Neverwinter Nights Forum News (Jun. 02, 06)

Remove all ads!

Here are today's Neverwinter Nights forum highlights, collected by NWVault. Please take into account that these are only single parts of various threads and should not be taken out of context. Bear in mind also that the posts presented here are copied as-is, and that any bad spelling and grammar does not get corrected on our end.

Chris Priestly, Quality Assurance

ID: Waaaay too much treasure

---

I.e. I know that you get a "Token of mastery" that enables setting dungeon options, but how does a new character get it?

---

A new character only gets it when they finish the game. If you are starting with a new PC, that PC needs to get their own token.

There is a way to cheat in the token. I won't tell you how on teh open boards, but if you want it, you can send me a PM and I will let you know.

Edit: Note, I'll only tell if you have finished the game once.

*Spoilers* ID: Cutscenes
There should be a cutscene each time you enter a "boss" lari. There should also be a few others. They should occur after your first portal back to Little Timmy, one after your third portal back, and one after the sixth portal back. And there's some nice ones to sum it all up around the end as well.

So you should be getting those, but no otehrs. Hope this clarifies things for you.

*Spoilers* All About Boss Monsters...

---

OK here is who (in order) I've beaten so far, my poor P3 800 can barely handle the dungeon size now but loading/saving gives me time to work on my novel.

Outer Circle Bosses
Lord Antoine Baccha, Visier de Guise (Wereboar)
Widow Hagatha (Ancient Cloud Giant)
Maggris the Hive Mother (Giant Spider)

Inner Circle Bosses
Masterius the Lich King (Archlich)
Harat, Lord of Darkness (Devil/Demigod)
G'Zhorb, the All-Seeing Eye (Elder Orb?)

Just started G'Zhorb's Pit but I think I'm getting close to the end!

---

Nicely done 420.

These are indeed the 6 "bosses" in ID. Due to the rndom nature of the dungeons, you can get them in slightly different orders. You will always get the 3 "Outer Circle" before the 3 "Inner circle" but the order you get them in can be different.

ID: spells won't regenerate
Hello HTM, BioWare's Evil Chris Priestly here.

Wow, that's a new one on me. I have played as a Paladin and everything worked fine for me.

But let me get some details from you. The camp you are resting at is it one of the camps randomly generated in teh dungeon, one of the portable camps you can buy and carry eith you, or one of the camp at the base of the staircases?

Also, you are definately finishing your rest and not interrupting it in any way, yes? As in you allow your PC to stand back up and don't tell them to move too soon?

A few details and we'll see what's goign on.

Stanley Woo, Quality Assurance

Infinite loot.
You can also adjust the loot settings once the game has been completed once. Since there's already a "too much loot" thread open on the front page, I'll shut this one down. Please use that thread to discuss.

ID: Brazier Puzzle ?? (spoiler)
Yes, and you're on the right track. Try to figure out what that extra brazier (the one on top) is for and what it does.

then you'd need to figure out why the outer ring of spheres lights up too.

James Henley, Technical Designer

Multiclass

---

What about the XP penalty?

---

In the case of Elhanan's suggestion, races that have fighter or rogue as their prefered class would not be penalized.

pure sorc
I've always been a fan of the sorceror's ability to work on the fly, since they don't have to memorize any of their spells ahead of time. They may not be able to learn every spell in the game, but a well chosen selection is more than enough to see you through.

Peter Thomas, Designer

ID: Serious Balance Issues, RE: Loot?
When you beat the game you get the Token of Mastery. This opens the dungeons customization options on the dungeon entrance and the journal.

Infinite Dungeons scripting questions

---

Okay heres a question: What would have liked to do, but couldn't script? Or rather, what did you try to do but had to abandon as not practical? Some things, like changing area tiles on-the-fly, are impossible and so wouldnt feature in the original design specs. I also havn't seen any projectile-based traps, trap doors & pit traps, or and "multiple area locations" for example the interior of a house in a tileset.

Many people have designs / plans / work-in-progress etc for Rogue-like modules / sections (myself included ) so getting some feedback from a large professional team about potential pifalls / issues is very useful.

---

There were a number of things we wanted to do that didn't happen. Originally, I had looked into constructing an area on the fly through assigning tiles, but that was ruled out. Until late in the project we had planned to have areas in hakpaks, but that would have ended up taking too much programming time. It's interesting in that areas in hakpaks can be accessed successfully when you start a new game, but when you try to load one, it doesn't recognize the resources. It's more complicated than you think.

We had hoped to include a racing element for competitive racing, with each level providing a challenge and having competing parties strive to complete those goals. The party and faction system in NWN can be... frustrating, and it would have taken a long time to make sure it wouldn't break.

At one point we wanted mini-puzzles in the game as well. These would be tiny things like 2 statues flanking a door and you turn them towards each other to open it, or light two braziers. There are a couple reasons why that wasn't put in. The first was time, and the second was that areas are actually getting kind of crowded as it is. There was a big list of requirements of what had to be in each area, and we couldn't keep adding more.

The safe zones in ID are the large spaces in which you could find camps or puzzles. At one point we considered alternate spawns there, like having a small chance of it just being a loot room. Nothing quite like walking into a room covered in piles of gold.

I would have liked time to work on the treasure system a bit more with regards to item properties. This is one thing that seems to have generated a number of complaints. The main thing about the treasure is that the properties an item can get are basically determined by value and the item property multiplier NWN uses to determine it. We had an ideal treasure value based on level and settings, and try to fit things in up to greater than or equal to half that. Doing it again, I would probably have to divorce everything from the itemprops 2da and manually compile a huge list of abilities. It would probably make it more balanced for gameplay (less likely to get 100% immunity for example, which actually isn't that expensive), but would need a lot of testing too. Also we would probably get rid of the level value formula and manually assign one determined by level.

There was also work done at one point exploring the possibility of taking an ideal item (preconstructed in the toolset), and removing properties to give a random version based on the theme of properties it posseses. The problem with that is that the value formula for properties is weird, so it would take a lot to determine what the item was worth, and there's also the fact that adding and removing properties doesn't happen immediately. I think it was that a call to add or remoe a property was an event instead just a command, and the game gets around to the next update cycle. So if you remove a property and try to get the value of the item, it won't change because no properties were added or removed yet.

We also looked into the idea of making treasure properties more class specific. In multiplayer it would be confusing though, and you'd probably have to scan through everyone to see what applies. If you only took the classes of the killing PC, you wouldn't drop anything for henchmen and support PCs. Also adjusting the generic formula we used to determine which property was assigned to take specific classes into account would have been a real mess.

Another thing I wanted, but wasn't added because of complexity and QA feedback was to have more dungeon modification options added right at the beginning. It was originally conceived that it would open in tiers as you defeated the bosses, but in the end we went to have everything open at once after you beat the game.

We wanted to have random random tile lighting for each tile in an area as well, but didn't have time to implement that. Scripting commands for it exist, though.

Traps are very generic because they had to be spawned in at a waypoint. We got the ability to spawn them dynamically and set all their abilities, and the custom script they run determines the type of damage they do.

---

Quote: What optimizations did you use, or consider using, to speed things up? For example, when dealing with large amounts of 2da data and storing it in the database, Ive had great sucess (i.e. faster) putting the data as local variables on an item in a creatures inventory and putting the entire creature into the database in a single operation. Is there any code to cache 2da data as local variables, or do you rely on the hardcoded 2da cache? And why did you decide that?

---

To speed up storage, each dungeon is actually a single item that is written to the database in one call. It still takes a while since there are a large number of variables on that item. There is only one active in the module at any given time to keep the size down and we limit the places in which the database write happens.

Most of the information used in ID is in 2das in the various unencrypted hakpaks included with the module. When the game starts, it reads in huge amounts of data and stores it locally. The fact that it is in 2das is because we wanted to give players the ability to add new things to the module. You can add creatures, items, themes, objects, etc. To stagger the places in which these 2da reads happen, we have a certain amount loaded at the beginning, theme specific things loaded on level generation, and tileset specific things loaded at area generation. Each of them only has to happen once, so the more you play, the less delays you'd have.

We also had some assistance from the programmers, speeding up 2da accesses. It should now store the last 2 or 3 2das in cache, so each subsequent access from that 2da would be a lot faster.

---

Quote: EDIT: Would it be acceptible for people to post their own versions of ID-like systems, and then critique/comment/feedback about how similar or not they are to IDs?

---

I would prefer that you not post other systems here and limit this to questions about ID systems. Feel free to start other threads to discuss those, however.


More:

---

Like Primogenitor, I had a random dungeon project that I had been working on and off for a while. Now I know why patch 1.67 had so many useful functions for such a project!

I'm also very curious as to why certain decisions were made, and how certain scripting was implemented. I would definitely like to know the answer to Primogenitor's question, and I'm also curious as to why you decided to go with the size you did for the areas. I was actually surprised that such large areas were used. Why not smaller? Or bigger for that matter? Is it only a matter of load times and database access, or was there some sort of player-based aesthetic involved?

---

The size of the areas was actually increased a couple times throughout development because of new features. The core areas we created had all the possible features that could appear. That means lots of encounter points, treasure points, a miniboss point, secret areas, safe zones, a puzzle teleport point, plot giver space, a rest tent (if a stairs up), and little safe areas around entrances. We could have made them a bit smaller, but then each area would lose features (which the area cataloging system can handle) and it might end up feeling like there wasn't much there.

More:

---

However, Ive used the bioware database for something different. Rather than leveling up each NPC via LevelUpHenchman(), I store one template in the database at level 40 and then reduce it to the appropriate level via SetXP() after retrieving it; in my tests this is a much quicker method.

---

There is a field in the creatures 2da that allows you to choose what class a creature levels up in. In the beginning the idea was to have a given creature level up in a single class, then branch as it gets to higher levels, with some taking different specialization routes, eventually going to epic prestige classes, etc. This would have been determined randomly for each creature.

Unfortunately, we did not have time to debug that thoroughly, so creatures now have just a base class and one they can level up in later. There wasn't time to completely rework the system.

I am contemplating trying to make a version that cuts out all the end-user customization. As you say, it should run faster than the current implementation. It will likely not happen, though.