My Crafting System – Part 1

I often complain about crafting systems in games but I have yet to make a good suggestion on how a good crafting system could actually look. Today will the first part in a spread-out series of posts capturing my ideas while creating my very own crafting system. If this gathers enough interest and I actually get that far into the process I plan to not only create a full-fledged design document on the topic but maybe even implement it in some way or another. This might also turn out to be a one-off article in which case you’ll never hear me talk about this again, but I have high hopes to get something productive out of this eventually.

Before we get into the meat of the whole business, please note that there are quite a lot of things I will ignore for now. Among those are the lore (or story) behind the game, the mechanics of the rest of the game, and questions of implementation. Whenever I mention something from those areas it will be purely used as an example and nothing else. I might talk about metalsmithing in one paragraph and the construction of a plasma cannon in the next, take either one as a representation of abstract crafting concepts.

Starting the development process, I know that I definitely want my crafting system to reward experimentation and I don’t want the results of those experiments to be freely shared on the internet. I have previously outlined two methods of achieving this – random tasks to solve while crafting and random player attributes that influence crafting. Eventually I want my system to have a bit from both worlds, but I’m not interested in going into the nitty-gritty of how the actual crafting will work just yet. I’ll therefore start with the idea of every crafter having a unique sets of attributes that influence his or her crafting.

A real world representation of this is easy to imagine. A smith with strong arms and big hands would be well suited to create strong and heavy weapons but will obviously be less skilled at detail work. When adding magic (or technical knowledge) to the system, things become even more obvious. A crafter skilled in the art of harnessing the power of fire will produce products that are clearly different from one that uses air instead. A mage might still be able to weave fire power into a weapon even if that is not her specialty, but it would be a crude piece of work, likely relying more on power than finesse.

There are various ways of deriving a set of attributes for the player, and role-playing games have used them all in the past. One could have players start with randomly generated attributes, one could allow players to choose their characteristics through a points-system, or one could base the whole thing on gathering experience in the related fields throughout the game. There are ups and downs to each of these. Random generation has the problem that players will feel treated badly when they don’t get the stats they would like to have and little is more frustrating than planning to make goldsmith and getting a huge brute of a character to work with. In the past, players have shown a tendency to re-roll their characters (generating a new random one) when they didn’t like the initial set of characteristics. This process is pretty boring and essentially only amounts to placing the points yourself because you will just re-roll until you get the desired outcome (or give up on the game through frustration.)

There are possibilities get around this problem. One could allow players to choose a basic direction they want their character to be in and then only slowly reveal the details of their abilities throughout the game. Players would be much less likely to re-roll a character that they’ve already invested significant amounts of time into and that at least roughly fits their ideas of how they wanted to play the game. A character might start out as an electrical engineer for example and only later discover that it has a penchant for constructing weapons with rotating frequency modulations. (Yes, that is sci-fi mumbo-jumbo that doesn’t make a lot of sense. That’s usually how sci-fi works.) Such a system is very susceptible to imbalances in the game design though. It’s tough to be stuck making particle beams when pure energy weapons are currently overpowered in the game.

Allowing players to distribute their points themselves is somewhat problematic from a balance point-of-view as well as players will tend to pick the most powerful builds and everyone would look the same. This can be counteracted through good game balance of course, but also by making each and every type of specialization useful and necessary from time to time. Niche spezializations would have less crafters choosing them, but that would increase the demand for them. The bigger problem I see is that this would allow players to put guides on the internet detailing exactly which points to choose for which outcome, which defies the intents of my crafting system. The game would have to somehow encourage a diversified distribution of points if free choice was to be allowed. One idea I can name for this from the top of my mind is that external circumstances could direct you in a certain way. You might (randomly) have the materials to make a couple of broadswords at the start and would put points towards the creation of good ones or face the consequences of producing terrible goods. Their could also be an influence of teachers (players or NPCs) and your stat distribution could vary by who you are learning from.

I personally like the concept of increasing the attributes that you use the most (as it is done in the Elder Scrolls series of games for example), but maybe this form of advancement is better placed in skills (which I’ll discuss on a later date) than attributes. Skills allow for a much larger degree of customization and don’t reward behaviour such as jumping a lot to increase your physical prowess for swordfighting. Dull excercise to strengthen you muscles or reflexes might be very realistic, but it’s not exactly fun.

The last option would be to ditch attributes alltogether and go straight into skills. This tends to lead to a quite confusing character creation though (when players get to pick between a huge variety of skills instead of a few attributes) and it reduces the amount of character customization possibilities. It seems to me that it would be more interesting to allow players to balance the attributes of their swordsmith instead of just picking a specialization in scimitars.

What do you think, would you rather have a (large) set of attributes that are determined at character creation and partially out of your control, or would yourather to have a smaller set off attributes that you can choose by yourself at the risk of meeting various people running around in the game with the exact same distribution of stats? Would you mind if your character slowly discovered traits and abilities throughout the game or would that maybe even add an element of discovery and surprise?

Also, are you interested in these types of posts or would you rather that I quietly developed the system and then presented it when it was done?