Should Beamdog improve AI in BG:EE and BG2:EE up to SoD's level?
WatchForWolves
Member Posts: 183
It seems to me that all original Beamdog content uses "SCS-lite" scripts for its enemies. This is most obvious in Siege of Dragonspear, hovewer any new content in BG:EE and BG2:EE - mostly the personal quest stuff for Neera, Rasaad and Dorn - will also utilize it.
Not only that, but if the creatures used in those encounters(or SoD) also spawn elsewhere, they will inherit this "smart" behaviour.
To those unsure what am I talking about, let me give a few examples. In the initial fight with Ekandor in Beregost, the Thay Mage will try to move away from your characters when he's not casting spells. The Tasloi, even ones randomly spawning in the game, will now try to use the superior reach of their spears when fighting. Bandits, also randomly spawned, will wander around even when out of sight, and when you attack one of them, others in vicinity will come running. I was running through the Mutamin's Garden the other day with Protection from Petrification on and after one round of fruitlessly trying to petrify my character, the Basilisks and Greater Basilisks moved into melee, "realizing" that their gaze is not working.
None of this is vanilla behaviour. But that isn't the problem. These kinds of AI improvements are really more of a "common sense" change and welcome, at least in my opinion. As long as Beamdog sticks to AI improvements, and not for example giving every NPC a small fortune worth of potions(*coughSCScough*) I'm cool with it.
The problem here is that this creates a seriously uneven gameplay throuought the saga. You start in Baldur's Gate, where most AI is "dumb", except for some beforementioned creatures and EE NPC quest encounters. Then you move on to Siege of Dragonspear, where all encounters use the "SCS-lite" scripts. Once you get used to the new, "smart" AI in SoD... you are kicked back to "dumb" AI in BG2:EE, again - except for some specific creatures and EE NPC specific encounters. Then finally you move on the ToB, where AI isn't exactly smart, it just doesn't matter because everything is a Mage with Time Stop.
So the question is - what should be done about it? Should Beamdog improve the AI in BG:EE and BG2:EE across the board, up to SoD's level, for the sake of consistency?
Not only that, but if the creatures used in those encounters(or SoD) also spawn elsewhere, they will inherit this "smart" behaviour.
To those unsure what am I talking about, let me give a few examples. In the initial fight with Ekandor in Beregost, the Thay Mage will try to move away from your characters when he's not casting spells. The Tasloi, even ones randomly spawning in the game, will now try to use the superior reach of their spears when fighting. Bandits, also randomly spawned, will wander around even when out of sight, and when you attack one of them, others in vicinity will come running. I was running through the Mutamin's Garden the other day with Protection from Petrification on and after one round of fruitlessly trying to petrify my character, the Basilisks and Greater Basilisks moved into melee, "realizing" that their gaze is not working.
None of this is vanilla behaviour. But that isn't the problem. These kinds of AI improvements are really more of a "common sense" change and welcome, at least in my opinion. As long as Beamdog sticks to AI improvements, and not for example giving every NPC a small fortune worth of potions(*coughSCScough*) I'm cool with it.
The problem here is that this creates a seriously uneven gameplay throuought the saga. You start in Baldur's Gate, where most AI is "dumb", except for some beforementioned creatures and EE NPC quest encounters. Then you move on to Siege of Dragonspear, where all encounters use the "SCS-lite" scripts. Once you get used to the new, "smart" AI in SoD... you are kicked back to "dumb" AI in BG2:EE, again - except for some specific creatures and EE NPC specific encounters. Then finally you move on the ToB, where AI isn't exactly smart, it just doesn't matter because everything is a Mage with Time Stop.
So the question is - what should be done about it? Should Beamdog improve the AI in BG:EE and BG2:EE across the board, up to SoD's level, for the sake of consistency?
- Should Beamdog improve AI in BG:EE and BG2:EE up to SoD's level?74 votes
- Aye70.27%
- Nay21.62%
- Ultimately, we are all Neutral Evil8.11%
0
Comments
Not so sure I trust Beamdog as much.
GBASLSK.BCS:
IF See(NearestEnemyOf(Myself)) Delay(12) !StateCheck(NearestEnemyOf(Myself),STATE_STONE_DEATH) THEN RESPONSE #50 SelectWeaponAbility(SLOT_WEAPON0,0) // Gaze Attack AttackReevaluate(NearestEnemyOf(Myself),15) RESPONSE #20 SelectWeaponAbility(SLOT_WEAPON1,0) // Bite AttackReevaluate(NearestEnemyOf(Myself),15) RESPONSE #30 SelectWeaponAbility(SLOT_WEAPON2,0) // Poison AttackReevaluate(NearestEnemyOf(Myself),15) ENDHowever, I would totally understand that people wouldn't want to have that kind of AI in the EEs, because not everyone likes to play with SCS. I know that the SCS-lite version seen in SoD is actually what was dislike by some people, which is an entirely valid position.
Can someone from Beamdog please advice if this is something feasible? Like is there a chance that this might be implemented for the existing EE games in the future? Thanks
Whether it's going to be done or not, is above my pay grade to say, though.
I'm surprised that people have noticed improved AI so much in SOD.
Where specifically is it most noticeable?
If I was to give a quick discription of SOD combat, it would be "large mobs rushing towards you.....pretty much all the time".
Where is the improved AI?
I've even seen an SoD mage use Teleport Field to fortify a choke point. I never even considered that myself, and I'm a long-running fan of Teleport Field.
But the best example of the improvement in AI is the lich in SoD. In vanilla BG2, liches buffed themselves with Spell Trap and Improved Mantle, and they summoned Pit Fiends without casting Protection from Evil first--all of which were foolish decisions. Lich spell level immunities and their saves vs. spell of 1 already made Spell Trap largely obsolete (single-target spells and disablers were no threat anyway; they were only vulnerable to high-level area-effect damage spells); their immunity to normal weapons made Improved Mantle utterly inferior to Protection from Magical Weapons; and a lot of them ended up getting attacked by their own summoned Pit Fiend. At which point they would spend round after round wasting valuable spells trying to get past the Pit Fiend's massive magic resistance, rather than fight the party. And when they did fight the party, they'd spam Symbol spells even when the party was already disabled and a couple damage spells would easily win the fight for the lich.
The SoD lich doesn't make those kind of stupid decisions. He buffs with PFMW and spell protections to compensate for his weaker spell level immunities. He nukes the party with Horrid Wilting. He'll debuff Viconia with a triple Lower Resistance Spell Trigger and then slay her with Power Word: Kill. He'll pin down low-HP party members with Power Word: Stun. He'll blast you with Skull Trap when you're low on health. And if you run away from him and try to wait out his defenses, he'll summon Skeleton Warriors to chase you down.
I previously criticised people who were in my current position in the past, but after the latest patch I haven't been comfortable at all playing the vanilla EEs, which leaves me with two options: playing the original BG, or using and or developing mods to make the game like vanilla.
Also the potions component of SCS is completely optional. Just sayin'.
- Archers constantly running away to get safe position for shooting when you try hunting them (really annoying)
- Blackguards using their Absorb Life ability (which hurts for 64 damage on LoB!)
- Assassins who sneak behind your lines and use poison weapons to cut down the Mages
And the best (or worst): Once Enemies have seen you they will constantly follow and search even when you go invisible again. I love this. Others hate it.
About Mages @semiticgod already gave some decent informations. Prebuffing is great! And their spell selection is - like described - better even though far from optimal as they for instance still try to CC you even when protected from Magic (One exception are the Demons in Avernus: They will somehow recognize your MR and stop their CC abilities until the protection expires).
Call for help could be upgraded too but with SCS in the background it works. For instance: Without SCS you can fight the spiders in the Troll Wood Cave one group after another. With SCS it works much better as the nearby group(s) will join the party trying to cut down your retreat.
Anyway: BG:EE and BG2:EE with SoD scripts would be nice!
Improved A.I. was one of the top items on my wish list of enhancements for BG since they announced the EEs in 2012. By "improved A.I." I mean good A.I. that poses a challenge by playing it fair, like SCS does; I was never a fan of giving A.I. "illegal" benefits like unlimited spells, more powerful spells or items than are available to players, artificial damage boosts, etc..
My ideal was:
I agree with @Kilivitz in principle: New content should be tailored to the original content and made to fit as seamlessly as possible. However I do not think this applies to A.I.. Dumb A.I. was a known limitation of the originals, rather than a feature, and improving it imo qualifies as a worthwhile enhancement. It's the kind of change that is on par with improving graphics or sound; adding new content, on the other hand, is where I'd want maximum fidelity to the original games.
Extra request: add magic-users to BG1 "nemey parties", so shall we have a renewed experience,
like shamans for kobold and orc groups, wizards fr hobgoblins, mages for ogres, you catch my drift?
That being said, I wouldn't mind had they did at all. It's not no 1 priority for me, but that would be nice.
Me personally, I'm always for improvements; then again I also never play without SCS so these particular ones are unlikely to touch me.
2) one of the best things with SoD is that insane difficulty adds more enemies, and these enemies are cool. You don't get this in SCS, and I'd love to see it in BG1/BG2. Imagine Firkraag's wife or brother spawning next to him on insane difficulty, or entire Draug Fea's grade school buddies coming to help him! I think this would be a breathe of fresh air in BG2.
3) you can enjoy ad.2) in SoD w/o the silly double-damage enemies, which is great.
To be honest I do not like scs or improved AI mods for every playthrough. More often I like to play leisurely for the role play, the story and the progression. Only one in every ten games is with a mod that kills me at every corner. Having that every game will be a chore and will take the fun out of the game.
I love the IE games and would love it if they were still getting updated to be compatible with the latest technology another 19 years from now; I'd love to play through them with my 4-year-old son someday on whatever kind of technology people are playing games on when he's 15 or 16. And my only hope for that is for them to retain a vibrant and engaged player base.
The one universal truth is that over time some percentage of any given population is going to lose interest. If there are 100,000 players right now, and 10% of them lose interest every year, then in 15 years the player base will be down to 20,000. Maybe that's not enough interest to sustain new content.
And that 10%-loss-per-year is ridiculously conservative, and it ignores things like critical mass and tipping points. (Maybe only 10% abandon the game when there are 100,000 because there's so much fan-made content being made, but once that drops to 80,000 the amount of fan-made content decreases and the rate of abandonment increases, resulting in a death spiral.)
So given that simple math, the survival of the IE games depends on replenishing the player pool with new converts. I don't mean to be hyperbolic, but the future and long-term survival of these games we love literally depends on this. If we don't manage to bring in new players at a rate equal to or greater than the rate we're losing old players, the games will die. The greater the disparity between those two rates, the faster the games will die.
As a result, I consider things that ease the acquisition of new players as the single most important goal for the EEs. And I consider things that assist the conversion of new/casual players into die-hard fans as the second most important goal. I believe these are literally existential matters.
So yeah, I don't think Beamdog should be making changes designed to cater to the people who are already diehard fans. The biggest thing that Beamdog can give those diehards is the continued survival of their games, and the best way they can do that is by creating new diehards.
Everything else is secondary, and to be honest is already filled pretty well by the modding community, who naturally gears its decisions directly to the diehards instead of the new players. (That's a fair division: Beamdog makes things for newbies, modders make things for diehards, and the result is a vibrant game that continues to thrive.)
I think improving higher difficulty levels, such as by introducing SoD AI, could help in the process of converting new/casual players into diehards. But if it comes at the expense of acquiring new/casual players in the first place, it's a net negative for the long-term outlook of the games. So the ideal for me is "stupid AI" on Normal and Core difficulties, (with perhaps some very low-hanging fruit addressed), with "smart AI" on Hard and Insane difficulties.
I had hoped that LOB mode was going to be Beamdog's version of SCS but it wasn't. I did have a good time with it but the HP increase really hurts the game balance to the point where a thief can no longer backstab an enemy spellcaster into obvlion at the start of a fight and that is oh so satisfiying. I don't like that amror class means even less because of the boosted enemy thaco. I don't like that the enemy saving throws are so high that many spells are next to useless and some are useless because they are dependent on enemy hit die which is boosted through the roof. Tactics get narrowed down instead of expanded and some classes become less useful. SCS managed to deepen the casting strategy, ambush tactics, spell use, and throw in increased enemy groupings without breaking the game balance.
It's not a problem for me to keep installing SCS but it wasn't updated and we have had to use subtledoctors quick fix to get it working. It would sure be nice to have an option that is already installed with the game and fully supported for those of us that want it. Heck, they have Gaider on the team, might want to ask him if it's ok to include an optional Ascension ending in the game options as well. Waiting for the beta to come out was painful and many of us can hardly stand to play the game without it. I don't expect anything though, they are mods and I get that. That doesn't mean it wouldn't be freaken cool though. They could label them as mod options in the game menu, this would allow users to easily select them without having to find them and quick fix them and install them over and over again. It certainly isn't expected but it would be a great thing for fans of the game to have the options available for some of the major mods like that.
I also don't know why you're treating BG/EE games like some sort of MMORPG that needs to keep a playerbase up for revenue. Industry data shows that most of game sales happen in the first two weeks after release, for a refubrished classic title like EE the dynamic would probably be different, but certainly not constant over 5-10-15 years.
In short, everyone who wanted EE already bought it(and most of them were people who already owned BG, BG2 and ToB). All new revenue from that title will come from players who never played BG before, and that's an extremely niche market that will likely be only purchasing it heavily discounted/in a bundle anyway. So even if we assume that 100% of players new to the BG:EE will absolutely hate the AI changes(which is absurd in itself, not only because of the 100% ratio but because since they never played the games beforehand they will not even realize any changes were done), that would have a negligible effect on Beamdog's revenue which will mostly come from new games they make/enhance, not five year old titles.
Besides, most of the time they wouldn't be able to get a refund anyway. So it doesn't matter what do they think about the game after they already bought it, heheh.
Ascension is a bit different. While it still has options, most of those could fairly easily be brought to some default consensus without annoying too many people. Since the bosses are also not ALL that difficult without SCS, it shouldn't be that big of a problem in that respect either (and frankly ToB always was a bit too easy in the default state imo). HOWEVER the big problem here is that Ascension very likely is a direct violation of the contractual limitations put on content changes. While it does supposedly represent the "true" ending as it was intended by some people at some point, that still doesn't change the contractual obligations. Even a toggle option would probably not work to satisfy those obligations, which leaves us with modding as the only alternative.
Most importantly, it would allow them to essentially make BG3 without having to make a game that takes place after TOB. TOB ends it so I think the only reasonable thing to do is to expand it until it is BG3. One of the great things about Baldur's Gate is the continuity between the games. I would hate to see BG3 take place after TOB or with a new set of characters and a new story. It might as well be called something other than Baldur's Gate at that point. Especially if it uses a new rule system and graphic engine like a remake would have.
As for difficulty, I am suggesting that they make a new difficulty like they did with LOB mode in SOD but make it more in line with SCS. Then, instead of us discussing if they should improve the rest of the game's AI to SOD standard, we would be begging them to improve the rest of the game's AI to BG3 standards.