Icewind Dale Remix Notes (Chapters 1-5)

Almost three years ago, I ran an online campaign of Icewind Dale: Rime of the Frostmaiden, WOTC's published campaign. I documented the first few sessions here, but didn't keep up with it (just like I haven't kept up with the documentation on any of my campaigns since Castle Xyntillan... sigh [EDIT: my AD&D campaign is being very well documented!]), though it ended well enough after a little under a year of play. 

Just now I got an itch to revisit it. While I've used published material before (the Ptolus sourcebook for 5e, Castle Xyntillan, various published adventures for Legend of the 5 Rings), that remains the only time I've run one of these big, campaign-spanning adventures. It also existed in this very productive space where it was good enough to care about and flawed enough to critique. 

So I've decided to go through and 'remix' it. I'm taking inspiration here from both Justin Alexander's Remixes, especially his treatment of Descent into Avernus (which also finished three years after the release of the original!) and Joseph Manola's Condensations of Paizo adventure paths. This is going to be somewhere in the middle between the two, offering both my opinions on the design of the adventure as well as modifying and condensing it, without being nearly so thorough or detailed as Alexander. 

Character Creation and Reaching Ten Towns

Icewind Dale starts out with character creation options: expansions on the standard backgrounds, the goliath PC race, and secrets. The latter is neat: every PC gets a secret from a list (possibly randomly assigned, possibly chosen from a selection). Now, some of these aren't really what I would call 'secrets' like the one that infects a character with slaad eggs, but a bunch of them are quite neat, add some campaign-specific flavor, and work to make the characters just a bit suspicious of each other. That last objective doesn't seem to stick around, as the rest of the campaign is not especially oriented toward intra-party intrigue, but it's a nice touch. It's a progression from the 'secret' mechanic in Descent into Avernus, published the previous year, in which the whole party is bound by a particular secret, and which by all accounts doesn't work very well. 

Kill it with fire!

Cold Open

After character creation, the game advises starting out with a two-paragraph narration about the state of the Dale, after which the PCs arrive in one of the towns and get either Cold-Hearted Killer, or Nature Spirits, or the local quest. The adventure punnily calls this the 'Cold Open.'

Punny, and wrong, and a waste, since an actual cold open would work quite well here.

Ten Towns Overview

The first section of the campaign is expected to take place in Ten Towns: a collection of settlements nestled among a few lakes in the middle of the Dale. Each town has its own quest, and there's a default assumption that the early campaign is going to involve traveling from town to town and picking up each quest until characters get to high-enough level that they can journey out into the far wilderness. 

The adventure nods toward the possibility of starting the campaign in any one of the towns, but this is, to put it delicately, less than fully supported. Several of the town quests, such as that for Easthaven, are probably TPKs for a first level party, assuming the mindset of 'fight everything and never run away.' More to the point, this sort of campaign makes the most sense if at least some of the player characters are foreign to the Dale, entering for the first time during the long winter. Following that, it also makes the most sense to simply start the campaign with characters entering the Dale, in which case it is natural for the first town they enter to be Bryn Shander, the gateway to Ten Towns and, apparently, the only town with a road leading to it from the outside. It also helps that Bryn Shander's quest is supposed to introduce the party to the blizzard mechanics and features a fight against a crew of goblins, as opposed to, say, a hag and giant skeletons. 

It wouldn't surprise me if at some point in development, Bryn Shander and its quest was the default introduction, but got folded in out of some desire to be more sandboxy. As much as I like sandboxes, and as much as it's a good thing that WOTC tried to move in that direction (we'll see how it broke down later), I think this was the wrong call: sandbox does not mean 'no guidance', nor does it mean 'lack of a strong introduction.'

What I did

I replaced the not-so-cold open from before with something of my own: the PCs (some locals returning to the Dale, some foreigners coming for the first time) journey into the benighted land. 

Two days ago, your wagon crested the Spine of the World. That was the last time you saw the sun. For just a few hours around noon, the sky turns to twilight. After that, there’s just the stars and the waxing crescent moon bearing down on you. 

Caravans in and out of the Dale are rare nowadays, and you ended up sharing the trip with a wizard and his bodyguard. He introduced himself as Dzaan, and his bodyguard as Vetala. Luckily, his magic has kept the covered wagon warm and cozy despite the deadly cold outside. 

With little to do on the trip through the mountains, he explained to the party how he was chasing rare elemental spirits called chwingas, and offered the party a deal. 25gp and a magical lantern in exchange for a live chwinga.

Dzaan is an NPC that shows up later, and I decided to give him some relevancy early on, both to consolidate a few questgivers into one, make it more impactful when the PCs find his clone, and give the Arcane Brotherhood a bit more presence and weight. 

Then the sled stops: there's another sled blocking the road, headed in the other direction: a pack of 4 wolves is gnawing on the bones of the driver and sled dogs. The party may fight off the wolves (as mine did) but they can just as easily scare them off with some fire, loud noises, and offering some rations. This is not supposed to just be a combat encounter, but one which sets expectations: Icewind Dale is characterized in large part by hunger. 

The driver is a man from Bryn Shander named Jericho, and he was killed with a spike through the heart, not by the wolves. His sled is too lightly supplied to be making a trip over the Spine, but there's nothing else for him to be doing headed in this direction. It looks like he was making a run for it. There are prints nearby, fresh in the snow, but they're quite far apart, as if the person making them could jump clear across thirty feet of snow and land soft as a feather. They lead straight to the gates of Bryn Shander, both ways.

As much as I whinge, gotta love the art

Intro Quests

The other wrong call here surrounds the wandering quests: there are two quests which the PCs can get right out of the gate which explicitly exist to encourage them to move from town to town, Cold-Hearted Killer and Nature Spirits. This is a good decision, and can help initiate players into a sandbox attitude; this was one of the much touted gimmicks of the campaign for a reason. 

Which makes it very difficult to understand why the campaign explicitly recommends using only one quest, not both. One is a lighthearted quest about searching for whimsical snow fairies, and the other is the search for a serial killer, so maybe this can be justified as a dial for GMs to set their campaign's tone, but it doesn't land for me: a lot of the campaign's best bits juxtapose the apocalyptic winterland with the absurd and whimsical. Every commentary on the campaign I've read has called this out, and recommended using both. This has benefits with juxtaposing two tones productively, and also makes sure the party has a reason to check out more towns. 

That said, there are some weaknesses here. In both cases, the introductions to the quests are pretty weak; both start with the party being approached by an NPC who hires them for the job. This makes some sense for Nature Spirits, but not for Cold-Hearted Killer. In the latter case, here's how it gets introduced:

Out of boredom and a sense of moral decency, Hlin has taken it upon herself to investigate the recent murders because no one else-not even the Council of Speakers-can be bothered. Hlin is studying the characters closely, trying to decide if they're worth her time. Ultimately, she takes the chance and draws them into conversation, asking them to help her take down her only suspect: a man named Sephek Kaltro.

Yawn-o-rama. 

There's a seed of something good here, in the subtext. An upright but desperate citizen who has failed to get the local authorities to move turns to the newcomers. It implies that Icewind Dale is a place where justice has fallen to pragmatism, possibly even one where the authorities conspire with the serial killer. But none of this is borne out by the quest as written. There's no sense that the fearful citizenry might try to stop the PCs from investigating, or that the authorities might apply any pressure. 

Instead there's just...apathy, 'no one else can be bothered.' The read aloud mentions how much gossip there is about the killings, but not mention of anyone else looking into it, measures people are taking to defend themselves. 

Not to mention that the boxed text sets up a faceless serial killer, but the questgiver comes out and correctly identifies him before the PCs have a chance to do any investigating (yes, the book says explicitly that this is a hunt, not a murder mystery: I still think it's the wrong call)! This is doubly silly because one of the major complaints about the campaign's first chapter is that Sephek is a tough combatant who stands a good chance of slaughtering a first or second level party, even alone and against freshly rested PCs. Laying out an investigation with each town providing, or having a chance to provide, some clue, going to the towns where the serial killer has struck and investigating there, asking after the friends and family of the victims, all of this would have the benefit of both building tension, engaging the players with an actual investigation, letting them think for themselves, and letting them level up a bit before facing him down. 

Nope. I know who the serial killer is, but somehow nobody else in this extremely paranoid and fearful tavern obsessed with the killings will believe me. Really? I would expect the townspeople to be organizing witch hunts and forming posses, not meekly sitting by. Tomfoolery. 

What I did

I used both, and also made modifications to each. The party met the serial killer, Sephek Kaltro, in Bryn Shander as soon as they got in: he thought the body they were carrying was a bounty, and offered to buy it off them and take it the rest of the way. I replaced Hlin with the sheriff of Bryn Shander, someone in a much more plausible position to offer a 100gp bounty on a serial killer. The party never took a proactive stance towards hunting the serial killer, so he showed up to bite them in the ass during the White Moose quest instead. Good times. 

In my version, Sephek could hop bodies, and Velynne Harpell helped them put him down permanently. In reality, she imprisoned the spirit and put it toher own ends. This didn't really come up again in my campaign, but it might come up in yours!

I also consolidated some NPCs: instead of getting the Nature Spirits quest from Dannika Graysteel, they get it from Dzaan. Instead of just being approached out of nowhere, he gives it to them on the sled coming in to the Dale. Also, I made it a bit more sinister: when delivered, the chwinga was upset by Dxaan, who put it to sleep in order to 'study' it. The PCs didn't like that one bit, and this cemented Dzaan in their minds. 

Ten Towns

After the starting quests, the various Ten Towns quests take up the rest of the first chapter. I'll go through the ones that have something of interest. 

Bryn Shander: Foaming Mugs

I wrote about this quest intro in Stop Writing Lazy Quest Intros, which just about tells you what I think of it. I'll recap briefly: if a bunch of pathologically trusting NPCs approach the party out of the blue and offer a bunch of money in an easily concealed form in exchange for a dangerous job, they're either doppelgangers or they can't complain when the PCs mug them. 

Also, this quest involves journeying out into the tundra to retrieve a shipment of iron ingots, but the reward is worth more than the ingots are. Now, that's not strictly bad, but you need to set up why that is: maybe iron is in short supply, and it's worth overpaying to get it back both so it can be turned into merchandise and actually make a profit, and/or because the reputation loss for losing the iron and the subsequent debt to the miners/smelters would be more damaging than losing gold. 

What I Did

In Bryn Shander, I moved the three questgiving dwarves to the House of the Morninglord, the temple/hospital, where they were recovering from injuries and frostbite after escaping the yeti. Since the party was bringing the body of Jericho here for proper rites, they encountered the dwarves quickly. It's more natural, and also builds up the environment as a threat. 

Bremen: Lake Monster

Pretend the fishing table doesn't exist and it's fine. I folded the questgiver Tali into Dzaan, who gave this to the PCs right after they gave him the chwinga.

Hey guys! Room for me in there?

Easthaven: Toil and Trouble

Another underwhelming quest hook I complained about before. The captain of the guard sidling up to the PCs while a wizard is getting burned at the stake and hiring them out of the blue will never stop being funny. My players never visited this town, but the quest looks serviceable and having one of these relate to the problems of food is good. 

Good Mead: The Mead Must Flow

The quest itself is fine, but the 'election' afterwards is bare-bones and pretty heavily railroaded. I addressed this in more detail here. 

Lonelywood: The White Moose

Good overall, the Elven Temple is a standout adventuring location among these early quests. The presence of the mummy, which my players freed but never really used, makes me wish 5e had proper henchman rules. 

Targos: Mountain Climb

I have to rant about this one. The premise of Mountain Climb is that a group of adventurers seeking a nigh-mythical werebear climbed the nearby mountain of Kelvin's Cairn with a guide, but got ambushed by yetis. So far so good. The guide is barely hanging on to life, and the only escapee is his dog, who goes to get help for its master. Okay, I'm interested, players tend to love loyal animals like that too. 

Now, put yourself in this dog's position. Your master is likely to die soon, and you need reliable help fast. Where do you go? Maybe down the mountain to the town of Caer Konig, which is the location of Frozenfar Expeditions, the company with which your master is contracted and from which the exploratory group set out? You know, the place which is nearby, inhabited, known, and which has people that will understand what has happened and have the expertise to mount a mountain rescue?

Or do you instead run across fifteen miles of dark, open tundra to get to the town of Targos, where your master's house and husband are, only to run to the first group of heavily armed strangers you come across and drag them to that house so that they can mount a rescue... maybe. 

Yeah, this hook just doesn't work if you have the map open. I moved this quest to Caer Konig and replaced it with an investigation into an underground Targos fightclub modeled on the WWE where the PCs were promised stardom by halfling Ron Jeremy and the druid developed a crush on the headliner, a goliath brawler named the Monolith, who beat him up. Good times. 

What I did

On Kelvin's Cairn, I added another room to the yeti cave, a control room for an ancient anti-spelljammer laser cannon hidden in the peak of the mountain. It has three shots left, can target anything above the horizon (like, say, a flying dragon), and can be controlled manually, though it still has functional targeting systems that strike at spelljammers that come in range. This comes up later. 

Chapter 2: Icewind Dale

After the party hits level 4 (which, given milestone leveling, is equivalent to doing five of the ten town quests), they are supposed to stop advancing by doing town quests (though they can still gain reputation) and should instead be turned toward the quests out in the wilderness of the dale. 

This division between the more-or-less civilized region around Ten Towns and the dangerous wilderness beyond is a good one, though the hard boundary here is less desirable. I'm informed that many parties chose to stick around and do all the town quests out of a sense of completionism. 

The big offender here is how these new wilderness quests are acquired. The book offers two options. The first is 'Tall Tales.'

By the time the characters reach 3rd level, they have garnered enough of a reputation that NPCs in Ten­ Towns will share tall tales with them. When the player characters are ready to explore more of Icewind Dale, use the Tall Tales in Ten-Towns table to entice them.

You can either choose which tales to share with the players, or you can ask each player to roll once on the ta­ble. Let the players decide which tales (if any) they want to investigate.

I recognize that the writers wanted to add some flavor to the hooks, but it doesn't work. First, all the information in these 'Tall Tales' is completely accurate. Second, the above level requirement ensures that the PCs won't hear about these quests/locations until they're almost ready to go out and do them. To the contrary, you'd want the PCs to hear about this stuff early, before they're ready, so they can look forward to it, so they can wonder, maybe get in over their heads a bit. 

The other option is for an NPC to hire them to go to each of these locations and either run an errand or kill everything there. Mind you, the setups for these are much better than the ones before: the PCs have built up some name recognition by this point, it makes sense for people to come to them with problems. Even better, some of these 'errands' make a good deal of sense: given that the wilderness is dangerous, getting the PCs to bring supplies to a remote location and check up on someone (Black Cabin) or supervise the delivery of valuable and scarce whale oil (Angajuk's Bell), actually fits. These aren't revolutionary hooks, but they don't have to be, they're just effective and make sense for the setting. 

The other notable aspect of the wilderness setting is a random encounter table: the gimmick here is that you roll both for encounter and blizzard, and the chance of a given encounter changes depending on whether there's a blizzard. It's a neat idea, one I may steal. 

So why, when I go onto forums discussing the adventure, do I find the prevailing opinion to be against the random encounter table?

There's a pat answer in here, something something 5e play culture, but I'd rather not descend into self parody. Frankly, I think it's just because the wilderness encounters are not very well integrated with the other mechanics. 

To be sure, there's a lot to like about it: at least a third of encounters aren't straightforward combat, and many others are unlikely to attack immediately; a few lean instead towards negotiation or other interactions, and a couple really deliver a nice moment at the table (mammoth-riding giants outlined against the rising moon? Hell yeah!). 

Hell yeah!

The issue is twofold. First, the encounters are pretty frequent. The adventure recommends rolling 1/day, with half of days having one encounter, a quarter having two, and the remaining quarter having none. That variance does get us some distance away from players blowing all their resources on the encounter because they know there will only be one (the Vaarsuvius Problem), but they still don't tend to have much of an impact on the journey. 

This is rooted in the short-term focus of 5e, with very little room for mid-long term resource management. So long as the players survive the 1 or 2 wilderness encounters, they'll get basically everything back with a long rest. There really isn't room for a wilderness expedition during which player resources experience attrition on a scale longer than a single day. As a result, these encounters don't successfully add time pressure to travel, nor do they incentivize the players to seek out shelter to recuperate, nor do they meaningfully limit how far and how long the players can journey away from civilization. They're speed bumps, and if you roll them as expected, you will have a lot of them: even leaving aside the shorter non-combat encounters, and assuming the DM ignores some of the weaker encounters which wouldn't meaningfully challenge the players, you'll still be playing through a lot of combats which will be forgotten the next in-game day. Maybe you've gotten your players to the point where combat moves fast in that system, but my campaign was online, and my players dithered. Moreover, I had a limited number of sessions in which to actually complete this campaign. I dropped the wilderness encounters after the first few times they showed up. 

All that said, the bulk of this chapter is a bunch of keyed locations, basically small, 5-15 room dungeons. I'll note down the ones about which I have something to say. 

Angajuk's Bell

Didn't get too much play in my campaign, but still one of my favorite locations. The presence of an awakened sperm whale that can guide the party across the Sea of Moving Ice is a great addition. If I were to run this again, I'd add more things to do in that area. 

Black Cabin

Another standout, frequently discussed on forums owing to the many near and complete TPKs that occur here. The implementation and metaphysics is a bit wonky, but worth remixing, which is an exercise left to the reader. 

Cackling Chasm

Didn't use it, straightforward hack and slash. Also the town quest leading here is far too twee for my tastes. 

Cave of the Berserkers

Never ran this. Out of all the wilderness locations, the only one that I read and think 'ugh, that would be a pain to run.' Doesn't help that the whole 'chardalyn' element of the campaign doesn't appeal to me at all. 

Dark Duchess

Standout: a wrecked merchant ship used as a secondary hoard by an ancient dragon. Likely to put the players face to face with something they absolutely can't take on in combat. 

Id Ascendant

One of my favorites, though I changed the exact setup: the illithids wound up becoming player allies after some misunderstandings. In my version, the Id Ascendant only appeared after the players got to the top of Kelvin's Cairn and accidentally activated the Ostorian anti-spelljammer laser cannon secreted inside the peak, in a secret passage beyond the yeti caves. The ship got shot down and the players had a very good reason to investigate. Good times. 

Jarlmoot

Boring as written, and an uninspiring hook (pro tip: getting an NPC to lure the players out to a remote location to murder them doesn't really work if the PCs have only just met this person). I modified it by putting a stoner frost giant sage there who served as a mid-campaign lore dump and smoked with the players. 

Karkolohk

Interesting by virtue of leaving room for negotiation and also being willing to throw a large number of weak enemies at the PCs. Solid overall, though the premise is just a bit too cute. 

Lost Spire

Has the potential to meaningfully change the direction of the campaign, but Dzaan needs to be set up ahead of time. 

Skytower Shelter and Wyrmdoom Crag

These should be great, but don't really come together for me. I think it's because the goliaths don't show up elsewhere, giving them a strong presence earlier in the campaign could make these stand out. I think it's also because the more lighthearted elements of these locations clash unproductively with the apocalyptic horror of much of the rest (unlike some other lighthearted elements of the campaign, which highlight the horror). 

My Changes

I worked up a list of 20 rumors cobbled together from the existing 'tall tales' table, obfuscated the hooks, added in some entries pointing at the Ten-Towns quests, plus a couple of red herrings, and let the party roll on that table from level 1. No reason to prevent the PCs from actually hearing rumors until they get to higher level: you want to build some anticipation, let them know about possible adventure out in the wilderness they aren't ready for. 

Chapter 3: Sunblight

I've mentioned before that Icewind Dale starts out as a sandbox, and then stops. This is the turning point. 

Sunblight is a multi-level dungeon with thirty-odd keyed rooms, the hidden fortress of a duergar warlord set on taking over the Dale. 

Duergar have showed up in the adventure before, spying on Ten Towns and causing trouble. Now they are revealed to be massing in force, on the cusp of wiping out the towns entirely, and they are constructing a superweapon: a robot dragon! The PCs must go and... take on a whole fortress by themselves? The adventure states that they might want to launch a preemptive strike before the weapon is ready, which would be quite reasonable, but it's already finished, and it gets unleashed right as the PCs arrive.

The book thinks this will be an 'agonizing decision' between storming the fortress (the whole point of 4-6 homidical maniacs assaulting a literal fortress was that they can make the difference by sabotaging a superweapon) and chasing down the dragon before it can destroy Ten Towns. It's the sort of thing that would probably work at the table, but players would start questioning this series of events at the fridge later. The in-world assumption represented by the initial hook (a party isn't enough to take on the fortress, but you can stop the superweapon) runs up against the mechanical assumptions, which is that a party in the 4+ level range is going to be entirely capable of fighting the fortress by going room-to-room. Also, given the transportation options available, there's really no way for the PCs to get back to the towns before the dragon has torched a few of them, and assaulting the fortress just doesn't take that long in this system of 6-second combat rounds. 

There are a few other problems here, like the overly-cute plot about which dark god exactly is pulling the strings here and the rather unsatisfying inclusion of a faction within the duergar that... don't really do too much... but Justin Alexander has the details on those, and I don't want to just repeat his points. 

My Changes

I foreshadowed the duergar plot more, and reworked Sunblight's plot. In my continuity, the duergar king realized that this group of 4-6 nutcases were the single biggest obstacle in the way of his plans, and they could probably take down the dragon once it attacked the town they were in. So he sends some disposable agents to leak information about the dragon (they think it's still under development, but it's actually been complete for a while), thus luring the PCs, along with a large part of the overall Ten-Towns militia, out into the wilderness. The dragon gets released then, so that it can torch the towns while they are most vulnerable, and the majority of the fortress' people are already out near Ten-Towns, lying in wait for the dragon's arrival. I played Xardorok Sunblight basically as a Bond villain who waited for the players to arrive in his chamber to unleash the dragon, gloat, and reveal the dozen crossbowmen waiting in ambush. Good times. 

Jim, what's the CR on that thing?

Chapter 4: Destruction's Light

This chapter also features one of the more confusing elements of the campaign, which has earned a lot of criticism. 

As the PCs are leaving Sunblight (either because they decided to chase the dragon or because they're done with the fortress), they're approached by Velynne, a wizard on a dogsled who offers them a list back to Ten Towns to deal with the dragon. 

The catch is that the sled dogs are undead, and the wizard is accompanied by kobold corpses as well. She's a necromancer. 

As written, this is a jarring introduction to an NPC whom many players will want to kill on sight, doubly so when they find out she's a member of the Arcane Brotherhood, the faction of wizards who have been repeatedly set up as treacherous antagonists. Even worse, this NPC is plot critical and basically exists to hold the PCs' hands now that the campaign has turned into a railroad. 

I already added Velynne into the campaign earlier (in my campaign she warned the PCs about Dzaan's treachery and helped them deal with Sephek), so this problem is nullified. 

As written, this chapter kills at least a sizable portion of the Dale's remaining population and torches at least a couple towns, even if the players are otherwise maximally efficient. I have no problem with that, but this needs to be set up tonally beforehand. This ought to feel like the sort of place that's on the brink of destruction before the dragon shows up. 

Chapter 5: Auril's Abode

We come now to perhaps my least favorite part of the campaign. 

The Island of Solstice is a huge, snowflake-shaped iceberg floating in the sea, in the center of which is an ancient giant fortress shaped like a huge skull, named Grimskalle. Love it, sign me up to adventure there any day!

But, why is the party actually coming here?

Well, because the railroad says so. 

In-game, it's because Velynne wants to enter Ythryn, the ancient floating city which fell to ground and was covered by the glacier. She knows where is it, but needs a special spell to break open the glacier at the right spot, and needs a sentient magic item that contains information about the city. Both of these are on the Island of Solstice: the spell is in a book called the Codicil of White made by Auril's worshipers, and the item, called Professor Skant, was stolen by another wizard who made her way to the island and died there. How Velynne knows this is unclear, but it doesn't matter for our purposes. 

What's more relevant is that Auril lives in Grimskalle, physically, and will be there unless the party sneaks in while she's away casting her spell to shroud the Dale in night. The players can end the Rime here by killing Auril (very difficult) or by killing her pet roc, without whom she cannot fly into the air to perform the nightly ritual that maintains the endless night. Please note that we are two large dungeoncrawls away from the end of the campaign. 

Unless she has left the fortress to cast her nightly spell over Icewind Dale, the Frostmaiden lurks here in her first form, living in fear that her divine enemies will find and destroy her. She normally occupies the room to the west but lurches toward the larger room to greet the characters as they arrive. The characters have no chance of surprising her.

I actually forgot about this bit. I replaced this with my own headcanon so early in the campaign planning that I didn't realize it wasn't the canon approach, much to my surprise when I was rereading the book for this post. 

This is all very, very silly, but it gets sillier.

Recall, the PCs aren't here specifically to kill Auril, but to recover some items. The professor orb is next to the other wizard's frozen corpse in a random spot on the opposite side of the island. Not impossible to find, but a bit dubious. The Codicil of White is in the basement, behind the world's silliest security system. To get into the chamber of the Codicil, the PCs must pass various tests representing the ideals of the Frostmaiden: Cruelty, Endurance, Isolation, and Preservation. This is a neat idea. 

Here's how it works in practice. There are four doors in the basement. You can go through each one, and you get teleported to a nomad tribe on the glacier which is presently experiencing a crisis related to one of the ideals. A couple are short (one is just a fight against a werewolf), while the others take a day and a week respectively. This is, apparently, actually happening in that moment, contrived as it is, and the adventure doesn't address how this is happening or how it is possible for multiple PCs to take the test asynchronously, as the book explicitly states can be done. The scenarios themselves are equally contrived, and if there aren't enough PCs who have passed enough tests, some NPCs come by to try to take it, opening the door in the process. Making all of this irrelevant. Now that's what I call a fragile scenario!

This section demonstrates that the campaign doesn't have a solid through line. Based on all the marketing, you'd expect the major conflict here to be the Rime. It's a situation which threatens everyone and makes the whole area almost unlivable but doesn't just kill everything immediately. 

But then the duergar show up, and they have no special connection to Auril besides opportunism, wanting to take over the now-dark surface. Then a dragon destroys between a little and a lot of Ten Towns, and you'd expect there to be some focus on repairing the communities or helping them survive, but no. In the last stretch it becomes a treasure hunt without any special connection to the Rime. 

This would be fine if RotF stuck to a sandbox structure past Chapter 2 and treated the Dale as a big situation that the party can't defeat with combat (until they realize that killing Auril ends it, but that ought to be really difficult). That situation can give rise to other symptoms, like a duergar invasion. Sandbox elements like allying with the goliaths and nomads can be pursued proactively to present a united front against these threats, and treasure hunts like the search for Ythryn can be attempts at finding solutions, or finding magical power to more effectively combat these problems. 

But since the campaign drops the sandbox structure after a while, that stuff is located in new school time rather than old school space, and the PCs are shuttled from one to the other without an obvious purpose. 

My Changes

In my campaign, I just removed the Frostmaiden from this location (her presence here no longer made sense in my continuity) and replaced the 'tests' with tests of my own. The players received dreams about the various ideals much earlier on in the campaign, and I tracked particular acts of cruelty, isolation, endurance and preservation they committed across the prior sessions. They had at least one person who had done each by the time they got to Grimskalle and made their way in without difficulty. 

If I were doing a more thorough remix, this is a section I would change up in the following way:

This is only Auril's Abode metaphorically. She doesn't physically live here. Rather, this is her temple, the headquarters of her frost druid worshipers, where she sometimes makes (or made) appearances to her most devoted followers. Most of the rooms can stay. The tests are more like my version than that in the book, ritual initiations which allowed worshipers to enter the inner circle of the cult. Each requires only that a person has committed some major act of cruelty, preservation, endurance or isolation, such that players who have done these things earlier in the campaign can come together to enter the secret chamber of the Codicil, which is not 'a primer on her worship' but a record of the revelations granted to the high priests of past generations. 

"This is where the plot goes."

Next up

The next post will cover chapters 6-7 and discuss the all-important question at the heart of the campaign: what does Auril want?

If this post interested you, comment below, share it around, and subscribe to the blog. Until next time, have an excellent week.