Main menu

"Goodbye, Jean-Luc, I'm gonna miss you. You had such potential. But then again, all good things must come to an end."
- Q, Star Trek: TNG

The Asgard Project: Debunking the Myths of High-Level D&D 3.5

by Ken Newquist / August 14, 2008

High-level play within D&D 3rd Edition is hard. Whether you’re playing 3.0 or 3.5, the end result is the same: thousands of feats, hundreds of prestige classes and gods-only-know how many spells give rise to complicated game mechanics that slow play to a crawl.  Iterative attacks, in which high-level martial classes like the fighter or ranger get four or five attacks every round add to the complexity as people calculate to hits and damage … and then have to do it all over again when they remember to factor in some party-buffing spell the cleric cast last round.

But is it unplayable? Or has everyone simply assumed it is?

Weaving the Myth

The myth of high-level play has been festering in my group for years, fed primarily by time-consuming expeditions into Maure Castle. The death-trap filled dungeon crawl saw lots of combat featuring 12th to 15th level characters, and consisted of the heroes spelling up, teleporting into the dungeon, battling some monster (or group of monsters) for three or four hours, looting the corpses and then fleeing back to their headquarters.

Combats took forever, some folks felt left out when their characters had no way of bypassing a given demon’s defenses, and the end result was a game that felt more frustrating than fun.

At the same time though, our 11-year-old campaign has spawned plenty of characters in the 13+ level range, as well as a goodly number in the 18-20th range. These are characters that everyone is eager to play … if we could just get high level combat under control. For a time at least some of us hoped that 4th Edition would be the answer, but I have doubt we’d be able to convert our high level characters to that edition and be happy with the end result.

This got me to thinking about how we could improve our game. I cast my mind back to other high level adventures we had run, outside of Maure Castle, and I came to the realization that we weren’t nearly as bad at high level adventuring as we though we were.

Indeed, almost every high level adventure we’d run in the last five years using the 3.x rules, be it a battle against frost giants in the Night of Storms, the assault on the white dragon Skel’s lair, the fight against an ancient mummy lord come to seek its revenge against the Blackrazor Guild, had gone off well. It wasn’t just that combat ran smoothly, it was that people enjoyed themselves, and we got to tell some memorable stories.

Identifying the Problem

The problem, I’ve come to believe, is three-fold.

First, we’ve let ourselves believe that high level play under 3rd edition is next to impossible, and thus we’ve avoided playing it in our weekly game. This is a self-fulfilling prophecy; we think it’s impossible to play, so we don’t play it. Thus, it becomes, by definition, impossible for us to play … regardless of whether the rules are really problematic.

Second, high level play is harder. 3rd Edition characters can be tremendously complicated at high levels, but that’s what makes them so fun. The challenge lies in knowing your character’s capabilities, and being prepared to act on them. For martial classes, that means knowing your feats and doing the combat math ahead of time. For spell-casting classes, that means being familiar with the spells you’re likely to cast and prepping their durations, DCs, and damage dealt ahead of time.

For all players it means using your time at the table wisely. You need to pay attention to combat, and if a situation comes up that you’re not prepared for (perhaps because of an opportunity to use a spell that you’ve never cast before) then you need to read up on that description before your turn comes up.

It also means that you need to practice. High-level play opens up new opportunities and new challenges, the likes of which both players and DMs haven’t seen before.  It’s a learning experience, and learning takes time, so we shouldn’t expect to run lightning-fast high-level sessions after only one or two engagements.

Third, game pacing is critical. Dungeon Masters need to keep things moving. Prior to the game they need to come up with rudimentary battle plans for the NPCs and monsters they plan to use during the session. Just like players, they need to make notes about spell and feat effects, damage dealt, and DCs. Everything they can look up outside of the game is one less thing they have to fumble with in-session … and one less thing that can slow down play.

When planning encounters, DMs need to make sure that everyone has something to do. If the Big Bad Demon has super-high spell resistance that is sure to stymie the mages, then provide some mooks that the wizard can blast away with chain lightning bolts. If the lich lord has ungodly damage reduction, make sure the fighter has some skeletons to cleave through. It’s not about throwing the players softballs; it’s about making sure that everyone has something to challenge and engage them.

Finally, DMs need to watch the pacing of the game. In our group, we call this the “fast combat” style of play, in which everyone is expected to make a decision in 10-30 seconds, with an understanding that play will move on if they can’t decide. Likewise, we ignore rules debates; if we’re not clear on a rule we ad hoc something that seems fair and debate the minutia after the fact.   DMs also need to build breaks into their high-level sessions, especially before and after a big fight. Before a fight players may need a little time to study up on their characters abilities. After the fact, everyone needs to blow off steam and geek out about what happened before.

These natural breaks in the flow of the game keep everyone loose and relaxed, and help prevent frustration with the rules (and let’s admit it, with each other as well) from spiraling out of control.

Forging a Solution

My group has spent countless hours debating 3rd Edition vs. 4th Edition. One of the recurring threads thats emerged from those discussions is that none of us want to give up our high level characters. We may never play them, but damn it, we don’t want to give them up.

Rather than spend the next 10 years complaining about what could have been, I resolved to do something about it. Thus I’ve proposed the Asgard Project, in which a small number of volunteers from my group will meet to play one-shot high level adventures. These adventures will serve as playtests to identify problem areas with high level play in 3rd edition, come up with possible solutions, and then bring those back to the larger group. In addition to identifying the areas where we run in to problems, this project will also serve to teach us how to run high-level adventures.

When the project concludes, we’ll know where the pain points are, and hopefully learned a lot more about how to run these kinds of high-level adventures. That in turn will hopefully lead to a renaissance of high-level play, allowing players to dust off some of our campaign’s signature characters for a new round of epic adventures.

Naturally, I’ll be blogging about all of this on Nuketown, so look for more posts about the Asgard project as we roll into the fall.


Excellent! Your observations are spot on. Right now my group is around level 14 and my current beef is the Mystic Theurge with his huge spell list being unprepared. We already use reference sheets with one line per spell, but it seems to me that what we really need is three or four sets that he can pick up without pouring over his spell list for ten minutes. We're already making jokes about the players of spell-casters having to come 20min early in order to prep their spells. :)

Thanks! Yeah, you definitely need to shift your habits when you get to higher levels. That Mystic Theurge may have been able to muddle his way through his spell lists at lower levels, but once you get to 4th, there's just too many options to not be prepared.

And although this kind of assumed in my post, it's something you need to talk with your group about and worth with them to come up with a plan of attack. When we ran our high level playtest in July, people really did show up 30 to 60 minutes early to review their character sheets, look up spells, etc. We reached a consensus, as a group, that we would work to speed things up ... including using "fast combat" and retaining the right to skip people if they take to long.

I think it's also important for high level spell casters to have "go to" spells, a small subset of spells they know they're going to use in combat that they're completely familiar with (e.g. maximized fireball or quickened magic missile or whatever). This way that have something to fall back on when the GM says "next!". Too often with high level play, I think players (particularly spell casters) spend too much time looking for the perfect spell or magic item for a particular action. Rather than perfect, they should be looking for merely "good".

Good post.

For me, the problem with 3e at high-level wasn't the play, but the preparation. Combats we could handle, and I found that even the most complex characters can be run by a capable player because it's the only one they've got to deal with. As a GM, it's down to me to prep the game though, and that gets more and more time consuming as the levels increase under the 3e rules. And as for generating classed NPC villains......... ick. I don't want to have to spend over an hour generating the Big Bad who's only going to be around for a few hours of play-time any more.

Thankfully, 4e has gone a long, long way toward fixing the prep-time problem at all levels of play. I'm finding it a godsend in this respect :)

Yes, the killer is prep time. That's something I plan to get into in the session report for the first playtest -- high level 3.5 stuff takes hours to prep for. I don't mind NPC creation so much -- I use helper programs to speed things up there -- but as I mentioned in the 3e vs 4e prep-time column, it's looking up all the individual powers and abilities that really kills me.

I've been looking over the Pathfinder alpha rules, and I think they're moving in the right direction. The simplified skill system makes it easier to create NPCs by hand, and I like how they've given suggestions for feats based on NPC role. I also like the guidelines for magical armor, weapons and such; you don't necessarily need it, but it's nice that it's there.

While I did find that my 15th level 3E adventure took a while to put together, I also want to try it again when I'm focused. I had Eureka on in a continuous loop while prepping for that game, and I think the lack of focus stretched things out. Two or three hours in my office working exclusively on an adventure would probably be much more productive (and faster).

That said, running a weekly high level campaign would be damn hard for me (at least right now, with two young kids, the day job and the freelance stuff). Doing the same in 4E would be considerably easier.

Honestly though, I'm willing to put in the prep time to make this work if we can get the game running smoothly at the table, and everyone has fun. If I put in 6-10 hours of prep time, and everyone walks away frustrated, well, that's just not fun.

My problems with 3.5 high-level play are two-fold:

1. "Instead of just giving you treasure, here's 20,000 gold pieces. Buy whatever you can afford." As a rogue, I needed more specific gear than usual to stay viable at high-levels: concealment to avoid getting hit (because your AC doesn't mean jack), item powers to sneak attack all the undead/golems/etc., the usual stat-boosting items, and a variety of scrolls/magic items to UMD with. I've spent 5-6 hours preparing the items for one high-level PC and still be scared wondering if I had missed something vital.

2. "You didn't cast Protection From Evil before you entered? Your warblade is possessed, turns around, and slaughters the rogue before he can blink." We had one prepared; we had just forgotten, in the preparation morass, to cast it; as a result, one PC was put out of commission & another PC was killed outright before the first round of combat. We had to cast 4-6 buffs before entering the dungeon to be effective, and woe on the party that chooses the wrong spells or (heaven forbid) has an absent-minded player on the wizard.

The end result? We trounced a cult boss with relative ease when we prepared for him ahead of time, but a lone ice devil random encounter that surprised us (no anti-devil spells, no frost resist spells, no holy weapons) nearly killed the entire party.

"Instead of just giving you treasure, here's 20,000 gold pieces"

I think this has been an overall problem in 3e to some extent. By finally providing actual item creation rules the players are implicitly encouraged to produce the equipment they need, especially when you consider the interplay of feats.

How many times have we heard variants of "that's a wicked cool magical spear but all of my feats are in bastard sword. How much can I get for it" quickly followed by "who can I get to further enchant my bastard sword". :)

I think it falls to the DM to counteract that a little bit. If you've got a high level rogue in the party, put in some treasure specifically for him. If your big fighter uses a bastard sword, but the enemy is using a wicked cool spear, change it to a bastard sword. :)

I'm not faulting the DM (certainly not OUR dm, because I like treasure and don't like dying. :) ). I've been known to use random treasure generators, too. But throwing in a personalized item here and there really helps to keep it feeling like .... well, treasure!

But you also have to have the right group. For years we had a player who just won every treasure roll he made. It was infuriating, and 'specialty' items for players almost never went to the correct player. And there was another who argued for every concession and extra share of treasure he could. Well, several like that, actually. :)

So an easily divisible sum of gold works much better in THOSE circumstances. But if everyone is pretty agreeable, I think it works best to put in specially tailored items.

A couple of thoughts here:

1) Regarding player characters being over-prepared for an encounter: if they did their homework, I'm cool with that. In a big uber boss fight, I think it's as fun for the players to kickass as it is for them to be super challenged. If they've taken the time to research their enemy, and prepare accordingly, I'm cool with them annihilating the boss. It can be a little frustrating ... but that's why the boss usually has a surprise or two of his own in reserve.

2) Regarding unprepared characters: that's always a challenge at high level in 3E. I think this goes away somewhat in 4E because you're not spending as much time preparing spells and such, your abilities are almost always useful, and high-level enemies and there aren't as many special defenses to counteract (e.g. where's my cold iron long sword? In 3E, I think this is a question of adventure design. If the players are going to the Paraelemental Plane of Ice, I'm not going to throw an Effreti at them just to screw with them because I know they won't have any fire magic up. I think this comes back to my earlier point of "know your party". I'm generally not in favor of random encounters to begin with, but at high levels IMHO every encounter needs to be planned, and part of that planning is knowing what your group's capabilities are.

That said, if the party is aware of a situation, and chooses not to prepare for it, well, then I say the demons should have them for lunch.

3) Regarding magic items: With high level characters (and low level ones for that matter) I try and include a good mix of magic items, including ones that only particular characters can use, as well as multi-use ones that everyone (or most everyone) can rely on).

I'll need to double check the Pathfinder rules, but I think they're going with a weapon group approach to a lot of combat feats, which should help with the "hey, how much can I get for this fiery burst dire spear +3 that I can't use?". Also, I wonder if increasing the feat progression (granting feats ever other level instead of every third) might also help with this by giving the martial classes more feats to play with (and thus, the possibility of being able to use more treasure out of the box).

One aspect of 3.5 that I don't like, and has been problematic for the group in Maure Castle, is the whole "golf bag of weapons" problem where players take to carrying one of everything in order to be able to affect high-level creatures (which typically have an assortment of special damage reduction capabilities).

I think this is something that could be handled through mid-to-high level feats ("exploit weapon" perhaps?) that would allow them to emulate certain capabilities. Or perhaps some uber weapon trait that allows it to bypass multiple forms of damage reduction (e.g. the "glorious" weapon is lawful and good aligned for purposes of damage reduction).

This is sort of a reply to the whole thread here in general. I have a group I play with every weekened and our sessions last about 5-8 hours depending on work. We've been playing together for about 15 years, so, as one would suspect we've gone epic level with campaigns several times (into the 30s a few times). We have three DMs in the group of 5-6 people playing so we're constantly cycling through DMs in different homemade campaign worlds. This allows for two to be building campaigns while the other DMs. That takes a lot of work off the DM because a lot is planned before hand. Also, we always use the same three campaign worlds. We play 3.5 by the way. We have a table (4'x6') with a plexiglass top with a grid underneath, and shelves and drawers all along the walls around us lined with about 4500 DnD minis easily within reach. We deligate responsibility by allowing some to take care of recording treasure, some to keep track of initiative, and one to take care of the 'kill box' so that after we just look through the box to see what died in the session. The room we play in is the DnD Room and is used for nothing else. Makes for dropping a session in the middle of something a lot easier. We have a standing rule that everyone figures out what they're doing before their turn comes up so we move through combats pretty fast. That being said, we have had 2hr+ fights using the entire table, but most of the time, those are strategic fights requiring lots of thought and end up being well worth the time. Other sessions we have no combat at all and those are just as rewarding. I think as players reach high level combat becomes less important as long as the DM allows the PCs to take part in the world at large. I've had my own world changed drastically by PCs and I love it. Let them delve into the intricacies of your world, let the game move away from numbers and preparation, and see if its more fun. I think it is. We never really have problems with prep time, but I think its because we drop into character upon entering the room and are totally into it all the way through.

For the magic items that people couldn't use, I had a small solution to that. Instead of having the loot be magic weapons, have it be a weapon upgrade : instead of the dragon's horde containing a few magic weapons the PC's can't use, instead I could put in a few glowing gems (or something similar) that the party's wizard could use to enchant magic items (with each item being usable only once). Not an awesome solution, but one that worked very well for a campaign of children playing mid-level characters.

I have been DMing the same game for 10 years and I have a couple ideas that I think will aid any highlv game, first any one that powerfull shouldn't be bound by fate the introduction of points to be used To avoid those little insta death spots and to aid them in Magor combat moments. 1fate point per 5 Lvs or what ever makes your fancy, and the dm can make up his own mind as to what the fate point does. Second, increase death by massive damage in enemies, Crits auto kill if the ecl of the monster is less then half that of the person dealing out the blow. Third, and this is my fav becuse it's so stupid simple, take off caps I.E. The once week Ray of frost that now does 22 to 23 dice of damage. And four dice maltiplucation, when The 32 lv ftr hits with his flaming shocking +12 full blade of wounding insted of haveing 20 sum odd dice bouncing around table just role one and times it by the ammount

I'm 33yrs old and have been playing with the same core group of 4 others since THAC0 was around. I have to agree with you spot on. I DM about one large campaign a year and split DM duties with one other in the group. We play once a week in a dedicated game room and combat typically looks like a game of warhammer with the fully crafted and modular environments and a large selection of minis that are organized into trays and tackle boxs that are labeled. Set up takes a minute but gives everyone a bit of prep before battle wich is nice. I find the harder you try to control a high level campaign the harder it is to keep a handle on it in 3.5. I haven't. Tried the parh finder rules set at high levels yet because we are still enjoying all the awsome adventure paths available but with 3.5 the issue for me is that after lvl20 leveling up just isn't that big of an impact on PC strength. I find that story telling becomes more important and I take a more free form approach with the rules. All treasure is specific when it comes to items but most of the time I don't write it down because at this point I know my PC's and I know what I want them to have in my head. Same with encounters. I'll do stat sheets for major NPCs and bosses and mid bosses but everthing else I go straight out of the monster manual. Know your players know your story and minimize your reliance on random generators unless an action is being performed. Random tables waste time you know what you want so do it. Also we cut sessions to one 5-8hr session once a week. My players do all there home work in between because there stoaked about next week and come ready to contribute locked and cocked. Mechanically I like the fast game play concept mentioned above and it makes a lot of sense that if your playing high level characters that you should know the game well enough to make a call. I mean in game 1 turn is what about 6 seconds? Why should I give you 2-3 mins to make a decision when you have all that time between turns to mull it over? It really all comes down to knowing the game(most of us have been playing 3.5 forever) and keeping it simple as possible. Its all about telling a good story and letting the players tell there story first then the mechanics can come in and make it realistic. If your killing characters all the time especially at high levels your doing something wrong. If combate rounds are takeing forever and people are feeling it then make adjustments like rolling intitive once for the whole encounter or roll all hits at once and all damage at once. Don't be a rules stickler just keep it going. The player will add all the excitement you need. Also idk about gear hoarding but portable holes and handy haversacks still do the trick at high levels for me. And really you know what. Your guys are packing don't throw wrenches at them by hitting them with chalenges they aren't prepared for. It just hurts the story in the end.

Also I never just start a high level campaign. If I know I want a game to go high(because of the CR of my protagonist) I'll still start the campeign around lvl5 or 6. This allows players trying an unfamilar class or an new build to ease into it with a few options at a time so by the time there up there in levels they know how to run there characters properly. Also if a player wants to play a character class not from the PHB then I expect them to know that class intamently. You should also be familiar with this class so there are no suprises. Prep time is getting an awful lot of flak in these blogs but honestly if your not familiarizeing your self with the basic core rules on a weekly baseis your wrong. If your haveing trouble with a certain mechanic of game play modify it or change it all together. The rules were ment to give you options not to be set in stone. There there to simulate basic physics within your world house rules can solve a lot of these problems your haveing with alittle trial and error if you have enough patience to think about it between sessiond and can ask your self honestly what's broken how would I fix it? I make calls on the fly all the time and my players know that hey josh is fair so after 18yrs of playing together they know to just roll with it but knowing the basic combat rules for most situations by heart goes along way. Knowing your source material and were to find the info you need is paramount to running any game. That's just being a good DM. Of course a good DM screen goes a long way to make your job easier(the path finder screen is awsome the best I've seen in years) and a good dry erase board to track combat and condition and status mods so everyone can see helps a lot too. You have to control the flow of information dureing combat and this is works for me.spell templates are great to just throw down instead of counting squares but a compass works great for AoE spell like fireball and you can use a protractor to figure cones in seconds. Its the little things that make it manageable but nothing beats just knowing the rules D20 hasn't changed much in over a decade for us old timers so we don't do as much shuffleing through DMGs and MMs because we've looked up the same stuff so many times. Truthfully if I have a brain fart at the table 9 out of 10 times my other DM knows the answer or were to find it so I'll put her to work but it doesn't happen often.

It's a bit late, but know that while I played 3.5 and 4ed, I did continue my main campaign as a DM with 2ed, where players of 3.5 come also into play (with some minor tweeks, it can be done, because 2ed is very versatile). We started 23 years ago. It's been around 10 years my players are on epic levels, and we still play and have a ton of fun.

So i'd like to tell you that indeed the system in itself (3.5) is a big disadvantage (while lower levels are more streamlined and functional).