probably all other ones you had researched), it decided that Fire Darts was the best choice? Why? Perhaps Fire Darts dealt more hp damage in total, just divided among the Niefel giants more evenly so that the net effect is weaker (esp. if they have regeneration). Perhaps the friendly fire was considered just as effective as the damage caused on the enemies, and the fact that this had much worse relative effect on your own units than on your enemies wasn't considered.

In short, I think the AI calculates the amount of damage dealt in total, not how much the spell will help in killing the enemy. Most often, the first is equal to the second, and the first is much, much easier than the second.

[calmon]

The AI didn't cast Fire Darts (it was researched), it used Fire Flies and Flying Shards.

[Endoperez]

calmon said:
The AI didn't cast Fire Darts (it was researched), it used Fire Flies and Flying Shards.

I don't know how the AI calculated the result, but in that case, Fire Flies was found to have done better results than Flame Bolt or Fire Darts. I don't know why or how. The host didn't happen with debug on, right? The log file would help a lot in trying to find out the why.

[Graeme Dice]

Endoperez said:
I don't know how the AI calculated the result, but in that case, Fire Flies was found to have done better results than Flame Bolt or Fire Darts.

Which is quite frankly, impossible, since the spell is strictly inferior to fire darts.

[Endoperez]

Graeme Dice said:
Which is quite frankly, impossible, since the spell is strictly inferior to fire darts.

That's why my answer was followed by "I don't know why or how. The host didn't happen with debug on, right? The log file would help a lot in trying to find out the why."

I don't know how the AI works. I don't know how it got result saying Fire Flies is a better choice.

[calmon]

I replayed one more and here a complete battle spell review (i've overseen 2 times the use of Fire Darts):

First 5 Battle Turns like i scripted: Summon Earthpower, Phoenix Power, Flame Bolt, Flame Bolt, Flame Bolt
Battle Turn 6 / Fatigue 16: Ironskin
Battle Turn 7 / Fatigue 17: Flying Shards
Battle Turn 8 / Fatigue 20: Fire Darts
Battle Turn 9 / Fatigue 21: Flying Shards
Battle Turn 10 / Fatigue 24: Fire Flies
Battle Turn 11 / Fatigue 25: Flying Shards
Battle Turn 12 / Fatigue 28: Flying Shards
Battle Turn 13 / Fatigue 31: Fire Flies
Battle Turn 14 / Fatigue 32: Flying Shards
Battle Turn 15 / Fatigue 35: Fire Darts
Battle Turn 16 / Fatigue 36: Flying Shards
Battle Turn 17 / Fatigue 39: Fire Flies
Battle Turn 18 / Fatigue 40: Flying Shards
Battle Turn 19 / Fatigue 43: Rout & Flee

Instead of using mainly fire magic (because of Fire Susceptible) there was an use of:
7 x Level 0 Flying Shards
3 x Level 0 Fire Flies
2 x Level 1 Fire Darts
1 x Level 3 Ironskin

Alteration 3 and Evocation 2 was researched.

I would never choose the level 0 spells, the level 1 Fire Darts is better in all cases. Especially for regenerating giants i would like to see flame bolts which costs just a little more fatigue (20/4=5).

To be on my suggestion: I had blacklist Fire Flies, Flying shards and Flare (because giants are big so bolts are better vs giants) and whitelist flame bolts.