I’ll be discussing the C2020 game mechanics and how to revise them from a top-down perspective. This is for “C2020 Redux”, a design project I outlined in an earlier article. Being familiar with the original game, or reading that intro article, is probably pretty necessary for following this one.
How C2020 dicing works mechanically
C2020 is a remarkably pure pinnacle of ’80s trad game design. The prior decade had seen this style of game emerge and be perfected, while the next decade would be more of an era of straining against the premises implicit in the tradition. The gamers of the time obviously didn’t see it this way, but to me the history of the trad style rpg ends up seeming like C2020 sits neatly at the high point of its style. It comes just before games like Vampire; you can even see a hint of a shadow of the future in it, in things like Luck, Humanity and Reputation.
This being the historical context, C2020 belongs in a school of design where having universal resolution (the same technical process is used to resolve all kinds of fictional situations) is the newest hotness, while resolution mechanics need to be completely subjugated to the GM’s control over the process of play. “Task resolution” is the rpg theory term du jour here, and understanding the different ways such can be utilized in games is key to GMing C2020 effectively. The game can be the perfect illusionist vehicle for a GM who wants to control everything, or it can be the chassis for organically emergent play, all depending on how the bland, toothless core resolution mechanic is used. From my viewpoint it’s like the game doesn’t actually say anything in its game mechanics, as it has very sparse procedural advice for determining when and how and why the task resolution rules are actually used: it genuinely depends on what you yourself want to be doing with a roleplaying game, the game text doesn’t have an opinion.
For the record, here’s how the dicing works in orthodox C2020. Each dice roll is usually assumed to resolve a single significant task that comes up in the emergent fiction, so you’ll be rolling for things like “lockpicking checks” or “driving checks” to find out respectively whether the character manages to pick a lock or stay on the road.
vs.
[GM ass-pull target number] + [GM ass-pull difficulty modifiers]
(That part about how the GM arbitrarily determines a target number and arbitrarily modifies that difficulty further is explicitly a presentation procedure explained in the text. As the game text says, players will better go along with the GM’s target number judgement if you set a low base target and then pile on explicit circumstance modifiers than if you just present them with a high target number in the first place. It’s the GM making the judgement either way, the difference is in whether you explain to the players how you came up with the number or not.)
This is all extremely familiar to old-time gamers, of course: stat + skill + die vs. target number is to my mind, and I expect to many others, the final and ultimate form of the traditional task resolution mechanic. C2020 was a pinnacle design, a remarkably clean implementation of this arrangement in its time. So much time has passed that I know gamers nowadays for whom this stuff would actually be novel, but for most people it’s still the standard of how rpg rules mechanics work. Modern D&D has been playing off this formula since 3rd edition, for example. I suspect that the arrangement having a shorter list of “abilities” combined with a longer list of “skills” hits some sweet spot of complexity-in-presentation that makes it so popular; it’s nothing impressive as a resolution procedure, but I think we gamers tend to like it because it makes the character representation look good on the character sheet. Essentially, we want to have abilities and skills on the character sheet, and this happens to be one of the most straightfoward ways to actually integrate those recorded details in the proceedings of play.
There’s not much else of importance to say about the C2020 core mechanics, really. The d10 used in the resolution rolls explodes (roll again and add) on a ’10’, and rolling a natural ‘1’ causes a fumble. As one might expect, in addition to deciding the target numbers, the GM also decides what needs to be rolled for and what success or failure on the rolls means. (This is what makes the system completely toothless: it’s not a procedure for play, but rather a mere description of a ritual the GM uses at their own consideration. The real system of play is always the “how we use this task resolution thing”, not the thing itself.)
How the character sheet gets filled
I’ll review the second important part of the core mechanics as well before going into my revisions. As I hinted above, I view the universe of trad rpg mechanics (of which C2020 is a very clean example) as a type of mechanical theater: the system is very universal and uniform, and it’s been atomized into a porridge of fiction and task resolution kernels that the GM can conduct freely according to their own vision. In the world of this porridge there is one thing above all that grows in significance in the eyes of the character players: the porridge may be fuzzy and messy and arbitrary, but your character sheet has real, solid numbers on it!
In this environment players generally care really, really hard about the dance of those little numbers. What’s your Strength score, or how many skill points you have in Firearms, those are in many ways more real than anything else that happens in the game. They may not affect the proceedings of play as much as you imagine as a player, but it’s like any human endeavour really: the small things you can know and influence are much more important than the big things that are invisible and outside your control.
C2020 believes fanatically in the desirability of point-buy mechanics: the best way to produce that stat array for a character is to give the player a big chunk of points, which they’ll then distribute into the statistical array to produce the exact character they want. What you get in practice is some pretty arbitrary min-maxing, because if you like the idea that your character will be strong, have a high Body Type (yes, that’s what C2020 calls Strength), then why would you set it at anything except the maximum value? That’s the cleanest representation of the idea that your character is strong, after all.
The point-buy idea is stupid and doesn’t work very well, but that’s what C2020 believes in, because it believes a lot in the ideal of letting the player develop their own character (but within bounds because fairness is important). All this is more understandable if you imagine the game portrayed against the backdrop of a monolithic Dungeons & Dragons looming over everything all the time: if you want the opposite of randomly rolled stats and inflexible character classes, then point-buy and long skill lists in lieu of classes is certainly one imaginable antithesis.
But anyway, the takeaway we need here is that in orthodox C2020 character creation you start by rolling 9d10 (because there’s 9 Stats, you see), sum the results up and then distribute the points into your character’s Statistics any way you like. 1-10, with ~5 being considered an average score. The Stats themselves are about what you’d expect, general character traits like Intelligence, Empathy, Technical Ability and so on. (And I’ll just let you imagine what I think about not only having pointbuy, but also having that point pool arbitrarily randomized. It’s like combining the worst worlds of random stats and pointbuy.)
Following the Stat distribution you do some lifepath stuff (important for chargen but mechanically freeform, so not relevant here) and then do Skill distribution, which works pretty similarly: you get 40 points to distribute between the ten Career Skills of your chosen character Role, and ~10 points to distribute freely between the hundred or so skills present in the skill list. It’s not as bad as many of these skill-based games that give you free run of the massive skill list, but you’ll still be spending quite a while counting points here.
The third character development rules mechanic we need to know about is the experience system, which is completely independent of the prior systems. Stats do not develop by experience, of course (well, it’s obvious if you’re immersed in the tabletop rpg culture), so we’re talking Skills here. This bit is actually important, so I’ll put it in a separate paragraph:
C2020 has a naturalistic experience scheme: all skills advance independently of each other, at variable mechanical speeds, and you gain improvement points (IP) as per fictional events. Things that you’d expect to work in reality to improve skills work here as well: Study gets you 1 IP per day (or more with teacher), successful skill use in traumatic adventuring situations give 1-9 points.
I emphasize this bit so heavily because unlike everything I’ve explained so far, this bit is not generic and obvious. Dungeons & Dragons doesn’t do naturalistic experience; C2020 is following in the footsteps of Runequest, the progenitor of the trad style rpg here. There are many possible implications to this situation: it may mean that C2020 characters actually do all kinds of interesting study and training activities in between adventuring, or it may mean that advancement basically doesn’t exist in the game. (The latter is how it played out for us when we played in the ’90s; I don’t remember character improvement having been much of a thing. I didn’t even remember that the game has a naturalistic experience scheme before reading it anew this winter.)
You can also invent new Skills, which makes perfect sense for a game that treat this whole skill thing in such a naturalistic way: the question is at no point whether there are enough Skills for all the Stats to be attractive, or if some Skills might overlap or not; the consistent concern is whether the Skills actually make some sense as things that are human skills in the game world. If you figure out something that should be a Skill realistically speaking, go ahead and add it to the list. The Skill list is not sacred, not interconnected with anything else in the game rules really: it’s just a list of various skills that might be important and useful for C2020 characters.
Revised Dice Mechanic
OK, so that’s enough background — let’s see what I want to do with all of this. The Redux project could get pretty far without touching the game mechanically, what with the trad style game having such a disjunction between the mechanics and process of play, but if I see something that I can improve on, it’s a shot I’ve got to take. After all, the game needs to be cool and polished to make me want to actually play it!
That said, my primary and most fundamental change is pretty radical; so much so that I’m not even 100% convinced of it myself. Needs playtesting. What I want to do is this:
Character Stats do not add to task resolution checks. The standard check is as follows:
vs.
[Target number] + [Circumstance modifiers]
A natural ’10’ on the dice explodes; a natural ‘1’ causes a Complication (like a fumble, but you can still get something success-like if you reach the Target number); you can get both on the same sequence.
A dicing result that doubles the target number is a Cool success, what I’m calling critical successes here. I’ll presumably discuss these more at some point, but a Cool success looks brilliantly effortless to on-lookers, allows chaining actions and so on. A result that doesn’t reach half of the target number is a Bunk failure, sort of the opposite of the Cool success; not quite the same thing as a fumble, but more of a situation where it’s obvious that the character has no clue what they’re doing, and shouldn’t have tried in the first place. A result that hits the brink of the target number (so TN or TN-1) is a Flimsy success or failure, a sort of partial success if you will — the character did succeed or fail as indicated, but not necessarily completely, and the performance seems shaky to onlookers. Routine tasks are always simple successes and are not modified this way.
The character trying to perform the task can mess with the formula by picking a special performance mode more appropriate to the situation than the standard check:
Defaulting: A character without the appropriate Skill (or a low value in it) may opt to default to relevant Stat instead of the Skill. They add a penalty die to the roll (roll an extra die, use the lower result), so this is usually only a good idea if your Stat is several points better than the Skill. Defaulting represents trying to figure out the task and the best way to perform it intuitively as you go; particularly obscure tasks (where basic intuition would lead one astray or the task absolutely requires specialized knowledge to get anywhere) may add two or even three penalty dice to the check at GM discretion.
Routine: A character may opt to not roll at all, simply taking their Skill value as their result. This has the advantage of ensuring no Complications, but obviously it only works out well if you can get your Skill + modifiers to match the target number without the dice. Routine tasks are always simple successes or failures (as opposed to Bunk/Flimsy/Cool). A task is never routine if the character defaults to Stat as per above.
Careful: A character with ample time and resources, and their full attention, may perform the task carefully. Doing so is worth a +3 modifier to the check. The die doesn’t explode when rolling carefully, but a natural ‘1’ is still a Complication. A Careful Routine (combining the two performance modes) is very possible, and often preferred in everyday tasks where Complications aren’t worth the risk.
Inspired: A character with skin in the game (emotional stakes riding on the outcome) can tackle the task with their full energy and attention. They gain a bonus die (roll an extra die, use the higher result) to the roll, but with a higher chance of Complication: if any of the dice in an Inspired check show a natural ‘1’ the task will be Complicated. An Inspired Default is possible, in which case the bonus and penalty dice counter each other, leaving just the one die.
Panicked: A character who is shaken by a crisis situation may be forced into panic, in which case they must default to Stat, ignoring their Skill value and suffering one or more penalty dice. A panicking character with Skill outright higher than the respective Stat gets a +1 skill modifier to the roll, but that’s all the advantage their Skill gains them. Needless to say, panicking in a crisis situation is a good way to make a bad thing worse.
I’m currently leaning towards not having degrees of success (and particularly no success-counting) aside from the Cool/Flimsy/Bunk thing, which people familiar with my design style may find unlikely at best, but we’ll have to wait and see where this goes. I’ll discuss the big picture plan on how to actually utilize the resolution results later. Having task resolution rolls be binary, without success-counting, would certainly be in the spirit of 1990. This is before dice pools became popular, remember.
Next, let’s benchmark things a bit. The original C2020 rules benchmark in a way that I would characterize as hypocritical: skill descriptions make a big deal of how skill values supposedly reflect fictional situations (“At +6 you can intimidate Sylvester Stallone or any moderate ‘tough guy’. At +9 you could intimidate Arnold Schwartzenegger.”) without the rules doing the least thing to actually back up these claims — they’re pure fluff. Meanwhile, task difficulty numbers are benchmarked with vague adjectives (a task is “of average difficulty”, and therefore difficulty 15). Perhaps we can do a bit better.
First, what the Skill ratings mean? What’s a low Skill, what’s a high one? Here’s three basic ranks I’ll be referencing in this regard.
A Newb is a character who doesn’t have any significant training in the Skill — rating 0‒2 or thereabouts. A Newb will often rely on defaulting to Stat, as most Newbs will have a Stat value ~5, which overshadows the penalty die from defaulting. The average Newb will roll an average of ~9 on a task using Stat defaulting, or ~6 without.
A Pro is a character with 5+ levels in the Skill. As the name implies, this tends to be a level expected of professional performance of the Skill; it’s also a level where the character will find it trivial to demonstrate their superiority in the Skill compared to the Newb. Many study programs and such strive to bring character Skill to this range. A Pro generally stops defaulting on rolls; I’ll estimate their average roll at when all’s said and done ~12. Opting to routine simple tasks is often key to how a Pro performs a Skill efficiently.
A Master is a character with 10+ levels in the Skill. A character can develop a Skill to a maximum of double the relevant Stat value, which means that an averagely-talented character may just barely reach Mastery if they put in the required work. Improving the Stat itself is also possible to take a Skill even further, but it’s expensive and generally beyond feasible for most characters. Mastering a Skill is not very useful for many Skills, as the tasks they’re used with do not necessarily require hitting high target numbers. Your average Master will roll ~15 and up, which suffices to trivialize most tasks.
So Skills go from zero to 10 like traditionally, with potential for exceptions on the high end, and weird probability interactions in the low end thanks to the Stat default rule. (As a rule of thumb, the penalty die will reduce the dice result by a 2‒3 points on average. Unless the heightened fumble chance worries you, default to Stat when the Skill is 3 or more points below the Stat value.)
As for target numbers in various tasks, I’ll give some rules for determining those:
The default target numbers (before external modifiers) look like the following. I added some simplified probabilities for benchmark characters; in practice the results would have more swing due to natural specials (fumbles and explodes) and various personal and circumstance modifiers, but these white-room numbers should still be helpful in wrapping your head around the difficulty classes.
| Target Number: | Difficulty Class: | Newb (Stat 5) Probabilities: | Pro (Skill 5) Probabilities: | Master (Skill 10) Probabilities |
|---|---|---|---|---|
| 3 | Trivial use of the Skill, generally not an issue, I’m just listing it as an option to be thorough. Things like throwing a ball or turning a computer on. | Success 100% | Success 100% Cool 100% | Success 100% Cool 100% |
| 6 | Basic use of the Skill; professionals perform these actions routinely, while newbies can take a stab, but aren’t really consistent due to Complications. Many tasks in adult life are Basic, performed routinely where possible. Driving a car on a public road or using a text editor on a computer are Basic tasks for their respective Skills. | Success 100% Cool 16% | Success 100% Cool 40% | Success 100% Cool 90% |
| 9 | Challenging use of the Skill; not really practical for the uninitiated, while professionals will need to work carefully to avoid risking failure. Challenging tasks are used in highly skilled occupations where substituting simpler tasks wouldn’t be practical. Flying an aeroplane and performing chirurgy are good examples. | Success 49% | Success 70% Cool 7% | Success 100% Cool 20% |
| 12 | Exceptional use of skill; even a Pro struggles and can’t guarantee success. Tasks at this difficulty level are only common when consequences of failure are low (e.g. many sports) or stakes are high and there is no other way to succeed, such as in war or medicine. And adventuring, of course, where exceptional solutions to exceptional situations abound. | Success 16% | Success 40% Cool 2% | Success 90% Cool 7% |
| 15 | Cinematic use of skill; a Pro could conceivably make it on a good day, but it’d be luck more than anything else. Society does not generally rely on tasks this difficult in regular activities, so they mostly pop up in exceptional crisis situations, circus acts and games. | Bunk 36% Success 1% | Bunk 20% Success 10% | Success 60% Cool 1% |
| 18 | Impossible use of skill; movie heroes succeed, real people almost always fail. Nobody would bank on succeeding in something like this, so the TN should only come up in exceptional hail mary situations. | Bunk 64% | Bunk 40% Success 8% | Success 30% |
It would be ideal to think of TN selection in terms of those categories rather than raw numbers: think “this is a challenging task, therefore TN 9″ instead of “the TN should probably be about 9-ish”. This should help a bit when avoiding conflating personal desires or game balance concerns or such with a task that is intended to be purely simulative. Assign TN on the basis of how difficult the task is in the abstract and then apply circumstance modifiers. Over time the different TN classes should develop character in your mind, such as it becomes obvious to say what are basic or challenging tasks for a given Skill. To summarize, here are the TN classes again:
Trivial — TN 3 — routine for most
Basic — TN 6 — succeeds untrained
Challenging — TN 9 — professionals succeed consistently
Exceptional — TN 12 — success can’t be guaranteed
Desperate — TN 15 — circus trick territory
Cinematic — TN 18 — not expected to be possible
The personal modifiers on character side and circumstance modifiers on the target number side should be understood as pretty formal, rather than arbitrary; I’ll need to delve into this in detail later but the Personal side includes things like Passion, Equipment and Reputation bonuses, while Circumstance side involves stuff like being distracted or in a hurry while performing the skill. The GM is allowed to lower or raise the TN by a single point on pure gut feeling, so when we say that a task is “Basic” (TN 6), we actually mean that it’s somewhere in the 5-7 range. If the GM makes this Gut correction after the roll (entirely GM’s choice, unless they established the exact target number in advance), it may affect the outcome, even turn a success into a failure or vice versa; them’s the breaks, and that’s how far I’m willing to entertain the GM’s gut feeling.
If this seems like a god-forsaken mutant off-spring of C2020 and Ars Magica, that’s not an accident. While the games are originally from the same era, with some similar attitudes, AM is on its fifth edition by now, and man is it sleek. Insofar as you think that the two games are trying to do similar things, you could do worse than stealing from AM. You’ll be seeing more of this particular crossbreed in this project.
You might wonder what I’m going to do with Stats if their only task resolution role here is act as desperate skill defaults. Read on, all will become clear.
Revised Skill Training
I want to drop the point-buy from the game entirely and have it lean heavily into the Improvement Point logic: throughout the character creation process you assign Improvement Points into Skills and this in turn determines the final Skill value. There’s more numbers, but I want to get the chargen out of the point-buy pit, and that’s much easier to do if we move the game to use naturalistic skill training uniformly. It’ll be possible to do things like have a character spend a year in a school and outright calculate how many IPs they’re supposed to gain given the time window and quality of teaching.
(Yeah, yeah, I’m apparently creating a cyberpunk-themed Ars Magica here.)
C2020 already has a relatively heavy-duty IP system, I’ll note; for instance, the system outright assigns training weights to Skills, which is something that similar games often avoid to control the complexity. (GURPS comes to mind as an exception.) Not so here, every Skill has a little number (X) after its name indicating how difficult it is to train. Even the aforementioned Ars Magica doesn’t dare to do this, and that game’s a heavy-weight that doesn’t even have any significant content outside training management.
That being said, let’s see what I’ve got for study rules:
Improvement Points — skill-specific points that are used to improve the Skill rating — are gained from various in-fiction sources on the basis of real study circumstances. Each source/activity has a Study Rating (SR) which tells us how much a character learns from it. Here’s a table of commonly useful Study Ratings:
| Study Type: | Study Rating: | Description: |
|---|---|---|
| Day of study | 1 | The character dedicates the best part of the day to studying the Skill, with appropriate training equipment and learning tools. The study session is assumed to take 4-8 hours; a character with extraordinary motivation could fit two (or even three) “days” of study into one day. Inefficient study (e.g. lack of equipment used with the Skill) is only half-effective, so has a fractional SR of 0.5. You won’t generally see the “day of study” used much in practical play, as characters usually study in longer chunks; this is more of an illustration of the underlying simulation. |
| Week of study | 5 | A sustainable pace of study assumes five days of dedicated work per week. Exceptionally motivated characters could double the pace at the expense of free time. I’m seeing this as the default scale used in SimPunk play, so e.g. if a character is a full-time student during SimPunk, they’d simply gain 5 SR per week for their troubles, neat and simple. |
| Year of study | 200 | We’re again assuming a fair amount of human-affairs-time, with summer vacations and such, so an exceptionally motivated character studying without distractions could well double the pace over the year. This SR is most relevant during character creation for characters who go to school. |
| Exposure | 1/ week, 50/year | “Exposure” is a generic term for situations where the character does not study a Skill, but witnesses it or uses it regularly. Characters working a job might be exposed to Skills, and hobbies usually count as exposure. There are all kinds of common life Skills that characters are exposed to as a matter of course in their lifestyle, which means that you tend to “pick up” minor Skill ratings without formal study. |
| Source Quality | variable % | The above numbers can and should be modified in particularly poor or commendable conditions of study. The basic SR assumes 20th century first-world learning conditions and the character possessing appropriate Study Skill (~3 for high school, 6 for independent study) to engage with the style of schooling in question. Halving the SR for major deficiencies (e.g. lacking a teacher for an academic Skill, or equipment for a trainable Skill) and a -20% penalty for minor deficiencies (outdated or biased learning materials, distractions) is appropriate. Similarly for bonuses, major advantages (virtual training environment, elite teacher) would be worth +50%, while minor ones (diligent student, excellent training tools) are +20%. The best study conditions in regular use in elite society might e.g. triple or quadruple the default SRs all told. |
| Practical Experience | 1-10 | Characters participating in NailBag events (adventures) often gain “practical experience”, a more intense form of exposure experience. When a character successfully uses the Skill for real stakes in a crisis situation, the player can request the GM for a SR: 0 for trivial stakes, 3 for minor stakes, 6 for major stakes and 9 for critical do-or-die situations. If a character gains multiple checks on a single Skill during the scenario, keep the highest SR and add +1 to its value. If the same stakes grant practical experience multiple times (e.g. the character uses several Skills in the situation, or multiple characters participate), deduct a -1 from each SR. The SR is processed into Improvement Points when play returns to SimPunk — after the scenario. |
This Study Rating business is intended to be an entirely emulative affair; you could calculate all of the above much more carefully if you wanted to without breaking anything (except maybe your patience). The tabulated use cases are intended as practical approximations useful for play, there’s no secret formal balancing or anything like that in there. Just clear conceptual assumptions and neat rounding. In practical play you should vaguely approximate or carefully calculate precisely as much as seems best for the flow of play at your table in the moment. (Using emulative rules effectively in rpgs is a separate topic, so consider it carefully if you haven’t done it before.)
When characters perform activities with a Study Rating, they produce Improvement Points (IP) for their respective Skills. The IP value for a given Study Rating is found by multiplying the SR by the appropriate character Stat involved in the study situation:
[Study Rating]*[appropriate Stat] = [Improvement Points gained]
The Stat multiplier represents the character’s natural talent for the field of study. The appropriate Stat to use depends on the Skill and how it’s being studied. I haven’t discussed Stats in detail yet, but it’s all rather straightforward — Intelligence for academic study, Dexterity for physical Skills, and so on.
Because the average Stat value is around 5-ish, we can say that characters will generally gain ~5 IP per day of serious study, but, as per above, the values can range from a single point up to a hundred or even more in ideal situations.
So there you have it: Stats, instead of contributing to task resolution, are an efficiency multiplier for learning Skills. I’ll note that the multiplier might seem radical (a character with Stat 10 learns 10 times as fast as one with Stat 1), but I’ll be rejiggering the way Stat values are assigned, and the pyramid pricing scheme in Skill values attenuates the difference such that ultimately a highly talented character won’t run away too much. Trust me, I’m a doctor — or a game designer at least.
Every Skill has an independent Complexity value that determines how many IPs you need to actually improve the Skill rating. The Complexity ranges from around 5 for the simplest Skills (First Aid, say) to several hundred for extremely complex ones (e.g. Diagnose Illness), which means that some Skills can be mastered to significant levels in a few weeks, while others require years to bring to a professional level. The formula for Skill level cost is as follows:
[Skill level cost] = [Skill complexity]*[Current skill level +1]
For example, a simple Skill like First Aid, with Complexity 5, would require 5 IP to reach level 1, 10 more to level 2, 15 more to level 3, and so on. Pyramid numbers familiar from many games.
Here’s a table of appropriate Skill Complexity values at different weights:
| Skill type: | Complexity: | Description and examples: |
|---|---|---|
| Simple Skills | 5 | Simple Skills can be learned to a Pro (~5) level in two weeks, and fully Mastered (~10) in a couple of months of diligent practice. The Skill being “simple” doesn’t mean that an expert cannot outperform an everyman (rather the opposite, as it’s entirely realistic for the expert to hone the Skill to its utmost), but rather that there’s not very much variety to how the Skill is performed, and therefore its formal study is quick and the rest is mostly repeat practice to ingrain the Skill. Examples: First Aid, Pick Pocket, Drive, Shadow, Swimming. |
| Language Skills | 10-20 | Languages vary a bit in how difficult they are to learn, but generally the SR conditions are much more decisive in making progress, as the rote memorization involved is particularly exhausting; languages are learned at a mere 2 SR / week (a -60% modifier for being unsuited to human cognition) in a classroom setting, not really much better than casual exposure. A high Study Skill helps, as does unconventional pedagogy. A Pro (Skill ~5) at a language is just a child or a fluent foreigner; natives generally end up much higher from sheer exposure as they mature, many reaching Mastery by adulthood. |
| Default Weight | 100 | This is a good baseline Complexity for your typical rpg Skill, such as many of the Skills listed in the C2020 rulebook. A character studying the Skill diligently (which I assume to mean half of the time, with supplementary Skills taking the rest of the time) will achieve professional competence in three years or so. It would be entirely reasonable to just assume that all Skills have this Complexity rating and then do reactive fixing when funny-seeming situations (characters learning unrealistically fast or slow) crop up. |
| Academic Skills | 300 | Academic Skills each represent a different field of academic study, which usually share the property of being as wide and detailed as the discipline’s practitioners have managed to develop them over time, with more complex fields being split up into their own disciplines over the centuries. The Complexity is high and becoming a Pro requires most people to dedicate several years of exclusive work to study of the discipline. A Bachelor is about Skill 4, Master at Skill 5 and Doctor at Skill 6, although there’s much variation in practice, and these are definitely Skills where people are compared to each other and consequently over-studying is rewarded by the academic community. The single most definitive task these Skills are used for is “understand recent research in the field” at TN 12. Examples: Anthropology, Biology, Physics, Mathematics, etc. |
As you can see, I’m envisioning the Skill Complexity factors and IP gain rates in a somewhat more naturalistic way than the original C2020 game text; it’s relatively conservative about this, presumably because the designer prefers to maintain a certain degree of gameability. I prefer to prioritize the naturalness, though, which means that I’m picking the numbers on simulative grounds: a master’s degree in this or that field takes four or five years of study, which is ~5000 SR, and assuming that it represents a score of ~5 in the Skill being studied (it’s a field where a single Skill is all that’s being studied through the degree), so therefore I know that the Skill’s Complexity rating should be ~300 to have the numbers make sense.
One more thing about naturalistic Skill training: it is completely in line with this system to track Skill deterioration — the process of losing Skills that fall out of use. Doing so would enable a more precise simulation of certain relatively common phenomena, such as skill maintenance drills used in various instances in the real world; those are necessary because people actually lose Skills they don’t use over time. However, the subject matter of C2020 being what it is, and considering the likely time-scale of a campaign, this is probably not something we need to concern ourselves with for now.
If the SimPunk endears you so much that you want Skill deterioration, I suggest making it so that each Skill loses 10% of the total IP invested in it each year, at the end of the year. (The deterioration rate could change as a character ages, of course.) This means that characters need to practice a certain amount, or gain exposure XP, to simply maintain their current Skill level. For example, if a Master in First Aid, a Complexity 5 Skill, has level 10 in the Skill, they would have a total 275 IP invested into the Skill. They would therefore lose 28 IP at the end of the year from the Skill. If the character uses the Skill professionally and thus gains exposure SR for it (50 SR a year, for an average 250 IP) they won’t actually start getting worse at it until the constant exposure stops.
A more expensive Skill could, however, end up requiring active maintenance even while enjoying exposure: a master doctor with Diagnose Illness (Complexity 200, let’s say) at 10 has accumulated a total of 11 000 IP in the Skill, and therefore loses 1 100 IP every year. A doctor of average talent would only gain 250 IP per year from exposure, leaving their Skill to deteriorate by 850 IP. I would assume that the doctor would arrange to study sufficiently to keep their Skills up to date, being as how they’ve gone to the trouble of achieving such a high level of Skill in the first place. They would, in fact, have to spend over half of the year in study just to retain their leading edge, assuming no study aids or great talent; such high Skill ratings are impractical for the untalented and poor to uphold.
But as I said earlier, I sort of doubt that tracking this sort of extra complication is worth it in C2020, at least not at first. It’s not a very edgy topic to worry about, and there are frankly much more serious facets of the game and playstyle to master before getting into this sort of stuff. Perhaps you’ll want to keep it in mind if the campaign achieves a high degree of mechanical competence and the group enjoys fuzzing around with simulative processes.
Further Elaborations
There’s yet more to be said about the core task resolution package, but I feel like we’re in a good stopping place here; we’ve had a cursory overview of how to resolve tasks, and how to learn Skills that are used to resolve tasks.
The next article in the series needs to go over the rest, though, before we move on to build larger-scale systems. Here’s some of the topics that need to be covered in Pt. II:
- Stats: what they are, how they’re assigned.
- Other traits: I intend to implement a range of non-Skill traits that work in a similar fashion, such as Personality, Passions, Relationships, Reputation; just a rich bunch of the sorts of measures you’ve seen in various games.
- The role of the Role: C2020 features the concept of character Role, with characters gaining special Skills and a list of favoured professional Skills from it. I’ll need to sketch out my plan on what we’re doing with that; it’s an important part of the game, but its mechanical implementation needs to change.
- Task resolution variations, there’s more to be said about that. Using Skills to support other Skills, long-term tasks utilizing multiple Skill checks, how to resolve actual conflicts via task resolution, etc. Routine stuff, but the core dice mechanic has to be applied properly.
- Perks, special privileges that a character gains by having a high trait. C2020 doesn’t have these, but it’s a popular mechanic that I’m seriously considering for reasons that will become evident.
I’ll come back to these ancillary topics in the next article, sometime over this month I suppose.
Once we go the route of naturalistic simulation, though, I wonder what the abstract skill ratings represent at all in the game world. The quality of one’s expected routine performance? That’s more of a fictional/narrative than a naturalistic/sim concern, really.
What would you lose by using IPs more or less directly as the dice roll modifier? Of course, the moment it gets any more complex than “add +1/100 IPs” we wander into ability score/ability bonus territory, but that too is perhaps a good way of looking at it. You don’t get to add +10 to your diagnose illness roll because your diagnose illness skill rating is 10, you get to add +10 because you have 1100-ish IPs invested in diagnose illness and that translates to a +10 bonus on the “difficult academic skills” column of the bonus table.
This might look like an entirely aesthetic concern (or, to some, entirely like a practical handling time concern) but I do believe there are, uhm, ontological implications.
I’m trying to benchmark the formal Skill values and task target numbers here in various ways. For Skills the basic benchmarks are to assume that the difference between Skill 0 and Skill 5 translates in the fiction into a perception of “professionality”; it is enough Skill that people can actually easily perceive the difference. There are many minor facets in what makes a professional, but they’re things like being able to perform Basic tasks of the Skill (TN 5-7) with Routine (not rolling); different specific levels of professional faced with different Basic tasks may need to be Careful (+3 to check) to achieve this, but that’s the general expectation: you’re a professional if you can perform the Basic tasks of the Skill routinely. (“Routinely” means, specifically, “without rolling the dice and risking a Complication”. A newbie can perform a Basic task most of the time, but roughly one time in five will result in a Complication.)
As for how we know which tasks are Basic or Challenging or whatnot, that needs to be determined the other way around: given an ordinary professional in this occupation, what things do we expect them to routinely perform without complications? What tasks are risky even for them, requiring the best person in the team to tackle them? These determinations are entirely independent for each Skill, and do not depend on e.g. adventure utility: for some Skills you’re training that Skill to be able to perform an important Basic task routinely, while for another Skill you want to do something risky and difficult and desire to improve your chances of success.
The goal is to develop the web of emulative concepts towards sufficient detail that it’s actually possible for us to benchmark things agilely and naturally. It should be possible for us to take an arbitrary Skill, say “Driving a car”, and then determine what a Basic task for that Skill looks like: it’s probably “urban driving”, incidentally. A trivial task would be “driving in an empty parking lot”. A “professional of Driving” would, perhaps misleadingly, be somebody with a driver’s license, as this happens to be a Skill that’s very common and performed outside a literal professional context all the time; actual in-fiction professionals would have Skill values several points higher, and possibly specialized other Skills not possessed by the lay-man driver.
As for using the Improvement Points directly as dice roll modifiers, sure – but if we’re still doing Skill complexity and the triangle numbers thing, then we’re basically just saying that the player needs to cross-reference their Skill value (which is now the bare IP point value invested into the Skill) against a complex table every time they want to actually use the Skill. It’s probably entirely superior for usability purposes to calculate the Skill bonus (the actual number you add to the dice roll) in advance rather than suggest that the player does it anew every time the Skill is used.
I think I get what you mean by the ontological implications, and I can say that as far as I’m concerned, the IP and the Skill value are one and the same here, as they are in Ars Magica: you may at any point convert the Skill value into its equivalent number of IPs to do operations, and then convert back when necessary. I’ll be using this process later when I delve into advanced Skill procedures; there are things you can do in a naturalistic Skill system like this that are best performed on the raw IP value of the Skill rather than the Skill’s rank.