DavidW

well, an example serves better in the context of defining cheats/abusing in the game than quoting the dictionary... Not really. The dictionary definition doesn't especially help, since this is a technical context. I already know that you think II+SI:Div is abusive, so requoting that example doesn't help either. What I want to know is what you actually mean by that term. (My experience is that 90% of the time it doesn't mean anything very coherent beyond "I dislike this", but feel free to prove that you're in the 10%.)

@Demi: Let's have a go at rethinking this from scratch (within SCS's parameters). I've been running through the problem this evening, and I can't myself improve on my original set of solutions, but this may be my own lack of imagination. Here's the basic problem. In BG2, Protection from Weapons spells (and, at lower levels, Stoneskin and MI) aren't supplements to hit points and AC: they're replacements for it. No mage can survive for any relevant period of time without them. Call these anti-weapon spells. In vanilla rules, Breach takes down anti-weapon spells, so one survives only until Breach is successfully cast. The version of Breach that's cast directly penetrates Spell Turning, so only invisibility is a shield against it. The version of Breach that's loaded into wands of spell striking can't penetrate Spell Deflection et al, so Spell Deflection and invisibility are both shields against it. SCS allows Spell Deflection to block Breach, even when cast directly. I'm very reluctant to lose that; anyway, for the sake of argument, let's assume it. (If you want to argue me out of it, go for it; I suspect you're in favour, though.) With that change made, taking down a wizard is a four stage process: (1) remove his Improved Invisibility, which stops you using Ruby Ray et al to lower his anti-spell defences (2) remove his anti-spell defences with single-target antimagic (3) remove his anti-weapon defences with Breach (4) cut him to pieces. This process can in principle be shortcutted by Dispel Magic. Pre-Taimon, this was basically ineffectual. These days, it's better, but still hit-and-miss at best. Now, here's the dilemma. A) If SI:Div is allowed, then in the vanilla rules there is no way at all to take down Improved Invisibility (short of the hit-and-miss strategy of using Dispel Magic). So the whole process can't get started and you're stuck at step 1. (That's the Tactics/IA situation). B) if SI:Div is not allowed, then any sane party has Truesight running as part of their pre-combat buffs (and even if they didn't buff, the cleric can throw it up fairly quickly). So step 1 happens almost automatically. Only steps 2-4 remain. In (A), things are annoying and boring. I don't like the Tactics/IA situation one bit. In (B), mages go down too quickly. (You can see this in SCS by looking at how quickly clerics go down: I'm just unable to protect them.) My ideal, unimplementable solution is for single-target antimagic spells to work even against invisible targets. In that case, you just hit mages with antimagic, in the presence of Truesight, till they're targettable, and then kill them. It takes long enough to cut through a wizard's defences that things are fairly even (I predict). In this situation, SI:Abj and SI:Div don't seem to me unbalanced: they're just one more step in the defence process. (I'm leaving out the legitimate irritation that SI:Abj doesn't seem to do what it says on the tin.) As I say, this is unimplementable. The nearest I can get to an implementation is the small area of effect used in SCS. That's imperfect, and you've noted above you don't like it; it's also the best I can think of to resolve the dilemma. At the moment, SR doesn't seem to help. Non-detection is penetrated by Truesight (as I understand it) so we're back in situation (B). If SI:Div is allowed, but area-effect antimagic isn't, we're instead in situation (A). Thoughts welcomed. I'm genuinely amenable to restructuring SCS's antimagic framework if there's a genuinely better (and not-much-more-disruptive-to-vanilla) solution out there. Area effects for antimagic is my least-worst solution, not my ideal one.

low cost, high effect. like SI:I On that basis, a high-level caster casting Remove Magic is cheap. the trick where you can make the project image immune to divinations is very abusive. That's an example, not a definition.

SR4 can introduce neat implementation of Spell Shield and PnP Non-Detection If you read up one line from that quote, you'll see that I was specifically discussing the situation relative to vanilla. That's a slightly odd comment. The point of the spell is to protect you from energy drain, so yes: the spell would be pretty useless if nothing attacked you with energy drain. Similarly, Truesight would be pretty pointless if no-one ever used illusions against you. Can you define "cheap"? In either SR or SCS, you can counter it with an anti-magic attack. Can you define "abusing"? Is learning SI close to automatic for you when you play? It's pretty low down my priority list for 5th level spells, personally speaking (and it rarely seems to get onto people's lists of recommended sorcerer spells).

I think the real difference (although it overlaps your "concept vs implementation" distinction) is that in SR, your thinking is basically "what would the best spell system look like if designed from scratch?", whereis in SCS I'm considering "what aspects of the existing spell system are sufficiently problematic that they need to be changed?". In addition, you're asking: what does SI look like when we allow for all the other changes made to the spell system in SR? I'm asking: what does SI look like relative to the vanilla spell system? If I were to ask your questions, rather than mine, then probably I wouldn't include Spell Immunity in my designed-from-scratch spell system. It's a cute idea, but on balance I think the costs outweigh the advantages. But redesigning the spell system from scratch isn't what SCS is about. (And, in this particular case, supporting multiple options isn't really viable as the buff/antibuff aspect of SCS's smarter mages is too core for me to be willing to support multiple versions.) So for me (and therefore, unfortunately, for anyone else who wants to change the mage defence system while maintaining SCS compatibility) the question has to be mine: is SI sufficiently problematic (either at all, or as a 5th level spell) that it really needs to be changed? The bar is relatively high here: comparable changes in SCS are allowing Breach to be blocked by Spell Turning, giving areas of effect to antimagic, and increasing the strength of Mantle - though the latter was a borderline choice for me. So, looking at your case-by-case analysis (and restricting attention to the vanilla situation, which has to be the point of comparison for SCS): - I agree that SI:Trans, SI:Conj, SI:Ill are unattractive. That doesn't especially bother me from an SCS perspective: they can stay as part of the overall flavour of the SI spell. - We're both happy with SI: Nec. - You think SI: Div is underpowered, which surprises me. It's a very powerful protective spell for a mage: II + SI:Div is the key to staying unBreached. I suppose I wouldn't go to the wall to stop it being 4th level, but +/- 1 level doesn't hugely bother me from an SCS perspective. (And of course if it's part of SI, then the overall spell might justifiably be 5th level even though this component would be 4th level as a standalone.) - You think SI: Ench is worth a 6th level slot. I'm not sure why (at least within vanilla; I concede that if you throw in protection from power words, things might change, but that's not salient for SCS). Chaotic Commands does everything SI: Ench does, has a far longer duration, keeps off psionic attacks that aren't in the enchantment school, and can be cast on others. But in any case, even if I were convinced it ideally should be 6th level, from an SCS point of view that's not a big enough change to justify shifting the status quo. - You say that "any 5th level Protection from Fire/Cold/Lightning/Acid spell simply pales compared to SI:Evo". As I recall, Protection from Fire/Cold is 3rd level in vanilla, but in any case, which would you rather have available when facing Firkraag? (And don't forget that protection from Fire can be cast on other party members.) Again, I think you're underestimating the benefits of long duration, castability on others, and protection from non-spell attack forms. And again, even if I were convinced that ideally SI:Evo should be 6th level rather than 5th, that's not enough of a problem to justify changing it in SCS. (I put my money where my mouth is here, incidentally: SI:Evo turns up pretty rarely in my prebuff routines.) - Conceptually speaking I've some sympathy with your annoyance at SI:Abj... but not much - I don't see a problem with the idea that some spells can bypass Spell Immunity, and if those happen to be mostly abjurations, so be it. Implementation-wise, immunity to dispel magic seems to be roughly worth a 5th level slot. Regarding your more general comments: (1) I agree that the technical constraints are annoying. Ultimately they don't bother me hugely, though: by all means let's suppose that sorcerors don't get access to these particuar spells. (There are no sorcerors in SCS, so it's not like I'm breaking that rule elsewhere.) (2) I think it's only true that all SI spells have equivalent spells if SR is installed. Relative to vanilla: - nothing can replace SI: Abj or SI: Div, so far as I can see - SI: Ench and SI: Nec are roughly but not entirely equivalent to certain priest spells, but don't have a mage equivalent - SI: Alt, SI: Ill and SI: Conj are fairly ineffectual, so unsurprisingly don't have equivalents - the nearest equivalent to SI:Evo is the (imo much more powerful, and so much higher level) Pro/Elem. (3) As a side point, do I take it that stacking SI: per se, doesn't bother you? (I'm assuming that because if you're replacing SI's various components with similar spells at different levels, presumably they will be stackable.) This is an interesting discussion, by the way. I'm deeply bored of people (not you) saying "SI is overpowered" or "it's cheesy to stack SI" or the like without bothering to give arguments. I'm mostly not persuaded by your arguments, but it's nice to actually get some!

I don't think this is a problem. Instead of spell protection (SI) there'll be specific protections (MB, ProEnergy, etc.) - AI won't detect vanilla SI, but it will notice the Chaotic Commands like effect.Of course, it requires DS to check for SR and execute an additional set of patches for it's changes, but I see nothing wrong with that. Nothing technical. But I'm not really keen on having the SCS version of DS keep track of every spell modification made by third-party mods, so I think if SR is making these changes to protection spells, it would make sense for it to ship with its own chunk of DS. The DS code is robust against being installed multiple times. EDIT: come to think of it, a cleaner solution is just for SR to supply its own version of the control table for DS, and dump it in (e.g.) the override. I'm happy to get SCS to look for it, and then use the SR version of the control table if it exists.

Leaving aside my points about Mind Blank's greater power, which you accept, I'm not much moved by pure PnP concerns. The BG2 spellcasting context is quite different (and in any case, 2nd edition spell level assignments are hardly immune from criticism). I'm also not sure what the significance is of your comment about "the vast majority of players" (even if it's correct, which I'm sceptical about - people who are happy with things aren't likely to comment). The salient issue, presumably, is the quality of the arguments, not the number of supporters. You do have a point here. The problem is that you have two caps to how many spells you can learn per level, an in-game roleplaying one (depending on INT, but having 4-5 SI in the spellbook isn't great even if you have INT 18) and a technical one (sorcerers can only learn spells from that 24 spells per level limited list, you can't have 8 SI in such list). Granted; in that case, I retract my support for getting rid of the single-use version. I won't quote the discussion of how SI is more useful to enemy wizards than player ones, because I don't particularly disagree. But if we're discussing whether it's overpowered for enemy wizards, then the fact that it's useable in eight different ways becomes basically irrelevant, because no enemy AI I know takes advantage of that flexibility. In this situation, it comes down to a direct comparison, and that comparison seems to work out okay to me. In more detail: grant, for the sake of argument, that Mind Blank should indeed be 8th level. It provides immunity to all mental attack forms, including but not limited to enchantments, and it has a long duration, and iirc, it's castable on other creatures, not just on the caster. In comparison, SI:Ench is caster-only, has a short duration, and protects only from enchantment-school mental attack forms. A three-level difference doesn't seem crazy in that circumstance. Certainly one level difference seems too little. (I wouldn't go overboard for 5th level versus 6th, but overall there's a game-interest case for spreading out spells a bit, and 6th is usually pretty oversubscribed... and in any case, in SCS I nearly always use SI via sequencers and contingencies, so 5th vs 6th is irrelevant) Similar arguments apply for Pro/Elements vs SI:Evoc. Now, granted: enemy wizards are usually less affected by these restrictions than PCs: they don't tend to get attacked by mind flayers, and they tend not to be too disadvantaged by duration (though I don't use SI as a long-term buff precisely because it's got a short duration, so the duration is only irrelevant for players who choose prebuff option 1). But that's not a reason to ignore the restrictions when assigning a level to the spell, any more than the fact that enemy wizards find Fireball very hard to use compared to the player is a reason to lower its level.

Hang on, something's wrong there. Mind Blank isn't in vanilla BG2, so the fact that you give it a level higher than SI isn't independent of your antecedent belief that SI is overpowered. Also, Mind Blank protects (I take it) from mental attacks that aren't included in the Enchantment school (notably Illithid attacks), and (correct me if I'm wrong) has a longer duration; similarly, (vanilla) ProEnergy protects against plenty of attacks not in the Evocation school (e.g. Flame Arrow, dragon breath) and again (iirc) has a longer duration. Finally, I'm very unbothered by the apparent benefit of being able to get 8 different versions of SI from one memorised spell because in practice in BG2 it's highly unlikely that you won't be reasonably clear on which one you need in advance; having said which, it would bother neither SCS nor me personally to remove the single SI entirely in favour of the eight specific versions. Put it another way. Is SI really the first spell every PC wizard memorises as soon as they get 5th level spell slots? Or even the 3rd? It's not when I play. SI:Abj and SI:Div are occasionally useful for PC wizards as part of an overall buffing pattern, but SI:Ench and SI:Evoc are rarely worth the slot and the time taken to cast them, given that any given wizard's attacks can probably be drawn from a variety of schools and given that you're only protecting one character.

Didin't we discussed it and reached a solution? If I'm not wrong SCS needs almost only SI:Abj and SI:Div, and thus "secretly" replacing spwi590 and spwi592 with Spell Shield and Non-detection respectively should be fine. If we add Mind Blank I could also replace spwi593 with it (I know we would have a mage cast an 8th lvl spell with a 5th lvl slot, but it's a minor issue imo). What do you think? Oh, that rings a bell. I use SI: Evocation occasionally too, though. I don't think I use any of the others, though I'm at work and can't readily check. On a different note, though, is this going to mess up Detectable-Spells-based targetting in SCS? After all, I check for SI:[whatever] when I target a spelll, but obviously I don't check for Mind Shield. (Indeed, there's a more general issue of how you handle detectability of new protection spells in SR - what do you do atm?)

Not for the first time, can I point out that removing Spell Immunity will break SCS compatibility.

SCSII uses it occasionally (1/4 of enchanters' 8th level combat slots). Variety is the spice of life.

Just to clarify: this is a list of what mods I expect to be compatible with the mod. It's not a list of mods that I'm happy being tested with the mod. For testing, I'd rather you don't have any mod installed at all (unless you're Kulyok and testing IWDNPC, or something similar). Probably minor ease-of-use mods are okay, but nothing more dramatic, please. The relevant parts of 1PP are now included in the core mod, so you don't need to (and indeed shouldn't) install 1PP. EDIT: I can live with Widescreen too, since it seems to be working okay.

The quest XP comment was facetious, to be honest - I don't actually think it does make sense to use quest XP to allow for spell selection. This is largely academic as far as SCS is concerned, for broadly the reason amanasleep gives: SCS isn't really in the business of changing XP levels, period. But someone suggested it was my call, so I thought I might as well answer. I continue to be unpersuaded that it is more effective for enemy NPCs to use summons than to use equivalent attack spells.

I'd be interested to hear the evidence that enemy spell selection is modelled through quest XP. Are you seriously proposing I should model this in SCS? (If so, it's pretty easy, in principle: a wizard whose spell selection includes a spell that summons a creature worth N experience points also yields a minus N experience point quest xp reward!) More generally, I think people are treating XP as if it's some kind of mystic energy, rather than an abstraction that represents a reward for the difficulty of the combat. Any such abstraction, applied along simple rules, is going to have occasional pathological cases that don't work properly - the monster-summoning wizard who teleports away is a case in point. But in general, not giving separate XP for summoned monsters seems to produce far fewer problematic cases than giving it.

Is it up to me? I didn't think I changed XP levels for summoned creatures? If it is up to me, expect no XP. I haven't seen a remotely convincing argument to the contrary on this thread. In particular, the double-counting argument (XP for killing summoned monsters is included implicitly in the XP for killing the summoner) doesn't seem to have been addressed. The only exception I can see is Aranthys's observation that summoning spells are substantially more powerful than other spells of the same level. I'm not convinced that's the case (whether I'd rather give an SCS enemy wizard Horrid Wilting or Summon Fiend is pretty situation-dependent) but if I were to be convinced, it's a case for altering the spell system, not the XP system.

Commenting on the narrow point of XP for summoned monsters: The argument against, as I understood it, was that killing summoned monsters was part of what was involved in killing the summoner, and so shouldn't give experience over and above that. Scenario 1: the Evil Wizard casts Horrid Wilting. You survive and kill him. Scenario 2: the Evil Wizard summons a Glabrezu. You defeat it, and then kill him. Shouldn't you get the same experience in both cases?

Some notes about converting an IWD mod to IWD-in-BG2 (mostly aimed at Kulyok for the moment). Most of it is reasonably obvious. Creature files need to be redone completely, since the IWD and BG2 versions are differently structured. (I have code that does this automatically, but unless you have dozens of creatures it's probably quicker just to do it manually. Custom spells and items may or may not be okay, depending on whether you've used IWD-specific resources, projectiles and opcodes. If not, you're fine. If so, you'll need to convert them. Again, I have code; again, it'll be easier to do it manually unless there are horrendous numbers of them. (See below if you're using IWD WAV or BAM files.) Stores need converting, but here the conversion really is simple, and might as well be done by code (which I can provide if desired). IWD resources are mostly still available under their IWD names. In particular, items, scripts, dialogs, creatures, areas have the same names as in IWD, so you can refer to them in the original way. Death variables are also unchanged: any death variable that picks out a particular actor in IWD will pick out the same actor in IWD-in-BG2. A small number of IWD spells aren't available in IWD-in-BG2 at all (Wall of Moonlight, Smashing Wave, Seven Eyes, Soul Eater, Mordenkainen's Force Missiles, (Great) Shout, Spiritual Wrath, Whirlwind), so if you've used them (or magic items that use them) you'll need to change them. More importantly, any IWD spell which isn't also in BG2 is likely to have had its resource name (SPPR123 or whatever) changed. This is because I want to avoid overwriting BG2 spells that aren't in IWD, as I think quite a few people will want them available and it's easier to hide unwanted spells than reintroduce overwritten ones. I suspect the simplest way to handle this is to write some WEIDU code to read in SPELL.IDS so that you have some array $spell_res which associates a spell resource name to each spell descriptor (so that $spell_res(~WIZARD_CATS_GRACE~) gets assigned sprwi214, say). Then add any spells you need at install time via ADD_SPELL, using $spell_res to set the filename of the spell to add. Alternatively, the converter outputs a file that tells you the new filename of each old filename, so you could just read that in and do a REPLACE_TEXTUALLY. (I can probably provide code for anyone whose WEIDU isn't up to this.) (When the converter is stable, it would probably be simpler to just use the BGT/TUTU dual coding trick. But that relies on the new names of the spells being stable, and they won't be for a little while - basically until I manage to convert and/or permanently give up on the unconverted spells.) (Scroll files still exist under their old names.) Scripts and dialogs should basically be fine provided that your IWD mod doesn't use any scripting that doesn't work in BG2. Variables are also unchanged, with one crucial exception: any variable of the form SOMETHING_DEAD now has the form SPRITE_IS_DEADSOMETHING. This is basically forced by hardcoded differences between the variables that IWD and BG2 set when an actor is killed. Given that many variables of this form need renaming, it's much safer and more bug-resistant to rename all of them (not doing this led to a significant fraction of the death-variable-related bugs reported in v4). The simplest way (I think) to handle this in your code, if you want it to work both in IWD and in BG2, is (i) get your code to check if it's installing on IWD or IWD-in-BG2 (e.g. by detecting BGMAIN). If it's on IWD, do OUTER_SPRINT ~SPRITE_IS_DEAD~ ~~ OUTER_SPRINT ~_DEAD~ ~_DEAD~ If it's on IWD-in-BG2, do OUTER_SPRINT ~SPRITE_IS_DEAD~ ~SPRITE_IS_DEAD~ OUTER_SPRINT ~_DEAD~ ~~ Then, whenever you want to refer to one of these variables (say ORC_DEAD), just put it into your code as ~%SPRITE_IS_DEAD%ORC%_DEAD%~ and compile using EVALUATE_BUFFER. That will compile to the correct value, whatever version you're playing. (Note that this only applies to things of the form XYZ_DEAD, not XYZDEAD. TALONITESDEAD, for instance, is unchanged.)

Here's a list of current issues that I can't (or can't easily) solve. It's probably incomplete - I'm working from memory here. I'll add new ones as I find them. The more significant an issue is, the harder I've tried (and failed) to solve it. GAMEPLAY Critical: Randomly, very infrequently, irreproducibly, the game crashes to desktop when you look at the map screen. I have no idea why this is or how to fix it.(CamDawg's rebuild of the map files ALMOST fixes this, but once in a long while it still seems to happen) I've got occasional and irreproducible critical bugs with the cutscene in the Yeti cave in the Vale of Shadows: occasionally, but not often, the game hangs here. (I think various scripting tidy-ups have fixed this.) Significant: I don't know how to stop the game ending when Player 1 dies. (I'm unsure whether this really matters or not) (Fixed now, thanks to ToBEx and Ascension64) I can't convert thieves' Evasion or Sneak Attack abilities (Evasion now done, thanks to Nythrun) Minor: I can't block resting in Kuldahar and Easthaven directly. (I fake it, but it involves new dialogue) In the Vale of Shadows, I think my converter handles spawning correctly according to the INI file that's supposed to control it, but that behaviour doesn't correspond to how spawning in that area actually works in-game. I think there is a hard-coded hack in the IWD engine which overrides the INI file. I haven't done the research to reproduce this, so spawning is somewhat more common in the Vale in IWD-in-BG2 than in IWD proper. (This isn't the same as the ridiculous overspawning noticed by players of v4: that was a bug in the converter, now fixed.) (After playtesting this a bit, I've just given up and blocked spawning in the Vale of Shadows. I have a feeling the playtesters of the original IWD may have had a similar experience) The "default" soundset comes with no biography. I have no idea why. SPELLS AND ITEMS Significant: I can't convert these spells: Seven Eyes, Soul Eater, (Great) Shout, Mordenkainen's Force Missiles, Spiritual Wrath, Mold Touch, Wall of Moonlight, Smashing Wave, Whirlwind.(Thanks to TheBigg, I've now got a (not quite perfect) conversion of Whirlwind) I can't convert these items: Three White Doves; those items that cast the effects listed above(Three White Doves now converted, thanks to CamDawg) My conversions of Cat's Grace, Power Word Stun/Kill, Blood Rage, Animal Rage, Chain Lightning, Static Charge are all a bit suboptimal (Thanks to ToBEx and Ascension64, Cat's Grace now works; actually, so do Kossuth's Blood, Kontik's Ring, Kossuth's Ring, the Ring of Edion, and Strength, all of which were a bit suboptimal but which I forgot to list) COSMETIC Significant: I can't get the white dragon animation to work in IWD: we're making do with the BG2 one (I could use Erephine's version, but at 45MB it's twice the size of the rest of the mod put together, and I can't straightforwardly automate its generation) The spell animations in IWD often have a sort of glow around them which I find a bit unattractive and which doesn't appear in IWD. I think this is because of some basic engine-level difference, but it might just be down to a VVC setting which I don't know. Minor: Something I don't quite understand about the GUI prevents the selected Inn in the room-for-the-night screen from getting the red border it ought to have. Because the BG2 move-to-expansion command is not reversible but the IWD one is, I can't give IWD and HOW separate maps, so they're combined into one map. The Kit selection screen in the character generator creates an unattractive shadow (Fixed now) Some of the selection boxes on the character generator could do with being bigger (Fixed now) The Bhaalspawn death movie plays at game-over (Fixed, thanks again to Nythrun) The maps are only 3/4 the ideal size (IWD maps are larger, so that the biggest IWD areas fill the whole map screen). Summoned monsters don't appear as quickly as I'd like (there's a puff of smoke, then a two-second pause, then they're there.) I could fix this if I knew the BAM file for the "golden bubble" animation that plays when monsters are summoned, but I can't find it. (note to self: apparently it's SPPLANAR ... no, apparently not, that's the more substantial animation you get from MS spells) For some reason, the floor in certain areas (wooden floors, mostly) hides the selection circle of characters standing over it.(Fixed by CamDawg) The rest icon is in the wrong place, and doesn't look quite as nice as it looks when it's in the right place. If you hide the left-hand display bar, there's no button to bring it back (the quick-key still works) Save-games in Heart of Winter get labelled as "-ToB". Selecting a fist (i.e. unarmed attack) seems to select all fists (i.e. put a green box around all empty weapon quick-slots). As far as I can see this is purely cosmetic; in any case, I've no idea why it's happening. IWD loadscreens show a picture of the area you're in. I can't manage to reproduce this. I can't convert the "spells" (i.e. graphical effects) INNATE_CONTAINER_GLOW, INNATE_CONTAINER_GLOW_BAD, and INNATE_BLUE_GLOW. These all make things glow in pretty ways, but I'd have to reengineer them from scratch. If anyone has the patience to do this, please send me the copy. The cursor disappears after you return to the credits page having completed the game. I have no idea why.

Do you randomize the components of the Equalizer et al? I ask because the SoA assemblable items actually have hints in their descriptions as to where they're to be found. So if you do, possibly that could be made optional for pedants like me?

No, if MI and RI are incompatible normally then I wouldn't put them in a sequencer in SCS, any more than I'd put multiple Project Images in a Chain Contingency. It's clearly an exploit. (Not that I care about blocking it. If players want to exploit the system, why should I mind?)

Commenting from an enemy-AI perspective: anything that gets you half as many hits as you'd expect over any significant period is worth having, if you can find the time to cast it. (There's a delicate tradeoff in wizard scripting: how much time to spend renewing buffs, vs how much time to spend doing damage, and given the need to keep up Mirror Images, Stoneskins and Pro/Weapons spells, as well as anti-spell defences, there might not be room for another short-duration defence spells in wizard scripts.) I don't think I'd be able to find the time to use RI in combat after about eighth level, at a guess, though it might be nice in a Minor Sequencer with Mirror Image. Incidentally, if prebuffing immediately before a battle, even a seven or eight round duration is worthwhile. My impression is that most BG2 battles are over, or at least under control, by about five or six rounds in (exception: wizard fights, where you have to wait out defences).

I'm not entirely clear what the new version does. But a 4 round duration probably makes it fairly useless at high level. A fighter/mage already needs to pause every 4th round to renew PMW or the like, not to mention stoneskin. They probably don't have time to keep a second protective spell up.

Indeed I know this (it's not in a hidden manual but the base 2nd ed MM), and that's another reason I never nerfed them, but like aVENGER and David I also think that being 100% true to PnP isn't always the best choice. Furthermore, if you really want to be 100% true then you should also know that these creatures cannot be controlled without a circlet of power, that a caster wouldn't be able to control more than one SW at once, and that PnP Animate Dead spell doesn't allow to summon any powerful undead, only lesser skeletons and zombies. Just one thing to remember about skeleton warriors: they can be turned. I don't think the vanilla AI ever tries to turn undead, but SCS(II) does. That presumably reduces their attractiveness versus non-undead creatures with the same stats.

It's not the way I do it (and it's fairly obvious from the CRE files that it's not the way the developers do it). And (unlike 3.0 or 3.5) the AD&D leveling system really isn't suited for this sort of method of assessing creature strength (and there is no direct map from creature HD, THAC0 etc onto character level, THAC0 etc.) Warriors only(Fighters, Barbarians, Paladins, Rangers+their Kits). And yes, the Barbarian is a Fighter Kit, but so what. Sorry, I meant warriors.

It might help to explain, for those not familiar with how CRE files work, that the "class" and "level" parameters in the creature file control very few of the creature's actual attributes. Simply designating a kobold, say, as a 20th level fighter/cleric, has no effect whatsoever on its spells, THAC0, hit points, saving throws, or proficiencies. The "class" and "level" parameters control only a very small number of hardcoded features, mainly: - the level at which spells are cast, if they are cast - the ability to turn undead (which, however, won't be used unless the creature was already scripted to) - the level at which certain spells (e.g. Sleep) affect the creature - whether additional attacks per round are gained at high levels (fighters only) It would in many ways be more logical if the creature's base statistics were all determined by its level and the rest of the CRE file just modified, rather than setting, those base statistics - but the Infinity Engine doesn't work that way.