I'm working on an RPG project and I'm currently trying to design & balance "procs" (properties that grant an effect based on a percentage of occurance during a specific action).

Let's talk about the most common ones: proc on hit. They can trigger when a unit is attacking another one. For simplicity sake, let's say that most offensive skills can trigger procs.

I'd like to find a design that allow procs to contribute to the character's customization depth. I'm having trouble dealing with the fact that skills in my game can have very different cast time (from 0 turns to a couple of hundreds at the extreme, with most of them in the 10-100 range).

In my game, all procs will have a fixed percentage of proc that will depends on the stats (items, talents, buffs...) of the unit. The effect of the proc (damage, heal or (de)buff length) may vary if we find it necessary.

In my game, there are mainly three paramters that can influence a skill's cast time:

  • The skill's base cast time
  • The unit's weapon attack speed
  • Item/Talents/Buffs that can improve these two things

Option 1: Proc have a fixed effect

If regardless of what kind of skill is used to proc it the proc always has the same effet, it heavily favors units have to cast fast. So low base cast-time skills and fast attacking weapon will have an order of magnitude (easily) more procs than a unit using a slow weapon and a long base cast-time skill. We can balance it properly by assuming it should be efficient only at fast (or very fast) skill cast so it's a perk. The biggest downside is that it will be useless (or close to) for all builds base either on long cast-time skills or builds using a slow weapon :/

**Option 2: Proc effect normalization **

We can also make proc effect to scale with the skill's final cast time (the result of the 3 parameters detailed earlier). This is a simple normalization process that will, over time, make it so every unit using the same proc will have the same cumulated proc power over a large period of time. Balacing it may be a bit more problematic since we will have the weakest proc be around 10 times less power full than the biggest ones possible. It has the quality of letting every kind of build use procs as efficiently as each other. The biggest flaws is that it is, I think, counter-intuitve for players whose primary reasoning when seeing a proc stat will be "If I attack faster, it will proc more" (note: proc more in our game means proccing with more power as the proc rate per attack is fixed). That is a fantasy about proc that Ideally i'd like to keep.

Option 3: Proc partial scaling on skill's cast time

This is what I'm leaning tower to at the moment. The idea is to keep the design goal of having proc beeing better with fast units but still make them viable for slow units. I was thinking of using a scaling between the final skill's cast time and the proc effect but not a proprotinal 1:1 as in option 2. Something like this:

  • Skills with 5 cast time (already very fast) - proc coefficient of 0.25
  • Skills with 40 cast time (the average base cast time) - proc coef of 1
  • Skill withh 400 cast time (extremely high) - proc coef of 3

In this setting, proc would be balanced against a medium cast-time skill.

Slow units would still be able to use procs but at a (400/40) / 3 = 33% efficiency comapared to units who use 40 cast-time skills. Fast units will have be able to use procs at a 200% efficiency compared to units who use 40 turns skills. (numbers are just here to illustrate my points, there are not very good yet).

This options would have the small drawback of beeing a bit hard to explain to players since the scaling is neither null nor pureply proportional to attack speed but likely logarithmic.

What do you guys think ? How would you attack the issue? I'm open to any question or criticsm :)

  • 1
    \$\begingroup\$ This kind of open-ended question, where you are seeking opinions or a discussion about which two balance options are "better," doesn't really fit the SE Q&A model of objective questions and answers. A good place to ask would be over at GDNet \$\endgroup\$ – user1430 Mar 5 '17 at 16:17
  • \$\begingroup\$ This came up in the "reopen vote" queue, but I don't feel I have a clear enough idea of what problem OP is trying to solve. More description of what the "proc" mechanic is, or example gameplay scenarios OP wants to balance (eg. "the player is in this circumstance. If they do X, Y happens. If they do Z, W happens. Both outcomes should have roughly the same expected damage/life/payout in the long run") might help resolve this and narrow it down to a clear, answerable scope. \$\endgroup\$ – DMGregory Mar 6 '17 at 4:31