To No Name:
"That is simply player error. And a player blaming the casting AI for mages going off-script due to friendly fire concerns (or range, lack of tvalid target issues etc) is just that player looking for lame excuses so that he doesn't have to blame himself for his own mistakes and failings as a player."
Nope. Not knowing the AI's flawed thinking in advance is not "player error". That is like....well...calling a bug an undocumented feature.
"Anyone who says this sort of thing is a game bug, casting AI bug, casting AI short-coming, or any other similar object of potential blame transference, needs to spend more time learning better scripting and positioning, and less time finding things to blame their playing mistakes on."
Anybody that makes a comment like this is compensating for something
"And for every player who says the casting AI should err on the side of recklessness, you will find another player who says it should prefer caution instead."
That is a debating strategem called a false choice or false dilemma. The choice here isn't between erring on the side of recklessness or erring on the side of caution (since casting a high fatigue combat spell before wading into melee carries its own risks and recklessness). The choice is between following a script as close as possible or deviating from the script and substituting choices that do not take into account the overall script or strategy employed.
" It's impossible to make both camps happy, and futile to even try. At the very least the code would need a massive re-write before progress cold even be started on making "both camps happy"."
Again, you are assuming the false choice. This conversation in proper context (the one that I had written my patch request with in mind) could be very simply implemented - a) mages do nothing when their spell would have a (to the AI) "reckless" effect, b) mages skip that step and go onto the next scripted order, c) -this would require more than 2 lines of code, but the casting AI could look at the last order (attack melee or cast or retreat) before deciding whether to cast high fatigue spells or not.
"There are far better things for Illwinter to be spending their time on if improving Dominions was the objective."
I know of no other areas needing improvement that have caused people to quit the game. What areas fail the potential of this game more than the battle AI?
"And far better ways of improving the casting AI than your suggestion. (if improving the casting AI/playing AI was the aim)"
I was not too specific, but I'm sure they could use any advice you have. All I was voicing was a wish for a battle AI that will stick to the script a little bit more often. There are many ways that could be done that I would be satisified with.
"Maybe think about scripting Cold Resitance instead of BoW next time. Or Cold Resistance followed by BoW if you really need the BoW effect."
No point having the CR in that case without putting BoW into play for that battle. Scripting both CR and BoW adds additional fatigue and in some scenarios could take more lines of scripting than you have for buffing a thug. If I had known the AI had this particular poorly prioritized concern for the safety of their allies I would have forgone the group buff and positioned them seperately. As it is, I had no reason to suspect that PBAOE cold auras took into consideration friendly fire since Winter Wolves, and casters that have already cast BoW have no qualms about bunching up as close to as many of their unprotected allies as possible during combat. Not to mention, several casters that have cold or poison auras will cast stone skin or invulnerability respectively on themselves given half a chance. No indication at all before having scripted this, that the AI would behave that way.
" Or Cold Resistance on the thug that appears higher up on the commander list, followed by BoW on the thug appearing lower down. As that way the first thug will be 100% CR by the time the second thug wants to cast BoW, preventing any friendly fire concerns (assuming no negative CR modifiers are being applied), and that way you will at least have BoW on one of the thugs."
Yes, I am aware that if you know in advance that the AI is prone to one mistake or another there are sometimes work arounds to prevent them from exhibiting that particular brain fart, but it doesn't mean that the AI doesn't need improvement in order to "deserve a place alongside modern titles".
Re: Dominions 3 no longer available from Shrapnel?
Linear Mode
