DavidW

I doubt that's a bug exactly, more likely they're using bespoke scripts written before I added cloud-avoidance. (It wasn't part of original SCS I, and I haven't really touched Sarevok et al for ages.

Yes, I see it. The variable that tells SSL that Refinements is installed is set too early and is being overwritten. Fixed locally.

The mod wasn't in very great condition by then... it might have influenced DavidW to remove the spells. It didn't. (If it had, it would have made it into the readme.) If they're not showing up, it's a bug. (Not implausibly something to do with the updating-Refinements-to-EE issues though.)

That sort of unfairness is definitely against SCS's design criteria. I agree that precast Stoneskin is essential for those casters, but they'll need to pay for it honestly with a spell slot.

I think you're right. TBH I ought to go over that code in general, I haven't looked at it systematically since about v15.

Not sure yet; I'll have a look at spell memorisations. (I'm reviewing the vanilla ones for v31 in any case.)

SR makes Stoenskin 8 hours, nobody complained about not really being in the spellbook.... Except me! (Or rather, I didn't complain, as I didn't know that, but I'm complaining now )

Replacing SI:Conjuration with SI:Abjuration in fighter-mage buffing (that routine already sometimes gives them SI:Abjuration) Mind Blank is supposed to be a very-long term buff (has the same duration as stoneskins, for example). Can u replace SI:Con with some other spell then (just not another SI)? SI:Con: ok, can do. Mind Blank: It's clearly a spell that should be a long-term buff *if* it's memorised. But it's not currently being learned so far as I can see. (I cast Stoneskin even when it's not learned, because it has a 24-hour duration so it's plausibly cast before the last time the mage slept. That isn't possible for Mind Blank, which has an eight-hour duration.) It sounds like I should be including it in the defensive spells roster?

But don't bother emailing the file - as noted in the pinned topic on the forum, IWD-in-BG2 is no longer in development now that IWDEE is out.

It's choking when it gets a level that's set to (or below) 0. Fixed locally. That's a general feature of level override; it's not very sensitive to multiclasses. Again, fixed locally. Lich or demilich? Yes, that's intentional. (It's roughly what's implied by her memorised spells, iirc). Cleric/mages aren't caught by the level_scale ini setting because they're done individually (so there's no code to work out what spells they should have if they have their level changed); that's not likely to change. Yeah, they shouldn't be on the general help loop in most cases. Will check.

I'm hoping to release v31 of SCS in the next few weeks, and so I've been going through these scripting modifications. Many are straightforward fixes to errors in SCS v30, and others are clearly sensible given SR. But I have a residual list where I'm not sure about the rationale and haven't implemented on my local v31. I thought I'd list them here to see if Kreso or others can advise! (I'm unfamiliar enough with SR that it's probable I'm often missing something.) I'm not: ignoring Spell Turning/Deflection, since I think it will lead to lots of targeting of protected creatures when unprotected ones are available (though elsewhere in v31 I do pay some attention to the need to force them sometimes) avoiding checking for Slay before casting FoD telling Anadramatis to cast Pierce Magic instead of Pierce Shield (he doesn't have it learned) letting Imix be subtle about spell turning. (Not everyone is perfect) worrying about SR school changes when the relevant specialist doesn't actually use the spell anyway dropping the saving throw check for the Wand of Paralysis dropping the pro/magic weapons check for touch attacks dropping the Mirror Image check for touch attacks and the Wand of Frost turning off the Stun check for Symbol of Stun casting monster summoning as a long-term buff (that breaks the rules for what counts as a longterm buff) Using Mind Blank or Protection from the Elements, since I don't think anyone has it memorised Replacing SI:Conjuration with SI:Abjuration in fighter-mage buffing (that routine already sometimes gives them SI:Abjuration) Memorising Summon Djinni, since we don't have scripting to cast it Giving Conjurers Wail of the Banshee, since it gets in the way of the thematic specialisations Giving necromancers Bigby's crushing hand rather than PW: kill (ditto)

That's a relief; thanks.

I officially support: - Baldur's Gate: Enhanced Edition - Baldur's Gate II: Enhanced Edition - Baldur's Gate: Siege of Dragonspear - Icewind Dale: Enhanced Edition (though this is experimental) - Enhanced Edition Trilogy (though a lot of my testing assumes SCS is installed after EET_End) - Original Baldur's Gate II, with Throne of Bhaal and the BG2 Fixpack mod - EasyTutu, including Tales of the Sworde Coast - Baldur's Gate Trilogy, with the BG2 Fixpack mod I do an install-test of the mod on these whenever I do a major update. My primary target these days is the Enhanced Editions but I don't anticipate stopping support for the older versions as long as they're widely used. I don't do install tests on vanilla BG1 TUTU (including Tales of the Sword Coast, plus TUTUfix), if only to keep my workload under control, but I'll probably be helpful fixing problems with it if they're not too difficult. I no longer support BGT or vanilla BG2 without the Fixpack (SCS will refuse to install unless it is present). I don't, and never will, support original BG, or vanilla BG2 without Throne of Bhaal, or (Easy)Tutu without ToTSC, or vanilla IWD.

Post them here.

You might be surprised at how easy that would be (unless I'm missing something). The BG2EE and IWDEE engines are much closer together than any two engines of vanilla IE games.

As a point of interest I avoid using Death Spell against PCs in SCS precisely because it feels so unfair.

The last released version of IWD-in-BG2 was v8, released in Fall 2013. In Winter 2013-14 I supplied a modified v9, designed for the Enhanced Edition engine and written to allow for engine modifications, to Beamdog. That modified v9 formed the core of IWDEE, but IWDEE was able to go a long way beyond what was possible in IWD-in-BG2 v8 because its development team (including CamDawg, the other main author for IWD-in-BG2) had access to the source code of the engine. To me (and I believe to CamDawg too) IWDEE represents a successful completion of the project we began in IWD-in-BG2, and with its release yesterday I no longer intend to develop IWD-in-BG2. (I apologise for not making this clearer in the couple of months since IWDEE was announced; I took it as obvious but clearly shouldn't have done.) There is some more discussion of the issue at this thread: https://www.gibberlings3.net/forums/topic/26767-is-a-newer-version-than-v8-in-progress/

It's totally appropriate, not a problem at all. And I'll post the sourcecode when I get a chance. (But at least *my* modding style requires that I basically keep track myself of any modifications to the source code, otherwise I just get confused.) On patches rather than sourcecode changes for intermediate releases, this is mostly because compiling the sourcecode to produce the distributed mod is a complete nightmare (largely due to poor design choices in my original code way back when). It's not realistic for me to do that, and to upload the 20MB result, every time something gets changed.

Thanks. This is (almost certainly) a one-line typo in the script-conversion routine. Will fix for v9.

Don't bother, tbh. The v9 sourcecode is likely to be fairly substantially rewritten anyway, and our pattern for v8.x, like v7.x, is probably to produce after-installation patches instead of changing the underlying sourcecode. But I'd be keen to know what the actual problem is.

It has it but it doesn't use it, at least where SCS is concerned. The relevant SCS targeting block is OR(2) HPPercentLT(scstarget,75) !CheckStatGT(scstarget,12,Level) OR(3) HPPercentLT(scstarget,50) Class(scstarget,MAGE) Class(scstarget,MAGE_THIEF) The player only has access to the Barely Injured/Injured/Badly Injured/Near Death categories, so in SCS, so does the AI. (This is ancient code and in hindsight I might redesign it slightly, but I'd keep to the general principle.)

I'm relatively sparing about immunity to Imprisonment - but I think dragons get it as part of "standardise dragon immunities". For AI use, I had a feeling there are easier ways to block Maze than imprisonment - but perhaps I misremember? (One of the Easter Eggs in SCS is an area where a spellcasting opponent has Imprisoned a group of balors as a counter to Freedom cast by the party...)

Hmm, hihihi, didn't see that, wish I had actually had ... no you can't hurt me. And I think my idea has merit. I think I just said "for the 34th time" or something. I've made this point lots of times before - but that's inevitable given the medium, and I wasn't actually irritated and didn't want to convey that I was. (I am capable of getting actually annoyed in a forum discussion and don't mind saying so when I am, but I wouldn't want to suggest I was when I wasn't!)

I am not in a position to alter the memorization choices of SCS mages as regards defensive spells in response to SR, because I need the same set of choices to work on SR and vanilla installs. Ultimately SCS is and will continue to be written for the vanilla spell system first and foremost. Over the years I have gone quite a long way to also allow for SR, but going this far is more work than I'm willing to do. (Edited to tone down excessive exasperation!)