Individual dwarfs at the same time have a name (their identifier), and their own traits which are randomly selected from a set of traits available to their entity, but not necessarily common to all dwarfs. So, you never end up with a dwarf who can fly, but you do end up with a dwarf that likes cows for their haunting moos and hates his neighbor Urist McJackass because he keeps a pet rat.

The real fun comes from giving the user a way to leverage this system to their advantage, or even actively manipulate it. Again to the dwarf fortress example, you can draft dwarfs into the military who have the appropriate traits (toughness, lack of empathy, strength) to create a military juggernaut that clears the fortress of all threats swiftly. At the same time, it is just as much fun to give the clumsy dwarf with poor eyesight and spatial sense an adamantine battle axe and have him spar with all the dwarfs you deem “unfit” for your fortress. As an added bonus, it gives your surgeons something to do!

I realized this morning that I forgot to mention the idea of social stats, which would be available only to the main character, and which would be used primarily to unlock more interesting dialogue options (which would, of course, have to lead to interesting consequences, or it’s a waste of time). As with so many other things, a full post on this subject is forthcoming.

I’ve also already begun to reconsider the trinity of attack stats, at least in their naming, and am considering calling the Agility stat something like Cunning or Finesse instead, the idea being that these are characters that use subtle, precise attacks to deal damage, rather than Brawn-based characters who primarily depend on overwhelming strength.

This rename will more nicely fit into the synergy scheme that social stats will support, and again, more on this soon.

You are of course completely correct about the problem of balance, which is why once this thing actually gets going it will need a great deal of playtesting and iteration, which is really the only way to get not only balance, but gameplay in general to work. While I will be writing many articles about how I think I want systems to work, I am not at all under the illusion that those decisions will actually hold up under playtesting.

I don’t want to go too deeply into the attribute system yet, because I’ve got some stuff already outlined on that front that will likely be up towards the end of the week, and it would make more sense to discuss it then.

As far as unit development, the first part you talk about is essentially what D&D 4E does, and I have some ideas along that front, but the second thing you mention is the crux of what I meant with that unit development section.

I do not, under any circustances, plan to have random attribute increases, it’s exactly the kind of thing I want to avoid, and the idea I am currently considering most seriously for this problem is very similar to the “template” idea you mention here. And yes, I will certainly be talking to you about all this stuff as it develops, brainstorming and discussing these sorts of things is another Very Important™ aspect of designing a game.

As far as the mechanics go, I hope to strike a balance between complexity and intuitiveness. I think no matter how crazy the mechanics might get behind the scenes (or how not simple), they need to be understandable, even to a relative novice. Any one feature should at least be easy enough to understand and explain that a player can immediately begin to use and get some benefit out of it, even if there is a lot of complexity in the background that an expert could exploit for even bigger gains.

Completely custom setting: Good luck, most everything here has already been done to death, especially if you are going with a high or low fantasy setting. Attributes and statistics play a big part in this genre, so even if you don’t have strength or intelligence, or any other of the D&D big 6, you’re still going to have some number tied to physical damage, or damage resistance, etc.

Player driven narrative: “But you must!”

Battle Mechanics: All about balance. Make any class combination too powerful and you’re discouraging alternative strategy. DragonMage anyone?

Custom Units: See above blurb on custom setting. You might consider classes playing a minor role, with most skills or abilities having prerequisites on attributes. Dedicating to a single attribute score (see below) will let you get to some of the more powerful abilities. Splitting between attributes might make your warrior hit a little lighter, but at the trade off of being able to heal himself and others. Maybe make the classes add flavor, with unique skills which have prerequisites both on class level and attributes. I guess what I’m saying is have a large commonly available set of skills, which players will constantly be using and always want to have access to in some way. Add in specialized class skills to throw in flavor and pull off those pure moments of awesome when you create a synergy between class skills and common ones.

Unit Development: It might be kind of interesting to make it so that class determines the primary attributes for each skill or ability. Example, melee damage for a warrior type is based on strength, but for a rogue it is based on dexterity. Maybe even a class that has their melee damage based on their charisma! “You fight like a dairy farmer.” “How appropriate, you fight like a cow”. Take it a step farther by doing away with random attribute increases on level up. Let the player manually distribute points or else select from “templates” to distribute the points on level up. Add in unique skills for each class which are unlocked after increasing an attribute to a certain point, to encourage specialization and also unorthodox class builds. Warrior class skills that have a wisdom prerequisite. Wizard class skills that have strength prerequisites. The possibilities are endless. Lot’s of ideas here, hit me up sometime if you want to talk about it.

Mechanics: No need to have the mathematics be absolutely insane. They just to be insane enough as to not be easily discernible to the average player, while still being decipherable to the power gamer. Randomness inherent in the system? Randomness depends on player actions and equipment choices (similar to tactics)? There’s also some fun to be had here.

End wall of text.

In other words, assuming that in a team game they average the skill of the two teams and then match based on that averaged skill, they could do the same thing with “difficulty” as long as difficulty was chosen per-team not per-person. That would probably be the easiest way to make something like that work, difficulty would be an option chosen when you go to get matched, and if you already have a team at that time, you would choose the difficulty for your team, and if you were trying to get matched with a team, it would match you with others looking for a team with that same difficulty.

Still, you’re right that there is a lot of complexity there, for probably little overall gain. They’re better off working on a desired pacing, which is exactly what you describe as far as difficulty ramping up, and breaks affecting the curve.

You mention getting a focus group to test these scenarios, and the cool thing is that this is exactly what the current closed beta is. Blizzard is getting all manner of interesting metrics from people playing the beta, and that will help inform exactly these kinds of problems, as well as things like “are there measures we can take to encourage people to play more than just the same race every game” and things like that.

I agree with battlenet throwing you an easier matchup when you’re on a losing streak. Doing so not only helps the player feel that they still are competitive, but also prevents them from simply getting frustrated and dropping the game entirely, as you did with the original.

I would like to see battlenet actually monitoring the time a player has been logged in. The longer you are logged in, the harder the opponents you are seeded against. That way, you could begin an evening with a warm-up (against a lower level opponent who has been online for awhile) with the games getting more and more frantic as the night went on. Doing it this way also might cause people to take more breaks during their play. Have the difficulty “reset” after 15 – 30 minutes or so of being offline (determine it by focus group testing, think “rested state” in WoW). Hopefully, this gap is large enough so that the greifers aren’t willing to wait, but short enough that the players who want to continue playing but not be matched against tough opponents don’t feel punished. I know I could use more breaks to rest my eyes, and with a mechanic like that, I would actually have a good reason to take more.

My two cents.

As far as setting up your own server goes, I specifically wanted to find the solution that would require the least manual configuration of servers and such. I have set up my own web server under Linux before, and it wasn’t terrible or anything, but I really don’t want to have to deal with any of the details these days, I’d rather focus on the development I find interesting.

With that in mind, I found the sites offering SVN hosting to be a fine choice, especially since I doubt we’ll need more than 200 MB, and if we did, a year-long subscription with a bigger limit is still cheaper than a single new video game.

As for servers, while many online services make SVN servers available as part of a package, I’m using VisualSVN Server for Windows at home, which is also free (the company charges for their VisualStudio Client PlugIn), and can be managed locally with a super-easy GUI interface.

Alongside this, I’ve installed Subversion Notify For Windows, which is a server-sde check-in email solution that is super-slick. It allows for complete control over recipients by path, and the check-in emails that it sends are glorious HTML formatted messages that display the files changed, check-in comments, and a full diff right in the email.

If you’re running any kind of “always on” server at home (like WHS), and you have a persistant Internet connection and a router that allows port-forwarding, you can run VisualSVN locally and use Dynamic DNS to provide a URL for your teammates. Then you’re not limited by service provider storage or bandwidth limits, and the configuration is completely in your control.

Race for the Galaxy is awesome. Even if you don’t win you can have a ton of fun. One game I started as Old Earth, and promptly settled Earth’s Lost Colony and New Earth. Then, I formed the Pan Galactic League so that all of humanity could have a say in interstellar politics, regardless of location. Unfortunately this isn’t even close to a viable strategy, but the shear luckiness of drawing all the cards to play thematically correct made it awesome.