Finishing off the opponents

[kreso]

....see the point?

why all the arguing, then?

I don't see discussions as arguing. Be that as it may, we can always agree to disagree about something. I'm not saying SCS fighters wouldn't do themselves a huge favour by finishing of party members, but it may lead to very stupid behaviour on their part. The current solution is fine. If you really want your disabled charaters dead, you can install aTweaks Undead, Demons, Elementals and you''ll have a blast, believe me. It even installs on top of SCS .

[Salk]

I throw in my two cents here to say that I appreciate the current implementation because of the positive impact it wields on the frustration factor, so I am happy that SCS adapts this rule of thumb when it comes to prioritize targeting.

 

I am instead less convinced by the absolute necessity to avoid blatantly wrong tactical choices. Ironically for the same reason why DavidW believes they must be a top priority: immersion. Human players are obviously fallible and mistakes can be made. The fact that the AI makes mistakes is, in my view, a point taken towards humanity. There can be a wide array of reasons why mistakes are made: presence of mind, stress, wrong tactical evaluation, etc.

 

The problem would be when the same mistake is reiterated by the AI (excluding of course, all those creatures that should not be smart enough to quickly learn from their mistakes).

[DavidW]

Don't worry, the AI makes more than enough mistakes without my having to add them deliberately!

 

More seriously, there's stupid and stupid. It's easy enough to think of mistakes that are so dumb they cause the player to think "bad AI" rather than "this feels realistic".

[Salk]

Absolutely.

 

I was just talking in general. SCS has fulfilled all my expectations and thanks to some very welcome customizations, it is possible to adjust it to become my perfectly sweet cup of tea...

[Bearwere]

I can do slightly more nuanced variants ("attack the helpless only if XYZ occurs") and I'm happy in principle to hear suggestions for XYZ, but a typical SCS combat script for a fighter is 3000-4000 lines long (compared to 27 lines in baseline BG2) and I don't really want to make it any longer!

Anyway, so this thread wouldn't go in vain, here's one concrete suggestion: webbed/incapacitated (stinking cloud, don't remember if there are more spells like this) NPCs make a grand target for ranged fighters and those meleers who are immune to the effects - berserkered, flying monsters, etc. (Although I'm not sure that "helpless" includes "webbed" in SCS)

[ancalimohtar]

Honestly, I for one really appreciate the concept that enemies are trying to win, rather than cause you to expend as many resources as possible before your next fight.

 

Given two options:

 

i) 70% chance of dying against the player while expending a tenth of the player's resources, 30% chance of winning against the player

ii) 95% chance of dying against the player while expending half the player's resources, 5% chance of winning against the player

 

It's highly frustrating and immersion-breaking when enemy AI chooses option B rather than A. Fight starts out, 6 PCs versus 5 enemies, when the player has killed 4 out of the 5 enemies, and the last enemy just wants to backstab your mage to make you throw something at the screen, if it's practical to avoid that scripting behavior, please god avoid it.

[DavidW]

Well, achieving (i) rather than (ii) is definitely the objective of SCS, but I'm not convinced by your example. If it's 6 PCs vs 5 enemies and 4 out of 5 are dead, the fifth is basically screwed. Backstabbing the mage might well be their least worst option.

[ancalimohtar]

Well, achieving (i) rather than (ii) is definitely the objective of SCS, but I'm not convinced by your example. If it's 6 PCs vs 5 enemies and 4 out of 5 are dead, the fifth is basically screwed. Backstabbing the mage might well be their least worst option.

Wouldn't going invisible, drinking potions, and trying to run into a corner be more realistic behavior than backstabbing (which immediately makes him visible and therefore dead)?

[DavidW]

Well, achieving (i) rather than (ii) is definitely the objective of SCS, but I'm not convinced by your example. If it's 6 PCs vs 5 enemies and 4 out of 5 are dead, the fifth is basically screwed. Backstabbing the mage might well be their least worst option.

Wouldn't going invisible, drinking potions, and trying to run into a corner be more realistic behavior than backstabbing (which immediately makes him visible and therefore dead)?

Going invisible and drinking potions already takes priority over backstab. Running into a corner isn't going to help win a fight (and early playtesting made very clear that monsters shouldn't attempt to escape from even hopeless fights - it breaks quests.)

[Bearwere]

Honestly, I for one really appreciate the concept that enemies are trying to win, rather than cause you to expend as many resources as possible before your next fight.

 

Given two options:

 

i) 70% chance of dying against the player while expending a tenth of the player's resources, 30% chance of winning against the player

ii) 95% chance of dying against the player while expending half the player's resources, 5% chance of winning against the player

You see, the only option to win for the enemy is to expend 100% of player's resources. Your percentages are wrong.

[Jarno Mikkola]

You see, the only option to win for the enemy is to expend 100% of player's resources. Your percentages are wrong.

Ouh, so the first enemy comes and steals all the players money and the rest won't give him a thing to sell as gold... good point.

[ancalimohtar]

Honestly, I for one really appreciate the concept that enemies are trying to win, rather than cause you to expend as many resources as possible before your next fight.

 

Given two options:

 

i) 70% chance of dying against the player while expending a tenth of the player's resources, 30% chance of winning against the player

ii) 95% chance of dying against the player while expending half the player's resources, 5% chance of winning against the player

 

You see, the only option to win for the enemy is to expend 100% of player's resources. Your percentages are wrong.

 

...No? Let me rephrase:

i) 70% chance of dying against the player (player is left with 90% of his HP/spells/other resources), 30% chance of winning against the player (player is obviously left with 0%)

ii) 95% chance of dying against the player (player is left with 50% of his HP/spells/other resources), 5% chance of winning against the player (player is obviously left with 0%)

[Bearwere]

Ouh, so the first enemy comes and steals all the players money and the rest won't give him a thing to sell as gold... good point.

You like to play dumb, don't you?

 

...No? Let me rephrase:

i) 70% chance of dying against the player (player is left with 90% of his HP/spells/other resources), 30% chance of winning against the player (player is obviously left with 0%)

ii) 95% chance of dying against the player (player is left with 50% of his HP/spells/other resources), 5% chance of winning against the player (player is obviously left with 0%)

No. I might say it's i) 50% of dying and ii) 5%. It's just words. Can you back them up? Some arbitrary numbers can't be an argument.

[ancalimohtar]

Dude, I was just giving an example. The point is, a strategy with a higher chance of success, but if unsuccessful, not being a huge annoyance to the player, versus a strategy with a lower chance of success but always being an annoyance to the player (like killing the player's mage but having a very low chance of actually killing the player's party). If the enemies didn't think of themselves as a speedbump on the way to the "boss," they would obviously not care how much of an annoyance they were, but rather want the highest chance of success at all costs.

[Bearwere]

Dude, I was just giving an example. The point is, a strategy with a higher chance of success, but if unsuccessful, not being a huge annoyance to the player, versus a strategy with a lower chance of success but always being an annoyance to the player (like killing the player's mage but having a very low chance of actually killing the player's party). If the enemies didn't think of themselves as a speedbump on the way to the "boss," they would obviously not care how much of an annoyance they were, but rather want the highest chance of success at all costs.

Your point is valid. My point (where this thread started from) is that in almost any imaginable situation (in almost any situation I've been in, anyway) killing a helpless enemy increases the chance of success.