NWN Neverwinter Nights (NWN & NWN2) Modules Thread

[Gargaune]

Beamdog's QA has always been terrible. Though to be fair, I expect this particular issue would only be a problem in rare cases, as almost every module series that might be using a database would either have been completed long before EE or only begun after it came out. Swordflight is quite unusual in having continuing development across both the EE and pre-EE era.

Sure, but it would've also been easy to address, and it should've also been obvious at the software design level, well before QA.

The DB dump is done in the course of the final conversation in each module: that with Hamad Galhadr in Ch. 1, that with Norab Tefaram in Ch. 2, the conversation triggered by the transition going down to the Underdark at the end of Ch. 3, and the conversation triggered by the transition leading up from the Underdark in Ch. 4. So if you have saves allowing you to go through all four final conversations again in sequence that should in theory remake the database in EE (though note I have not actually tested this and relying on theory is one of the ways you get terrible QA). Make sure you do Ch. 4 last or your database will have the wrong version of Zarala.

Thanks again! I'll have to check, I don't remember off the top of my head whether any of my final saves are past those end conversations. If they are though, I should have other saves not too far back, can probably rush the endings again from there. I'll check to identify the right saves for each chapter, delete the .sql DB and run through them in order.

Hey, the hoarder's upside is that sometimes you do need that shit later!

 

[Popiel]

Gargaune You seem to know your way around NWN2, is Purgatorio module playable?

 

[KainenMorden]

So can anyone else confirm if divine might can give you extra attacks even if you're at the AB cap?

Also just to clarify, when under tenser's you're just considered an elf and not undead?

 

[notpl]

If your base attack bonus - listed separately on your sheet - is 16 or higher, then you cannot gain any more attacks. If it's 15 or lower, then divine power will give you an extra attack. No cap or anything is relevant to this.

 

[Gargaune]

Gargaune You seem to know your way around NWN2, is Purgatorio module playable?

Sorry, not familiar with it. Vault comments say there's some mileage in it, but that it's incomplete.

 

[VentilatorOfDoom]

Divine Power doesn't give you an extra attack. It improves your BAB to that of a fighter of the same level. Only if this BAB increase would result in an extra attack (according to normal attack progression) will you get that extra attack.
For example a lvl10 cleric has two attacks, a lvl 10 fighter has also 2 attacks. In that scenario Divine Power doesn't result in an extra attack. Otoh at lvl11 the cleric will get an additional attack from the spell because a lvl11 fighter has 3 attacks while a lvl11 cleric still has only 2.

 

[KainenMorden]

Divine Power doesn't give you an extra attack. It improves your BAB to that of a fighter of the same level. Only if this BAB increase would result in an extra attack (according to normal attack progression) will you get that extra attack.
For example a lvl10 cleric has two attacks, a lvl 10 fighter has also 2 attacks. In that scenario Divine Power doesn't result in an extra attack. Otoh at lvl11 the cleric will get an additional attack from the spell because a lvl11 fighter has 3 attacks while a lvl11 cleric still has only 2.

Can you explain what relationship it has to AB cap? Sorry I really find this confusing

 

[VentilatorOfDoom]

Can you explain what relationship it has to AB cap? Sorry I really find this confusing

In NWN the BAB increase of this spell counts towards that +20 cap.
NWN has a cap on attack bonus a character can get from weapon enhancement bonuses plus spells/spell like abilities. All weapon bonuses plus bonuses from spells or SLAs can never exceed +20.
For example if you have a +10 weapon and cast divine power for a BAB increase of say +4 and cast bless (+1) and Aid (+1) and Prayer (+1) and true strike (+20 AB for the next round) you will only ever get +20 AB because that's the cap.
You still might get that extra attack if applicable but for the purposes of making an attack roll bonuses to attack will be capped at +20.
I don't know how many cleric spells you can stack for AB increases but I'd guess you won't reach that cap anyway.

 

[KainenMorden]

Can you explain what relationship it has to AB cap? Sorry I really find this confusing

In NWN the BAB increase of this spell counts towards that +20 cap.
NWN has a cap on attack bonus a character can get from weapon enhancement bonuses plus spells/spell like abilities. All weapon bonuses plus bonuses from spells or SLAs can never exceed +20.
For example if you have a +10 weapon and cast divine power for a BAB increase of say +4 and cast bless (+1) and Aid (+1) and Prayer (+1) and true strike (+20 AB for the next round) you will only ever get +20 AB because that's the cap.
You still might get that extra attack if applicable but for the purposes of making an attack roll bonuses to attack will be capped at +20.
I don't know how many cleric spells you can stack for AB increases but I'd guess you won't reach that cap anyway.

Well, I will have to do some experimenting with my SF build when scrolls of divine power are available again. Nearing the end of chapter 4 now and I don't believe scrolls of divine might are available.

As far as I can tell, my ab reaches the cap from weapon enhancements + ability modifier + tenser's as a lvl 25 sorc.

Also, can you explain what the 2 free attacks in the description of tenser's mean?

 

[VentilatorOfDoom]

passive stuff like weapon focus feats or other feats that increase AB as well as STR or DEX mods aren't applicable for that +20 cap. Only weapon bonuses and spells.
those 2 free attacks just mean 2 additional attacks at full BAB (instead of iteratives at -5 each). Not sure though.
Divine Might is a feat, not a spell, fueled by turn undead attempts per day, there are no scrolls for that.

 

[Sabotin]

I was browsing a bit for database migration and encountered this, was it never implemented or it's just for servers?
https://forums.beamdog.com/discussion/76471/bioware-campaign-database-replacement-info

Also a fun fact about Divine Power: If you get any extra attacks they're made at full BAB, not at the tail end like it should be (e.g. lv20 cleric would have [20]/20/15/10 instead of like a lv20 fighter with 20/15/10/[5]).
Also be careful with character sheet, it may show wrong info, look at rolls.

While we're talking caps, you can see/change them in EE in options/game/ruleset/advanced.

 

[KainenMorden]

passive stuff like weapon focus feats or other feats that increase AB as well as STR or DEX mods aren't applicable for that +20 cap. Only weapon bonuses and spells.
those 2 free attacks just mean 2 additional attacks at full BAB (instead of iteratives at -5 each). Not sure though.
Divine Might is a feat, not a spell, fueled by turn undead attempts per day, there are no scrolls for that.

I have the divine might feat, there are scrolls of divine power in SF.

I will have to take some screenshots or take another look at my combat log in some of the fight videos on my YouTube channel. I just remember my AB was capped without using anything to enhance ab beyond improved invisibility and tenser's

 

[KainenMorden]

https://youtu.be/PM51ZZ5wYUo

Uploaded the Vartakor fight, it was a very fun final fight for the chapter. For some reason, I thought Balors were immune to fear but mine needed protection from evil for this fight. Tenser's on its own wouldn't cut it, I wasn't able to do physical damage at all when polymorphed because tenser's sword is only +3. I'm usually not much for RPing but I did ponder the implications of a Lawful Good character summoning an evil creature. I see it as doing what needs to be done for the greater good, even if that means colluding with demons and liches and other evil entities at times but bringing evil into the world would be seen as an act of evil to others regardless of circumstance. Also, the irony of summoning a Balor to defeat a Pit Fiend while in the employ of Celestials is not lost on me.

Also uploaded the grey watcher quest balor fight, azoophun fight and shadow dragon fight. Long battles but a user on reddit said I wouldn't be able to beat this chapter with melee, I took this as a challenge and wanted to see if it could be done.

I enjoyed both ch.3 + 4 very much and Ch.5 is already installed and up and running.

My house is currently being remodeled, amongst other personal time commitments, I don't give a fk, I miss out on sleep at times to play SF. Can't thank rogueknight enough for creating these incredible modules.

Funnily enough, one player that criticized SF on reddit but never played it, now recommended it to a new user who was asking about SP modules. Maybe he finally tried it for himself and realized the error of his ways? One can only hope.

 

[KainenMorden]

https://imgur.com/CEQ7RK9

So here's a screenshot of my SF fight posts on reddit, it appears that a large amount of users look but few comment. I'm not entirely sure how the insight system works on reddit but I'd imagine its fairly accurate.

Now for some reason, my latest posts that contained the rest of the chapter 4 fights were not viewed by anyone, I assume I'm shadowbanned or something on that sub reddit.

I just want to get eyes on SF even if my long fight videos are not likely to be viewed by many. Either way, I suppose I'll register another account there and try again....

 

[KainenMorden]

First impressions of Ch.5, I "respecced" to get epic spell focus evocation and the idol quest was tough. The battle outside the cave took quite a bit of reloading but part of it is also me having to learn how to use offensive magic. Sunburst is invaluable, I've also been using that greatsword that casts icestorm on hit.

If I were to use the holy avenger rune on it, would that negate the penalty on the weapon anymore than greater magic weapon would? I imagine it wouldn't but was just curious. The spell resistance granted from the holy avenger property might be helpful on this weapon but I can also become almost immune to cold damage between spells and equipment as it is.

 

[rogueknight333]

Both GMW and Holy Avenger add a +5 Enhancement bonus. There should be no difference between them with regards to that.

 

[Gargaune]

rogueknight333, I just spent some time testing and setting up my Ch.5 start if you're interested in the academic results. Correlating my inputs with the modified date on the EE's .sql DB for Swordflight, I've made the following observations:

1) The Swordflight SQL DB is touched (meaning updated or created if there is none) whenever the EE accesses the character localvault, regardless of whether you're loading a Swordflight module or an unrelated one, but only if the old DE format DBs are present in the database folder;

2) The Swordflight SQL DB is touched when loading any Swordflight savegame, but only if the old DE DBs are present and only when loading from the main menu (loading one SF save while already playing SF doesn't touch the DB);

3) The Swordflight SQL DB is touched whenever manually called by your module via the final conversations, as expected.

I've also had a brief look at the SQL DBs size, but wasn't thorough enough to lay it out. In a nutshell, though, the SF SQL DB that gets created when accessing the localvault or loading a savegame stays the same size (76KB in my case) regardless of which ones or how many times you load. It only increases when you go through the module finales. Note that an SQL DB built with DE DBs present will start and remain larger than one initiated with no DE DBs present, which suggests that data keeps getting added next to the presumably ported DE stuff rather than overwriting it.

Going back to the modified dates it seems that the EE does have a legacy compatibility process to check and translate old DE databases. This is called when the localvault is accessed for any module or when loading a save related to that database (though, curiously, only when loading from the main menu), and will attempt to translate old DBs if they are present.

As for my own setup, I did what you suggested - removed the new Swordflight SQL DB I had in there (as well as the localvault entries), and just loaded the final saves I had for the prior chapters and took each to completion in chronological sequence. I did keep the old DE DBs in there, so a port process did happen, though broken (see below). Upon starting Ch.5, Zarala's alignment is correct per her ending state in Ch.4, as is her inventory, and querying the Ch.2 henchman variable returns 2 for Chelys. So, unless you've got another variable I should check, it appears that my progress history has ported correctly?

Then it hit me as I was writing this and tested the obvious next step - indeed, if I delete the SQL DB and start Ch.5 with just the old DE databases, Zarala's reset to default equipment and alignment and the henchman variable returns 0, just as if I had no DBs in there at all. So, to revise that conclusion two paragraphs up, it appears that the EE does have a process to port old databases to new saves... and that said process doesn't work. It's definitely carrying something across, but maybe it's failing to match the data to the character, I dunno. Final tests I did were going through only the finale of Ch.4 with either no databases or the old DE databases, the results were the same - Zarala's state and inventory are saved for Ch.5, but the henchman variable is 0, so historic DE data isn't getting loaded, it's just dead weight in there.

 

[rogueknight333]

So, unless you've got another variable I should check, it appears that my progress history has ported correctly?

If it has gotten Zarala's alignment and which rogue henchman you had in Ch. 2 correct, most likely everything is. If you really want to be thorough you could check a couple other things, e.g., if you acted as a spy for Olim Khachar in Ch. 2, there should by a variable "KHACHSPY" in Ch. 5 set to 1. Also there should be a variable "HENCHCASTDEAD" that should be set to 1 if you took Talu with you on the final quest in Ch. 2, or 2 if you took Snajagh with you. Neither of these actually have much effect on anything in Ch. 5, other than some slight differences in dialogue.

... it appears that the EE does have a process to port old databases to new saves... and that said process doesn't work.

Yes, the Beamdog forum posts Sabotin linked to also seem to indicate that databases were supposed to be compatible. So the quality of work we have come to expect from Beamdog, I guess?

 

[Gargaune]

If it has gotten Zarala's alignment and which rogue henchman you had in Ch. 2 correct, most likely everything is. If you really want to be thorough you could check a couple other things, e.g., if you acted as a spy for Olim Khachar in Ch. 2, there should by a variable "KHACHSPY" in Ch. 5 set to 1. Also there should be a variable "HENCHCASTDEAD" that should be set to 1 if you took Talu with you on the final quest in Ch. 2, or 2 if you took Snajagh with you. Neither of these actually have much effect on anything in Ch. 5, other than some slight differences in dialogue.

Thanks, I'll double check when I get the chance.

Actually, very vaguely related, meant to ask about saving campaign variables if you don't mind. DBs are for porting data across distinct modules, but what would be good practice on storing progress variables within a single-module campaign? Been saving to the module object so far to avoid having to scrub the PC at export, but I'm in two minds about how that would work for coop multiplayer compatibility. It's not a priority, but wouldn't mind doing it the right way if there is one. The third option I was pondering was saving to a plot-flagged object in the PC's inventory.

Yes, the Beamdog forum posts Sabotin linked to also seem to indicate that databases were supposed to be compatible. So the quality of work we have come to expect from Beamdog, I guess?

Quite disappointing. Not sure what's worse, neglecting to plan for a DB port or implementing one that just doesn't work. Sabotin, I didn't see your post, could you link again? Just curious about it.

 

[luj1]

Bumdog

 

[Sabotin]

...Sabotin, I didn't see your post, could you link again? Just curious about it.

It's just a couple of posts up: https://forums.beamdog.com/discussion/76471/bioware-campaign-database-replacement-info

Caveats: They're talking about PW and only mentioning RetrieveCampaign*** instead of GetCampaign*** which is used to fetch stuff in SF (and in general?).

Btw, if you're only looking to port variables, the .dbf file can be opened with excel and then you can basicaly copy/paste them into the sqlite db with an appropriate browser or use console commands to set them ingame (rogueknight333 gave an example of this in one of the previous posts a few pages back).
I assume the .fpt with the henchman inventory and stuff can also be opened/exported with some application but what I've found on the internet seemed pretty involved, with some db knowledge/setup required.

Also I never thought I'd have a backlog of NWN modules after a decade xD.

 

[rogueknight333]

Actually, very vaguely related, meant to ask about saving campaign variables if you don't mind. DBs are for porting data across distinct modules, but what would be good practice on storing progress variables within a single-module campaign? Been saving to the module object so far to avoid having to scrub the PC at export, but I'm in two minds about how that would work for coop multiplayer compatibility. It's not a priority, but wouldn't mind doing it the right way if there is one. The third option I was pondering was saving to a plot-flagged object in the PC's inventory.

I do not have a lot of experience making multi-player friendly modules myself. In principle you can set variables on any object in the module that is convenient for you. If there is some reason to prefer one type of object over another as a general rule I do not know about it (other than obvious cases like not setting variables you need to preserve on something that could be destroyed). Multi-player modules often set variables on the specific PC since that way you can set conversation & quest states just for that character and still allow other players to start them from the beginning, though if you expect players to go through the module playing as a single team that might not really be necessary (Recommended procedure when playing co-op in a module not really optimized for MP is to have one player act as the party spokesman and handle all coversations with quest givers).

Variables set on PCs or items do not survive if exported. One can still use items as an alternative to using databases to transfer information between modules by giving players items with a custom tag in one module to mark certain quests as completed or other data, and then check which of those items are present in the sequel module. One module series I know of transferred data by giving players, upon leaving the module, a bunch of custom bullets with a unique tag to indicate certain variables, and setting the size of the bullet stacks to set the variable number, and then in the sequel checking for these items, setting the indicated variables in that module, and then destroying them. Something like that could avoid some of the database issues we have been talking about, though a possible downside is that if someone plays an exported character employed in such a system in some other module than the intended sequel, which would lack the check & destroy script, he may find a lot of weird junk in his inventory.

 

[Gargaune]

It's just a couple of posts up: https://forums.beamdog.com/discussion/76471/bioware-campaign-database-replacement-info

Sorry, I was (and still am) quite tired, went skimming the prior pages instead of just scrolling up.

Btw, if you're only looking to port variables, the .dbf file can be opened with excel and then you can basicaly copy/paste them into the sqlite db with an appropriate browser or use console commands to set them ingame (rogueknight333 gave an example of this in one of the previous posts a few pages back).
I assume the .fpt with the henchman inventory and stuff can also be opened/exported with some application but what I've found on the internet seemed pretty involved, with some db knowledge/setup required.

Thanks, I wasn't aware that there were online .sql viewers, but now that you've mentioned it, I found this: https://inloop.github.io/sqlite-viewer/

Which also solved the fucking mystery:
- DE databases record the player key as "Player FirstName LastName"
- EE SQL databases record the player key as "FirstName LastName"

So I was right, the data's ported but it is failing to match the player, I just didn't imagine it was something this ridiculous. Both are truncated at the end, but identically so. I did try trimming the Player prefix from the DBF in both Notepad++ and Excel, but the resulting database is invalid, the SQL port will produce an empty one from it. And it wouldn't cover the CDX and FPT either, so no point.

rogueknight333, two items in relation to this:

1) While redoing all my chapter finales did construct a new SQL DB, I noticed there is one variable missing in the new one, HAMADTOLD. I think I might've installed updated versions of your modules on the EE relative to the DE ones that made the original DB, not sure anymore. Is this relevant in Ch.5 or do you expect it to become relevant in Ch.6 going forward?

2) You might actually have a solution here, though I haven't had an opportunity to test it. Since the DB can't be messed with easily, you can approach it from the other end - the character. You could have affected players use GFF Editor to open their character .bic in the localvault and prefix their FirstName with "Player ", start the module with it and make a save, then use GFF Editor again to open the player.bic inside the save's folder and change it back.

EDIT: Upon reflection, even if this works, it'll still kick the can down the road to Ch.6 at which point your Ch.1-4 data will be keyed to Player Bob and your Ch.5 to Bob. So you'd either have to repeat the GFF rename just before Ch.5 does its DB dump, or learn to love playing Swordflight as renowned hero Player Bob Bobson.

I do not have a lot of experience making multi-player friendly modules myself. In principle you can set variables on any object in the module that is convenient for you. If there is some reason to prefer one type of object over another as a general rule I do not know about it (other than obvious cases like not setting variables you need to preserve on something that could be destroyed). Multi-player modules often set variables on the specific PC since that way you can set conversation & quest states just for that character and still allow other players to start them from the beginning, though if you expect players to go through the module playing as a single team that might not really be necessary (Recommended procedure when playing co-op in a module not really optimized for MP is to have one player act as the party spokesman and handle all coversations with quest givers).

Variables set on PCs or items do not survive if exported. One can still use items as an alternative to using databases to transfer information between modules by giving players items with a custom tag in one module to mark certain quests as completed or other data, and then check which of those items are present in the sequel module. One module series I know of transferred data by giving players, upon leaving the module, a bunch of custom bullets with a unique tag to indicate certain variables, and setting the size of the bullet stacks to set the variable number, and then in the sequel checking for these items, setting the indicated variables in that module, and then destroying them. Something like that could avoid some of the database issues we have been talking about, though a possible downside is that if someone plays an exported character employed in such a system in some other module than the intended sequel, which would lack the check & destroy script, he may find a lot of weird junk in his inventory.

And thanks again for all this! My rationale is that, if possible, I'd like to keep things coop-friendly though I'm not too worried about story consistency, since it's designed as a single-player experience. I'm assuming that if people were to play in multiplayer, they'd all be in the same party and progress at the same time. If NPCs end up acting like they know players they don't etc., that's no big deal.

 

[rogueknight333]

1) While redoing all my chapter finales did construct a new SQL DB, I noticed there is one variable missing in the new one, HAMADTOLD. I think I might've installed updated versions of your modules on the EE relative to the DE ones that made the original DB, not sure anymore. Is this relevant in Ch.5 or do you expect it to become relevant in Ch.6 going forward?

That variable is set in Ch. 1, and only relevant in Ch. 2, so does not matter at this point. Possibly you made a different choice when redoing the final conversation in Ch. 1 this time, resulting in it not being set (since IIRC correctly that is just a 0 or 1 variable).

2) You might actually have a solution here, though I haven't had an opportunity to test it. Since the DB can't be messed with easily, you can approach it from the other end - the character. You could have affected players use GFF Editor to open their character .bic in the localvault and prefix their FirstName with "Player ", start the module with it and make a save, then use GFF Editor again to open the player.bic inside the save's folder and change it back.

EDIT: Upon reflection, even if this works, it'll still kick the can down the road to Ch.6 at which point your Ch.1-4 data will be keyed to Player Bob and your Ch.5 to Bob. So you'd either have to repeat the GFF rename just before Ch.5 does its DB dump, or learn to love playing Swordflight as renowned hero Player Bob Bobson.

Possibly an option, though rather complicated and not ideal. Another potential issue is that a player's MP account name (normally what is used, "Player" is just the default if you have not made such an account, e.g. for me the name would be "rogueknight333Bob Bobson") added to the character's first name might possibly exceed the character limit, if both are very long.

Which actually suggest another possible database issue that might come up, unfortunately (and this one is not even Beamdog's fault since AFAIK it would be equally relevant in either EE or DE). If a player starts a new MP account in the middle of playing Swordflight or has multiple accounts and sometimes switches between them, stuff might end up getting written to the DB under multiple player names thus preventing some of it from being retrieved properly.

 

[Gargaune]

That variable is set in Ch. 1, and only relevant in Ch. 2, so does not matter at this point. Possibly you made a different choice when redoing the final conversation in Ch. 1 this time, resulting in it not being set (since IIRC correctly that is just a 0 or 1 variable).

Thanks, I'll ignore it then. I did rush those end conversations when rebuilding the DB.

Possibly an option, though rather complicated and not ideal. Another potential issue is that a player's MP account name (normally what is used, "Player" is just the default if you have not made such an account, e.g. for me the name would be "rogueknight333Bob Bobson") added to the character's first name might possibly exceed the character limit, if both are very long.

Realistically, I think the only option for people who don't have their saves at this point is to have the EE port their DE database and then manually adjust the name key with an SQL editor. I did have a skim last night for some GUI-based solutions, but I've never used one, wouldn't be able to recommend one over another.

Which actually suggest another possible database issue that might come up, unfortunately (and this one is not even Beamdog's fault since AFAIK it would be equally relevant in either EE or DE). If a player starts a new MP account in the middle of playing Swordflight or has multiple accounts and sometimes switches between them, stuff might end up getting written to the DB under multiple player names thus preventing some of it from being retrieved properly.

Maybe that's why Beamdog dropped the player account from the entry, trying to fix BioWare's earlier cockup? The EE shouldn't have this problem anymore since the player account doesn't feature in the campaign DB. Unless, of course, the EE DB does add an account key if it's not just "Player", but I can't tell, I don't do MP.

Anyway, I can see now how they'd stumble on porting the player ID from the old format, there's no apparent way to parse it back to individual fields, but then I don't get why you'd bother implementing a DE-EE translation process that just won't work.