ometimes in reading through blogs and forum posts, I see people wonder why we’re not making faster progress on the game. Realistically, if we wanted to hurry things along, we could have all the classes and races out in playtest form, hordes of monsters, and piles of treasure. Instead, we’ve seen six classes and four races, and I can assure you that we’re a couple of iterations away from seeing more. So, what gives?
We know based on surveys we’ve done and data from the 4E online tools that the four basic classes are by far the most popular options. The more we can do to get them right, the happier everyone will be. Those four classes also cover a far wider conceptual area than our other classes. A lot of character types live within them, so they need to be more flexible than the other classes we’ll create.
Also, there’s something that I’ve noticed in working on anything creative. The more time you can spend actively working on and thinking about something, the more likely you are to have a eureka moment that leads to a big breakthrough. If you want to become a better distance runner, you start running. By the same token, the best way to produce a better fighter is to keep plugging away at the class, trying out new ideas and either embracing them or setting them aside based on how they work out. For every personal best time you set while running, there are plenty of days that you don’t come close to setting a new mark. Game design is similar. For every idea that ends up in the packet, we have dozens that never even make it past a conversation.
So, how is this really interesting beyond maybe justifying why we don’t have a monk or ranger in the playtest? Well, I wanted to share some basic concepts with you that might not end up in the game, but are hopefully an interesting experiment. They relate to spellcasting and how it functions.
Systems or Classes?
Last week I wrote about the results of our playtest, and I wanted to save specific comments on the warlock and sorcerer for a later date. There’s still more to talk about regarding those classes, but this week I want to look at this idea: each class embraced a different spellcasting mechanic. Our original idea was that each class would use magic in a different way. Trying to cram several different magic systems into one class proved problematic and confusing. The thought was that you could use the classes to explore different styles of magic.
While we were working on the classes, the concept of using different styles of casting in one class remained in the background. You might have noticed that the wizard hasn’t changed too much between the different packets, since the arcane traditions we wanted to add to the class have taken some time to coalesce. Then, at one point, we decided to try using the traditions to embrace different casting mechanics. I actually went as far as to write up a tradition that used spell points and another that used the 3E sorcerer’s mechanic. That approach proved to be a dead end, since once again we were left with a class that had sprawling mechanics that could prove problematic in play.
It seemed like we had an intractable problem on our hands. How could we possibly present multiple spellcasting options for one class without turning the class into a Frankenstein’s monster? The answer actually proved fairly simple, but it required a change of perspective.
Rather than put multiple casting mechanics into a class for a player to choose, we could simply move those mechanics to the DM’s side of things. After all, a magic system is big. It defines part of a fantasy world, and building the world is mostly (in many groups, entirely) the DM’s job. Why not let the DM pick and choose, and then make those options available to the players? A player who wants to use a specific system could just ask the DM, in much the same way a player might ask to play a lizardfolk wizard or a warforged character in the Greyhawk setting.
This shift solved our seemingly intractable problem. All we have to do now is create a core magic system that focuses on a set of key concepts that all the systems would have to cover. For instance, all the systems have to work within the idea of spells and spell levels. We don’t want to recreate the entire spell list for each system. The systems would also need to handle at-will spells, signature spells, saving throws, and so on—game elements that the wizard and the arcane traditions refer to. In essence, this approach embraces what we want to do with modularity, where we build a simple core that relies on a few key concepts to function. As long as an optional system plays correctly within those assumptions, swapping in one system for another or even using multiple systems in the same campaign works smoothly.
The only thing that changes is the table for the wizard’s spell progression. Using spell points, slots, or traditional D&D casting does require some modification, but at the table you could run several casters, each using a different system, and the game functions fine. Those casters could share magic items, trade spells, and so on, without any issue.
As part of making sure that the modular approach to casting is easy to use, we’ll provide world background and flavor for each system. This approach makes it easy to use all of the systems at once in your campaign—perhaps competing empires or arcane academies use markedly different approaches to magic—with a minimum of work. If you want to reskin things or change the story, you can do that if you want to.
This breakthrough took place recently enough that you won’t see it in the very next playtest packet, but you should see it in the one following ones. If we do a good job with it, it should also provide a good model for what rules modularity will look like in D&D Next.
Mike Mearls is the senior manager for the D&D research and design team. He led the design for 5th Edition D&D. His other credits include the Castle Ravenloft board game, Monster Manual 3 for 4th Edition, and Player’s Handbook 2 for 3rd Edition.