Before I step back onto my soapbox, I have to clarify a few points, I consider a sandbox as non-linear gameplay resulting in emergent gameplay, though even these terms are not all encompassing, if they were then there would be no need for the term sandox. I also do not subscribe to the idea that free-roaming is in any way necessary for a sandbox and vice-versa. Here I purely look at the perspective of building things and knocking them down with the view to achieving an open-ended scenario in which the players decide how they are going to achieve their goals, both game designated and self-appointed.
Right, up onto my pedestal where was I?
Asset Capability Against Resource Cost
In this blog I will use chess pieces to explain my points. While not a sandbox pe se, using superior damage reduction capabilities of ablative armour against its poorer resilience for example would need clarification.
Imagine how chess would differ if the pieces in it had 'hitpoints'? Imagine if a knight could remove the first pawn that attempted to take it but fall to the second or a rook could only be taken by knight if the taking piece is also back row and supported (can be theoretically taken) by another piece. Imagine if the pieces had differing hitpoints and support values, or even support values that were variable to the position being taken and the position doing the taking.
The nature of sandbox games is that there are invariably a great many more parameters to consider for every asset. I have used the word asset, though this is interchangeable with piece (as in chess) item, position or even location in some cases.
When developing a game system, it is fundamental to start with as few parameters as possible. In chess, this effectively stops with assigning a starting location for each piece and a method of moving. From this the game evolved, establishing strategies (fool's mate) and tactics (forks and skewers) based on these two parameters. The nature of sandbox games however is that assets invariably have a lot more parameters that often are integral to a much more developed combat and/or interaction mechanic within the game.
From a development point of view therefore, it is much better to first establish a baseline in any mechanic. For example, any piece can take any other piece simply by moving onto its square. A more advanced mechanic for example could have the baseline, all weapons deliver their damage value, while all armour stops their defensive value. This can still be applied to chess by presuming all pieces have a weapon that delivers 2 damage and all pieces have 1 armour and 1 hitpoint.
Advancing this mechanic is very straightforward assets can be given differing damage, armour and hitpoint values.
Armour can be given variable defence, such as preventing an average damage based on their armour value. This introduces concepts such as critical damage (where the armour just happened to stop zero damage).
The introduction of the term average is a big step forward, moving the game from purely predictable to probable outcome. The nature of play therefore changes from the bare minimum to ensure victory to the perceived minimum to ensure victory. From experience, where the outcome is not guaranteed, many players will opt for overkill.
This last point brings me onto the purpose of this blog assets against resources. As a designer of a sandbox, you can virtually guarantee that players will strive for the utmost overkill. This allows you to weight the resource cost against the asset capability disproportionately while still confident that the asset will be sought after and utilised in the game. In fact, failing to give superior assets a disproportionate resource cost to asset capability makes all inferior assets in the game largely pointless.
As the quantity of parameters inherent in the sandbox increase, so to does the risk of design oversight. This is where two or more assets (which could be two or more pieces, a number of pieces and a location or some other situation) combine to skew the advantage unfairly in one direction. The simplest way of looking at this is if the combined asset capability under relatively common situations is greater than the resource cost of an asset with the same capability, then there is a design oversight. In our modified chess game, this would be allowing any piece that reached the opposition's back row to convert to a queen. Another example would be allowing the weakest character to equip powerful weapon where the effort to become a powerful character is magnitudes higher than the effort required to acquire a powerful weapon.
One option is to pre-nerf assets as they are added to the original game design. This essentially means that the degree of disproportion of the resource cost to the asset capability is overly exaggerated at the time of inclusion. This is achieved by a number of methods. The first is simply making the time factor to acquire said resource exceptionally long (multiple hoop jumping due to low drop rates, or time to completion being vastly inflated). Another option is to distribute items into the sandbox, limiting their presence in the game. This allows for observation of the asset for a period of time before allowing them to become mainstream or introducing modified variants. The last (common) option is to simply dissallow combinations until certain hoops have been jumped, e.g. low-level characters can't weild powerful items.
In the next blog I will look at Construction Risk. This feeds heavily back into asset capability against resource cost.