Play Report: The Tower of Zenopus, Session 2

I ran D&D over Zoom again this weekend, with much more success this time.  I've cast my net pretty wide among my D&D-playing acquaintances, and managed to rope in some guys that I haven't gamed with in over 20 years.  Last week we had a bunch of connection issues, but this time around everyone who wanted to play was able to get on.  It was only three people, just one more than last week, but the game went much longer and saw the party experiencing far greater success.

The novel thing for me is that none of the players who showed up carried over from the last game.  This is the first time I've ever experienced running multiple groups through the same dungeon setting, and I'm looking forward to doing more of it.  This week's group took an entirely different path than the previous adventurers, so it didn't make much of a difference.  Plus the dungeons beneath the Tower of Zenopus are small, and I don't think I'll get more than one more session out of them.  But with my current set-up, where I run for whoever shows up on a Saturday night, the possibility is there for some proper old-school multi-party campaigning, especially when I start running some more expansive modules.  I can tell already that it's going to be a lot of fun.

This week's group consisted of a halfling, a cleric and a thief.  They played things smart by hiring a man-at-arms, which they were clued into by one of the rumours from Zach Howard's excellent Ruined Tower of Zenopus.  I called the man-at-arms Hew off the cuff, and had a lot of fun investing him with the persona of a simple everyman slightly overwhelmed by the situations he was finding himself in.  The players took to him pretty quickly.

The party headed east from the entry stairs, and fought a fairly easy battle against the five goblins in room A.  I used the morale rules from Basic D&D, and the goblins rolled well and ended up fighting to the death.  This went somewhat against the module text, which states that they surrender once half of their number is dead.  Ah well, you can't remember everything.  Unfortunately, what I also forgot was some of the loot: the goblins are meant to have two bags with 500 silver pieces, but I only gave the PCs one.  I think I'll move the second bag somewhere else in the dungeon, so that they can still find it.  I don't want to cheat them out of the treasure completely.

From there they went south, and it didn't take them long to figure out the trick with the doors and the statue.  Continuing south to the base of the thaumaturgist's tower, they fought and killed a giant snake.  The cleric got constricted a bit, but a single monster like this rarely fares well against a full party, and the snake didn't last long.

From there they headed west, to the room with the sundial and the bronze mask.  The mask will answer any question asked of it, but only if the party figures out its clue: "I will not speak til it be four".  The cleric player figured this out with suspicious quickness; I'd suspect him of having read the module beforehand, but he's always been a sharp dude.  I'll have to keep on my toes, it's going to be hard having a player who is obviously more intelligent than I am.

They kept their eyes on the prize, and asked the mask where they could find the most valuable treasure.  I didn't have this info readily at hand, so I had the mask direct them to the sea cave and the smugglers.  In retrospect I should have directed them north to the catacombs, but I didn't want to bog the game down while I scoured my notes.

They followed the directions south, west, and then north, having a fight with a giant crab along the way.  It lasted a few rounds due to a tough AC, but my rolling for the crab was abysmal, and it didn't score a single hit.  My rolls for Hew, on the other hand, were on fire, and he was undoubtedly the MVP of the session.  He might have made things a little too easy, but then again it could all have gone very differently if the dice had swung the other way.  One bad roll and Hew would have been dead.  Still, an easy game for the players isn't such a bad thing now and then.

From there they headed north, where they fought four smugglers.  This was the final encounter of the session, and it ended in a real anti-climax.  The smugglers failed their morale check after the first of them fell, and two more of them were cut down as they fled.  One of them escaped, fleeing north into the dungeon tunnels.  I suppose he'll still be there next session, if he's still alive.

The party found the treasure chests in the smugglers' boats, as well as a kidnapped Lemunda the Lovely, daughter of the local lord.  There was serious discussion about them taking one of the boats and rowing out of the sea tunnel, but that would almost certainly have resulted in disaster at the hands of the giant octopus lurking at the bottom.  I figured that Lemunda would have heard the pirates talking about the octopus, so I had her warn the party against it.  Sometimes the DM must be kind.

The rest of the session was spent lugging the two treasure chests out of the dungeon.  I probably should have made more of a big deal about them carrying them across a 3-foot-deep river, but it was getting late so I fudged it.  What I didn't fudge were the half-dozen or so wandering monster checks I made while they were struggling to get the chests out of the dungeon.  Not a single one came up.  I probably made about 20 wandering monster checks for the whole game, and none of those came up either.  Sometimes all the luck just runs the way of the PCs I guess.

All told, a fun and quite successful game.  The PCs played well (lots of sharp decisions, listening at doors, not lingering too much), the dice rolls went their way, and they managed to win a few fights and make off with some decent loot at little cost to themselves.  They also had some luck with the paths they chose, and avoided the most difficult encounters, any one of which could have resulted in a TPK.  Those encounters are still down there, of course, and it just makes it more likely that the party will encounter them next time.

ENCUMBRANCE HOUSE RULES

I'm still using my encumbrance house rules, and this time I actually got to put them into practice.  They worked about as well as I was hoping, and we were able to easily keep track of what the PCs were carrying and how it would affect their movement.  They even had to leave some treasure behind: 2,000 copper pieces in a chest that they found after defeating the goblins.  Copper pieces are a proper encumbrance trap for dumb PCs, being of such low value, and I was pretty pleased to see this party making the smart decision to leave them behind.  The system I've come up with isn't precise, but it's close enough that it made the PCs question whether to take that chest with them, and that's exactly the sort of thing I was looking for.

One thing I need to come up with is a rule for  characters carrying things between them.  As it is, nobody in my current rules is capable of carrying a chest with 2,000 coins in it.  I allowed them to carry the chests with one character lifting each side, and dropped all of their movement rates to 30'/turn as they struggled their way out of the dungeon.  It would be nice to have concrete rules to fall back on for this kind of thing though.

I also need to know how much weight a character can carry over the amount that drops their movement rate to 1.  I'll look into how the various editions have handled this and work something out.

LIGHT SOURCES

I was tracking light sources pretty rigorously, which I've never really done before.  The party were using torches, which burn for one hour.  Combined with ten minute turns, and D&D's frightfully slow dungeon movement rates, those torches run out really quickly.  A little too quickly, I felt, but then again I have no practical experience with burning torches.  The players didn't complain, and they were smart enough to take and use the torches that I mentioned the goblins had burning on the wall.  So I had some misgivings, but I'll stick with the rules as written for now.

COMBAT SEQUENCE

I've been using a combat sequence with group initiative and phases: spell phase, missile phase, movement phase and melee phase.  Last week I toyed with the idea of adding a second melee phase that would happen right at the start of the round, but I decided against it on the grounds of over-complication.

I'm not entirely thrilled with what I've come up with.  The main problem I'm having is that it's a bit arbitrary as to which players go first during their turn in melee.  It doesn't really matter, and didn't cause any issues during this game, but I know that there are players that are territorial about  getting their "kills".  I think I'll start using Dexterity scores, and just running down the characters in order based on that.

EXCEL GRID MAP

I mentioned last week that I was using screen-sharing and Excel to show the players the room dimensions, and to track things during battle.  So far, all of this has worked surprisingly well.  My main concern was that it's only good for square and rectangular rooms, but I've solved that problem as well, as you can see below.

Rooms K and S2 of the Tower of Zenopus adventure.

You can set background images in an Excel worksheet, and doing that I was able to mock up the sea caves and the round rooms in the thaumaturgist's tower.  It's a little time-consuming - especially getting the grids lines up correctly, but it does the job if you're not into using things like Roll20.

I expect to get one more session out of the Tower of Zenopus, possibly one more in Portown if the players decide that they really want to go hunting for smugglers.  After that, I think I'll send them to a certain Keep on the Borderlands, and the Caverns of Quasqueton nearby.  Beyond those, I'm not sure, although I'm leaning towards Caverns of Thracia and Jeff Sparks' version of the Holmes' Skull Mountain dungeon as a bridge to get them strong enough for some of the mid-level TSR modules.

Play Report: Zoom Conferencing in the Tower of Zenopus

Holy crap, it feels good to be playing D&D again.

With everyone in lockdown, I've been having a lot of Zoom conferences for work, and while I was zoning out in the middle of one I had the thought that it would be a really good way to play on-line D&D.  I've looked into Roll20, but I'm not all that interested in the fancy mapping tools and the dice rollers and all that.  What I really wanted was an easy way to set up video chat with a whole bunch of people, and Zoom fit that bill.

I'm not all that tech-savvy though, and I figured that running my first ever on-line game would be complicated enough, so I opted to run a simple system: Moldvay's D&D Basic Rules.  Of course, I spent a week drilling down into them, house-ruling a bunch of stuff, bringing in things from other editions, and filling in gaps in the rules, so it's not quite Moldvay Basic.  It's still pretty simple at the table though, especially in comparison to 3rd edition onwards.

I also wanted to opt for a simple adventure.  My plan is to run a continuity-light campaign, which will consist of a series of classic D&D modules, most of which I've never played or run.  Originally I was going to do a combination of the classic modules In Search of the Unknown and Keep on the Borderlands, but there was a fair bit of prep involved with merging those, and I was concerned that Keep would eat my players' characters alive; I've never played or run it, but it looks to me like you need a decent number of PCs to survive it at 1st level.  So I went even simpler, and chose to run the sample adventure from the Holmes Basic Set: the dungeon adventure often referred to as "The Tower of Zenopus".  I combined it with some ideas from Zach Howard's excellent Ruined Tower of Zenopus, which is a conversion of the original to 5th edition.  It has some good background material though, and some ideas for additional encounters that I liked.  I've used most of those additions, including the map of the area surrounding the tower.

I was mega-excited to be running D&D again - I barely slept the night before - but we had quite a lot of problems.  One of my players couldn't connect at all, due to some bad internet at his end.  Another of my players was on long enough to create a character, but then his connection dropped out for good.  One player had a prior Zoom commitment that ran way longer than expected (and I berated him mercilessly about it afterwards).  So I was left with two players in total, to tackle a dungeon using old-school rules...

One created a thief, and the other played an elf.  The second of those was a great choice for a small party, and he had sleep memorised, so I liked their chances of getting through at least one encounter and possibly claiming some treasure.  Alas, they turned left at the first junction, and came to a room where they were attacked by four skeletons: enemies that are sadly immune to sleep.  They were also resistant to every weapon in the players' arsenal, so the skeletons were taking half damage from every blow.  The players were quickly surrounded, but did pretty well all told, managing to defeat two of the skeletons.  Then the elf went down, and the thief decided to get the hell out of there.  The remaining skeletons failed their attacks as he fled, and he made it out alive.  (He was fleeing during a round where I was supposed to check for wandering monsters, and he got lucky on that roll as well.)  So not a total TPK, but not exactly a rousing success either.  They would have done better if the elf had remembered that he had some holy water in his backpack, so it's not entirely my fault.  Still, I probably should have advised them to hire some men-at-arms.  I even had some rules for that in my notes, but it completely slipped my mind.  My bad, I'll definitely correct that for next time unless I get a lot more players.

Everyone had fun though, and the actual play went really smoothly (as it tends to do with a small group).  Playing over Zoom worked well aside from the connection issues, and I was able to use its screen sharing and private chat features to good effect.  It helped with one of the biggest issues that I was worried about, which was visualising the dungeon for mapping and combat.  It's been ages since I've played purely "theatre of the mind", and I like to have at least some kind of map.  I decided to use Excel for the purpose, as you can see below:

The site of the PCs' downfall. Most of the action happened
in front of the top left alcove.

Each of those tabs is a different room of the dungeon.  As soon as the players entered a room, I switched to the correct tab, shared my screen, read the room description, and there was an instant visual for them.  If anyone wants to map they can do it without me tediously explaining the room dimensions or drawing it myself, and in combat you can use letters, number and symbols to show where everyone is.  Excel is great, I use it for mapping old-school CRPGs as well.  The only problem with it is that it can't draw rooms with irregular walls.  There are a couple of circular rooms in this dungeon, and some caves as well, so I'll need to figures out a solution for those before the next game.

I'll do a run-down of some of the house rules that I'm trying out.

• I'm letting the players roll 4d6 for ability scores, dropping the lowest number and arranging as desired.  I hoped it might offset the numbers disadvantage.  Besides, we've been doing it that way for decades, and it generally results in good characters but not necessarily amazing ones.  Even with this system, the elf still had a Charisma of 4 (I guess nobody will mourn for him too much).
• Clerics in Basic don't get spells at 1st level, but I did a quick rejig of their class table to allow it.  Seems a bit ridiculous to me.
• I'm using ascending Armor Class and attack bonuses, as in 3rd edition, but the math is the same as Basic.  As much as I like descending AC, the 3e system has been working like a charm for so long that I can't justify not using it.  And I've never missed explaining how THAC0 works.
• I also dropped clerics, magic-users and thieves to one level below fighters in combat.  As it is, Basic starts with everyone having the same chance to hit in combat.  I changed it so the fighting classes start with a +1 bonus, and everyone else fights at +0.
• I'm using the level limits for dwarves, elves and halflings, but I'm allowing their prime requisite modifier to allow them to advance a level or three over what the book says.
• I also gave fighters their mook sweeping ability from AD&D, and I'll let them have some extra attacks as they advance.  Although I'm thinking of maybe just rolling that into the mook sweeping ability, and letting it be dependent on the monster's Hit Die and the character's level, so that a 4th level fighter could sweep two 2 HD creatures (for example).  Not sure if that would be too powerful though.  I just feel like Basic fighters are a little vanilla.  I don't know why you'd play one instead of a dwarf, except for flavour purposes.
• Oh yeah, I upped the hit points for thieves, clerics, fighters and dwarves to their AD&D equivalent.  I also let them start at full hit points at level 1.  Sorry old-schoolers, I like to give my players at least some sort of a fighting chance.  I'm certainly not fudging die rolls in their favour at the table, so they need it.
• One of my players wanted to play a dwarf wielding a two-handed pick, so I house-ruled that up.  I didn't want it to do 1d10 damage, so I gave it 1d8, but I needed a reason for it to be a worthwhile choice.  I gave it a +1 to hit armored foes, to simulate its penetrating power.
• We did some house-ruling on the fly when we saw the absurd price for a grappling hook in the Expert Rules.  25 gold pieces?  What the hell?  It's worth more than a two-handed sword!  We all agreed to knock it down to five, and while we were doing that we knocked down the crowbar from 10gp to 5gp as well.
• I've overhauled the encumbrance rules, as I detailed in this post last week.  As I suspected, it's a bit of a bitch to set up and explain, especially over video chat.  It's no worse really than calculating regular encumbrance though, and we had accurate movement rates for both players.  It was dead easy to see the effect on the elf's encumbrance when he dropped his longbow.  Unfortunately the game was too short, and they didn't find any treasure, so I didn't get to really test how it works at the table.  The jury's still out, but I'm optimistic that I've cracked it.
• I also overhauled the combat sequence, using something similar to the one from Chainmail.  I kind of got sick of the way initiative works in 3rd edition, where one player runs a full move and then a monster makes a move, and everyone can watch while this happens and react to it.  It works, but it's a bit unrealistic, so I want to try something else out. To summarise, both sides roll initiative on d6.  There's a missile phase, a spellcasting phase, a movement phase and a melee phase, and the side with the winning initiative goes first in each (although movement happens simultaneously, which I'm iffy that I can pull off with a large group).  I've already changed this slightly based on play.  We had a situation where a skeleton burst out of an alcove to attack the thief, but he didn't get surprise.  So the elf got to fire an arrow, everyone else got to move, and then the skeleton made its attack.  I'm adding in another melee phase at the start, for those who are already engaged in melee.  I think it will work a little better, but it could overcomplicate things.
• As a part of the above sequence, I've come up with a house-rule for casting times that makes it a little more unpredictable when spells go off, and gives a chance for spells to be interrupted.  Basically, the level of the spell is the chance on 1d6 that the spell won't be cast during that phase.  So if someone casts a sleep spell - 1st level - they roll a d6 during the spell-casting phase.  If they roll a 1, the spell doesn't happen yet.  Anything else, and it goes off.  If they rolled a 1, they roll again at the end of the missile fire phase (provided they don't get shot).  It goes on down the line like that until the round is ended, and if it still hasn't been cast it happens then.  It still makes low-level spells fast, and high-level spells slower on average.  I've also tied this in to the elf's ability to cast in armour: heavier armour gives a penalty to the spell-casting roll.  So the spell will still be cast in the same round, but if you're casting in plate there's a good chance that it'll happen at the end of a round.  It makes magic armour and especially elven chain more desirable (actually, the entire encumbrance system should do that).
• Basic characters die at 0 hit points, but that's too deadly for me.  I thought about just going with the AD&D "death at -10" rule, but I figured this was my chance to try something else out.  I'm using something similar to death saves in 5e, where someone dropped to 0 makes a save.  If they succeed, they're unconscious, and if they fail, they're dying.  Once dying they keep rolling saves until dead, so it's still pretty lethal.  Bandaging can save them though, as can healing magic.  I used the Death Ray/Poison saving throw category, but I think I'm going to have to change that, because Dwarves and Halfling have absurdly good saves in that category.  I quite like it that higher level character don't die as easily, but I'm not really down with a 10th level character only needing a roll of 2.  I might have to add a separate Death & Dying category to cover this.
• Of course I'm using the standard double damage on a natural 20 rule.  It ain't D&D without it.  I've gotten rid of 3rd edition's stupid confirmation roll though, it took all the excitement out of a natural 20.  I just say that if you need a natural 20 to hit, you can't score a critical.
• Unarmed combat isn't covered much in Basic.  I made unarmed strikes do 1d3 damage plus Strength bonus (with different die rolls for creatures of different sizes; Small creatures do 1d2 for example).  The damage is non-lethal, but the only difference that makes is if you drop someone below zero hit points: you can't kill someone with non-lethal damage unless you drop them to -10.  This allows a strong person to kill a weak character with a single punch, which I'm down with (or for a character to pummel an unconscious foe to death).
• I also covered grappling, which I'm doing as a hugely streamlined version of 3rd edition grappling.  It's just a series of opposed attack rolls, basically.  Monster's don't get a Strength bonus, but they get hefty size bonuses (or penalties if they're little).  A halfling, say, would be at -4, and an ogre would be at +4 (plus another 4 for Hit Dice).  The halfling might get a win here and there, but on average he's going to get creamed by that Ogre in a grappling contest, unless he's higher level.  (Actually, a halfling tops out at 8th level, with +6 to attack.  Factor in that penalty, and the maximum Strength bonus of +3, and the best bonus he's going to get is a +5.  He'll never be good enough to grapple that Ogre without magic or extra numbers, and that works for me.  After years of 3e, it's nice to go back to a system with a flatter progression.)

There's other stuff, but those rule changes are the bulk of it.  It's stuff I've been thinking over for years, while I've been digesting other D&D clones, old-school blogs, and the original D&D rulebooks.  My ultimate goal would be to have a ruleset that can be expanded or contracted  to mimic every D&D edition up to 3rd - the fabled "D&D Nth Edition" that I've written about before - but that's probably a design goal that's beyond me.  For now I'm happy enough to try out the ideas I've been working on for years, and give them a proper road test.  And more than anything, I'm happy to be playing D&D, and not just pontificating about it.

Play Report: Journey to the Sandbox

So I played some D&D yesterday, the first game I've had since... 2011? Can that be right?  Alas, yes.  It's one of the perils of growing up, and not really enjoying gaming with people I don't know.  It's left me with a small playing group of people who have very full lives, so chances to game have been scarce.

As to the game itself, it's a continuation of a 3rd edition campaign that's been running on and off - mostly off - since 2004.  The PCs just wrapped up all of the adventure hooks and subplots in the initial area (a wilderness fortress) and the goal for this game was to get them to a new area, the largest city in the world.

I haven't played much in recent years, but I've been thinking and reading about the game a lot, and I wanted to transition away from obvious pre-planned adventures and more into a sandbox style game.  So I mapped out the city, the catacombs beneath and the wilderness.  I came up with random encounter tables.  I wrote brief descriptions of a bunch of other adventure sites that the PCs could pursue or ignore if they wished.  To be honest I felt a little underprepared, but I'll have to get used to that if I'm going to keep things open-ended.

The first surprise to me is that the whole session went by without any combat, and barely any dice-rolling at all.  I've never experienced this before, and after the game I felt a little guilty about it.  There were certainly opportunities for conflict, and one scenario was right on the brink of it, but it just didn't happen.  The players who showed up were the cautious ones.  I probably shouldn't worry about it, because everyone was engaged and having fun, and interested in exploring their new home base.  But it doesn't quite feel like I gave them a real game.

In addition to that, I could see them struggling with a lack of obvious hooks.  I'm hoping they'll adjust in time, and become a bit more proactive.  I won't rely on it though, and for the next game I'll give them a choice of maybe three obvious things to do.  I've also told them outright that I don't care where the game goes, and that they should pursue their own goals both long term and short term.  I'm confident it'll work out, but it's going to be an adjustment.

The only other issue we had was a small bit of debate about how the spell detect thoughts works.  Does the target know that their mind is being invaded?  In this case the spell was cast right in front of the target, so I ruled that he knew.  In other cases, where the target hasn't seen any spellcasting, I'd rule that they don't know.  I need to look into this to see if there are any specific rules, but it's not a huge deal.

Other than those misgivings, it was great to be back in the saddle, and I'm looking forward to more.  The current plan is to make myself available to play once a month, regardless of who shows up.  The groups might be smaller, but I'll settle for it.  I'm not having another 6 year hiatus, that's for damn sure.

Play Report: A Siege, the NPC Railroad, and Problems With Resilient Sphere

Miracle of miracles, I have managed to play some D&D.  This has been a long time coming, with a good three or four still-born attempts to organise a session since the last game, but we finally made it.  It looks as though the next game might be fair way off for a completely different reason: the PCs will probably move on to new pastures, which necessitates me designing a whole new adventuring space.  That's a different post, though.

The game went a lot more smoothly this time, as is to be expected when the time between games shrinks from three years to four months.  I didn't burn out this time, as I had made sure to get plenty of sleep the night before, and there were no non-gamers wandering about to dissuade me from general DM acting silliness.  I felt a lot more comfortable running things, and that's an encouraging sign.

The game resumed with the PCs still in a castle under siege by an army of orcs.  They had recently resurrected some serious heavy-hitting NPCs, the sort of guys that I could not justify taking a back seat, and I was sure that this was going to be a problem.  The PCs had been in charge in all but name before this, and I was concerned about leading them by the nose and having NPCs giving them orders and suggestions.  On the other hand, it was their own choice to bring these guys into the game.  If you resurrect a legendary hero, you have to expect that you're going to take a back seat.  Resurrect three, and you're lucky to have a walk-on role.

I opened the game with a warm-up battle against nearly forty elite orc warriors, dropped on the fortress roof by dragons.  This was a little more time-consuming than I would have liked.  That's what happens when you generate the numbers with a random roll, then roll high.  It was a good hour of grinding through orcs, but I have always found that a battle eases me into the game very nicely.  We got through a decent amount of other stuff this time around, so I wasn't too bothered.

At this point, after the characters had rested and healed up, they actually got to the business of figuring out which of their powerful items could break the siege.  They have an arcane warsuit, a sort of mech loaded with magic wands, but that wasn't powerful enough to tackle a whole army; a sphere of annihilation, also impractical against so many foes; the Skull of Vecna, which could be used to create and control vast numbers of undead (and with the Eternal Battlefield, where skeletal armies of the past fight endlessly, just miles away); a barrel of distilled dwarven fire oil, the closest thing to a WMD this society can muster; the Hand and Body of the Light, actual fragments of this world's dead sun god; and lots of other goodies.  In the end they opted for the use of the Ram's Horn Staff, which had the power to animate the trees of the forest to do the wielder's bidding.  It was a good choice; even though the trees were eventually beaten back by a barrage of giant-thrown boulders and dragon breath, they destroyed a third of the army, and netted the PCs a lot of Victory Points.  (I was operating on a Victory Point system here, whereby various actions taken by the PCs or NPCs would gain or lose points.  The target for reaching the endgame was 100 points, and this single action got them to 50 all by itself.  It was perhaps too many points awarded for the risk to the PCs, but I felt justified in it given how deadly the adventure where they gained that staff was.  Blink dogs with levels of rogue and evil elves who can use true strike once per day are serious business.)

Following that, the leader of the army showed up, an orc called the Reaver who was thousands of years old and wielded two axes of sharpness.  He wasn't in my original plans for the siege, but I had noted him down as the most powerful figure around in my campaign world some time ago.  So when his ancient enemy King Peramis I comes back from the dead and shouting his name from the top of the battlements while waving severed orc heads around, I figure that I'm justified in having him appear.  He challenged the king to a one-on-one duel, which was accepted.  As a way to get the PCs involved, I had another NPC suggest to them that they try to find a way to break the rules and kill the Reaver before the duel can be finished.  This worked pretty well.  They had a vial of poison, made up of a few drop's of heart's blood from the god of evil, and given that the Reaver wore no armour they had a good shot at killing him outright.  If only the elf had not stopped to cast true strike, King Peramis may have survived instead of losing an arm and then a head.

(As a side note, it was only during the game that I noticed there is no equivalent to the sword of sharpness in 3rd edition.  Yes, there are vorpal weapons, but I didn't quite want to go that far.  I fudged it by having the axes of sharpness sever an appendage on a successful critical, which worked well enough.)

(As a second side note, the PCs are seriously lucky that they killed the Reaver here.  I had no qualms about throwing a divinely powered 20th level fighter at them at this stage of the campaign, and had he made it inside the fortress there would most probably have been a bloodbath.  I was looking forward to it, but alas.)

The killing of the Reaver pushed the PCs over 100 Victory Points, at which point the orc army went bananas, the Reaver's lieutenant lead a flight of dragons to attack the fortress, and a smaller dragon force tried to retrieve the Reaver's body.  There was fighting on two fronts, with some PCs defending the fortress while others tried to retrieve the body of King Peramis (not for altruistic reasons, but to loot his corpse).  At this point I would like to point out (to the surprise of nobody) that dragons are tough.  These weren't even true dragons; they were big and they had breath weapons, but they had little more intelligence than animals.  That two claw/two wing/bite attack routine is murder, though.  There were 4th level NPCs that were getting chewed up at the rate of two a round.  The Dwarf barbarian went from 80-odd hit points to under zero almost instantly.  Again, the PCs were saved because I have been super-gonzo with the magic items recently.

During all of this, the main NPC antagonist (named Elmyr) tried to make his escape.  He was an ally of the people who lived in the fortress, but he had been needling the PCs for a long, long time.  I was working up to a final confrontation, but it got lost in the siege and the arrival of the super high-level NPCs.  It couldn't have worked out much better though.  He summoned his erinyes ally, and together they tried to fly away.  One PC tried to capture them with Otiluke's resilient sphere, and here I had to make a few judgment calls.  Can this spell enclose a creature in mid-air?  I ruled that it can, which I usually do when presented with something that isn't in the rules.  Then the question arose: does the sphere stay suspended in mid-air?  And if it falls, are those inside hurt by the impact?  I ruled that the sphere stays in the air, as I didn't want to create a precedent that could cripple airborne foes in the future.  So the PCs had five minutes to figure out how to deal with those trapped within, which they did by preparing another poisoned arrow.  One shot and the erinyes was dead, while Elmyr fell 200 feet to the ground.

At which point everyone declared him dead, but just out of curiosity I rolled damage by the book.  He survived with three hit points left, which gave him a chance to surrender, and me a chance to have him explain his motivations a bit.  Best of all, the PC that he landed next to was the one that he'd had the most friction with, and he got to deliver the killing stroke.  I couldn't have planned it more perfectly.

It was a fun game, with I feel just enough chaos to offset the NPC railroad that had come about through no planning of my own.  I have problems for the future, though.  Besides having to design a whole load of new adventuring material, I have a party averaging about 7th level with an absolute shit-ton of powerful magic items.  It was a great thing to have when running an epic finale for the first stage of the campaign, but I know it's going to cause me headaches in the future.  I'm thinking I'll just leave it as is for a while and see how things go.  I would have frowned on this kind of magic-heavy party a few years ago, but my attitudes are a lot looser about this sort of thing now.  As long as all parties are enjoying themselves, and the game is still a challenge, things will progress apace.