Upgrade your account to a Community Supporter account and remove most of the site ads.
Rocket your D&D 5E and Level Up: Advanced 5E games into space! Alpha Star Magazine Is Launching... Right Now!

Reply to thread

Message: <blockquote data-quote="Ainamacar" data-source="post: 5846811" data-attributes="member: 70709">Sorry about the delays, real life has been hectic!</me edits post, whistles nonchalantly> Sir! You are mistaken, and I take umbrage!!! <img src="https://cdn.jsdelivr.net/joypixels/assets/8.0/png/unicode/64/1f642.png" class="smilie smilie--emoji" loading="lazy" width="64" height="64" alt=":)" title="Smile    :)"  data-smilie="1"data-shortname=":)" />I think you're both on-the-spot on the difficulty of doing this relationship well.  To make the system work as a whole we'd need to generate some sort of conceptual consensus (or at least clarity) for what roles the DC and the level of training each play.  These are probably the biggest obstacle to making this kind of system work well, especially in D&D where such things don't have much of a history.  In this post I'll address KesselZero's question of how I envision things working.  That will eventually be followed by a post more focused on presentation and other ways to conceptualize the system.  Eventually I'll get to complications and the suggested tweaks.So far I've been describing both DC and training level with the vague notion of "difficulty", and in fact using that similar language was a poor choice on my part.  I think it would have been better to describe the DC as "difficulty" and use different language for the UCEM part.  I think of the latter as serving two primary purposes: first as comprehensiveness/extent/magnitude, and second as a soft threshold for who can achieve results of minimal comprehensiveness.  Speaking generally, more comprehensive results are usually more difficult in an absolute sense (i.e. final probability of success), but there is a sense in which that change in difficulty is incidental to the task itself.  The second aspect, that of soft threshold, is a way to set just who can access the minimal level of success without significant additional aid or resources.  (You'll notice that this is a very results-oriented conception.  Why I arrived at this particular concept, for better or worse, will be in my next post.)I think the aspect of comprehensiveness is fairly straightforward.  Whenever we can imagine a person doing a particular task but succeeding at qualitatively different levels we have an issue of comprehensiveness.  Frequently it can be phrased to answer questions like "how much?" and "to what degree?"  It is just how much of a translated text is understood.  It is whether one can sense a presence or pinpoint the source.  It is the level of detail known about a thing, and not necessarily 1, 2, or 3 details but "very little, quite a bit, or essentially everything."  In the traditional d20 skill system, it replaces the qualitatively more awesome results previously given when beating the DC by 10 or some other large margin.  The issue of comprehensiveness doesn't really care what precise number of successes represents the minimum amount of success, it cares about how additional successes beyond that matter.I readily grant that analyzing the appropriate degrees of success on the fly can be difficult, so skills where this would be common should probably be written beforehand with enough clarity that the DM knows how to apply them without thinking through the situation in detail.  Things like Decipher Script and Track are perfect examples, and I think we've been mostly concentrating on these cases because they are the most complex.  The DM should, however, have enough guidance to improvise freely.  We all know how this works: sometimes a player would get a 33 on a DC 21 check and look knowingly at the DM, as if to say "Look how I did! What extra stuff do I get?!"  The DM didn't think about it beforehand, there's certainly no special rule for this case, and so now he has to think about whether beating the DC by 12 on this random check is worth something extra, and what something extra that might be.  The good part about this classic d20 scenario is that it deferred such considerations until they were necessary.  The bad part is that it gives the DM little guidance in actually answering the questions once it arose.  I'd like to keep comprehensiveness in these situations as a way to defer the questions (even more so than before given Skill Tricks) while giving the DM much stronger guidelines to determine exactly when something extra crops up (an additional success) and how much it's worth.  Really good DMs have always done this sort of thing naturally, but there are lots of DMs who could use a tool to help them.How does this comprehensiveness aspect relate to the DC?  When I said above that the increased difficulty from increased comprehensiveness is incidental, what I mean is that the task itself is unchanged.  For example, in the decipher script example both the mentor and student go about the task in essentially the same way.  The process is unchanged for both of them, and it is the process (not the result!) which is represented by the DC.  At the table, however, I tend to think of results first (what do the PCs hope to accomplish) and then set the DC from that.  I think this is common.  One can therefore see that setting the DC based on the minimal degree of success (the only level of success the DM should have to consider on every single check) is functionally the same as the old version of looking at the desired result and setting a DC from that.  And regardless of the number of successes required to achieve that minimum result, for a person with training at that level a single d20 is rolled.  So setting the DC with this rule-of-thumb isn't just the same conceptual process, it uses the same single d20 math with which we're already familiar.That leads us to consider the second aspect, where UCEM serves as a soft threshold which determines who can achieve these minimally comprehensive results.  I think this is where there is more difficulty getting on the same page.  (Understandably, since it isn't specifically addressed in my earlier posts, and the process of writing this post is what's really moving my understanding of it from the intuitive to the concrete.)  To explain it, I first want to strike a contrast between "soft threshold" and "hard threshold".  I consider the DC to be a hard threshold because a very small change in DC can lead to a very large change in whether or not a given task is impossible to achieve or impossible not to achieve.  (If that weren't the case we probably wouldn't be thinking about this topic in the first place.)  This means the game is inherently unstable to small variances in bonuses between characters for very easy or very difficult tasks, as well as for the same character facing difficult challenges with a small variance in DC.  The simplest solution is to stay away from the edges of the d20.  That works, but it also means almost anyone can try anything and have, perhaps, an unreasonably large chance for success or failure.The math of multiple simultaneously-required successes quickly makes everything less likely than it was before, both for those with large or small probabilities of success on a single roll.  This admirably keeps the undertrained away from things which should be out of their reach (without completely excluding them) but simultaneously suppresses the likelihood of characters with even very good bonuses succeeding.  Such is the nature of p^n!  This is why highly trained characters definitely need fewer successes to achieve more comprehensive results.  The bonus dice for less comprehensive results are perhaps less necessary in a statistical sense (either way the probabilities of success will be far from the edges) but it does serve to emphasize the difference between characters despite the relatively small range of DCs.  Therefore DCs can be kept moderate throughout the entire game, but the probabilities for succeeding can move smoothly toward 0 or 1 for all tasks for all characters.  A soft threshold in the minimum number of successes required to achieve any level of success is a way to keep all similarly-trained characters achieving at a similar level amongst themselves (despite some variances in their bonuses) while enforcing a more reasonable gap, when necessary, from those with considerably less training.So how should the soft threshold (minimum number of successes) and the DC relate?  The basic question I would ask is this: Is the minimal useful result I can imagine for this task beyond the basic resources of an Untrained or Competent character, for any DC I might choose?  If yes, then select the level of training one ought to have to be able to succeed on a single roll.  If no, then leave it as 1.  Then set the DC for however easy or difficult it should be for someone with the minimal proper training.  This is a macroscopic, conceptual, and results-oriented view.  To use this sytem well I think it should be the first thing the DM considers when making a check, and since it makes sweeping generalizations it should be an easy one.  Whatever the result of the deliberation, one set the DC such that a person with the minimal level of training deemed appropriate succeeds with the desired frequency.  All other considerations should be reflected in the DC.  To set the DC one does not worry about how difficult the check is for anyone except the person with a skill corresponding to the minimal result. (For example, the DM decides that a particular Diplomacy check would normally require an Expert in diplomacy because the "target" has a strong bias against the PCs.  He then reasons that a typical expert could probably succeed about 30% of the time and sets the DC accordingly.)  Almost always this should be 1, because almost always characters can meaningfully attempt a task.The cases which appear most counter-intuitive are probably those where more than a single success is required, but the DCs are low.  I think these correspond to the cases where a person without training basically has no idea where to start, whereas a person with training finds it pretty easy. (A corner case, to be sure, especially in an adventure game.)  Consider something like basic calculus, for example.  A person without training in calculus would probably be at a loss for the meaning of the integral of x dx, much less how to calculate an answer. A person with training in calculus, however would find that incredibly easy.  Now, should that check have a large DC so the person without training can't get it while trusting the extra dice of the person with training to make it work?  Or should it have a low DC so the appropriately trained person gets it easily on a single check and the undertrained person needs extra resources?  It's clear that we could use either method and set the DC in such a way that the probability of the trained character (for example) is equal in both cases.  Conceptually, however, what is the more natural call?  In my mind, at least, it makes much more sense to determine the probabilities based on the person with training appropriate to the task.In summary, here is my proposed process.<ol> <li data-xf-list-type="ol">Determine effect of simplest useful result.</li> <li data-xf-list-type="ol">Determine if minimal success should require special training or resources to make sense.</li> <li data-xf-list-type="ol">Determine the DC based on how a person with the least required training should fare.</li> <li data-xf-list-type="ol">Roll the check.</li> <li data-xf-list-type="ol">If more than minimally successful consider additional results or use Skill Tricks.</li> </ol>I don't think this is drastically different than the thought process we normally use.  Step 2 is more involved because instead of trained/untrained there is UCEM, but it is a gut check, not a deliberation.  Step 5 may actually be simpler than in the standard skill system, since "extra successes" is a firmer criterion than "rolled really high."Here are some ways the basic combinations of DCs and minimal successes might work.  Notice that I arrange them not by the possible number of results, but simply by the lowest training required to not require extra resources.<ol> <li data-xf-list-type="ol">Low DC, U or C lowest result.  The tasks of everyday functioning.  Most  checks at this level can be ignored unless an extraordinary result is  desired, or something hangs in the balance.  This is stuff like swinging  on monkey bars, preparing a non-gourmet meal, simple personal taxes,  detecting an obvious fib, convincing a person who wants to be convinced,  etc.</li> <li data-xf-list-type="ol">Low DC, E or M lowest result.  The specialized tasks that look really hard, but are actually easy once you learn them.</li> <li data-xf-list-type="ol">Mod DC, U or C lowest result.  The bread and butter of adventuring!   Everyone can contribute, but the highly trained will shine.  The bulk of  the game should reside here.</li> <li data-xf-list-type="ol">Mod DC, E or M lowest result.  Specialized tasks that look hard, and  actually take some effort. I think a lot of crafting, spell research,  and serious lock-picking will belong here.  For example, a masterwork  sword might actually require a master.  The iconic tracker who can find  tracks after a good rain has probably succeeded at this level.  Even the  highly-trained tend to seek out additional resources when possible so  they can benefit from their Skill Tricks, or because they are the only  one who is really capable of competing this task in the first place.</li> <li data-xf-list-type="ol">High DC, U or C lowest result.  These are the things almost anyone can  imagine themselves doing, but would not want to have to do.  It includes  the brutal forced march and other daunting physical tasks.  It is a  great spot for esoteric (but not necessarily complex to grasp) knowledge  of which even learned sages often know nothing or only the scantest  details.</li> <li data-xf-list-type="ol">High DC, E or M lowest result.  This is where truly epic tasks happen.   Wizards recreate magic long thought lost, perhaps even dabble with minor  artifacts. Knowledge that has been carefully concealed or obscured  probably occurs here.  Most parties will find it necessary to seek out  NPCs or go on special quests to accomplish these tasks.  Even parties  with highly trained members think that's a very good idea.  Success at  this level is noticeable to anyone who is paying attention.</li> </ol>Now, for some tasks with many levels of success and also many possible modifiers this entire scheme is tricky, especially when circumstantial modifiers could reasonably scale from "minor inconvenience" to "unavoidable obstacle".  I think tracking is emblematic of this sort of situation.  For example, should "soft terrain" vs. "hard terrain" correspond to a largish change in DC or by increasing the minimal level of successes?  What about "trail is fresh" vs. "trail is very old"?  Given the complexity of this kind of check the DM will probably need specific guidance, and that means specific rules.  When I think about the meaning of expertise in tracking the thing that pops out at me from fiction isn't just the ability to see a track, it is the ability to correctly interpret subtle signs that others see but which simply don't register as important.  That suggests to me that the DC in this case is closer to "seeing something is here" and the minimum number of successes is closer to "knowing that it means something."  Time and weather tends to destroy the subtleties first and the obvious things later, which suggest to me that in this case those effects increase the minimal number of successes.  Creatures that cover their tracks are usually trying to cover the obvious signs, which increases the DC.  Creatures with special expertise in covering their tracks are also removing the subtle signs as well, or even using it for misdirection.  Therefore I think the may require a tracker to gain additional successes. (In general I think opposing creatures with equivalent expertise in a skill should be evenly matched whenever possible.  Not that I have a strong conception of opposed checks yet.)A version of track might then look something like this:I think this works more-or-less as desired, and with a coherent internal logic a DM could expand upon if necessary.  (For other complex tasks where the "right" way to construct checks is unclear, or even contradicts the usual rules-of-thumb, I think internal coherency has to be present to cue the DM.)  Almost any creature has at least some chance of help tracking on soft and hard ground, and should be able to follow them with time.  Just about any trained creature has a chance of getting at least a single success whenever tracking, teleporting pixie druids in driving snow notwithstanding.  A single success for following fresh tracks lets the game go on, while most creatures will have to devote time and/or team up if the trail is starting to go cold.  But they'll have a chance, and an expert or master can really help out.  A master ranger tracking a master druid might find himself on equal terms, with the druid leaving behind details only a master could interpret while keeping pace.The example above also has fixed DCs in it, so I guess I'll address the range I'd like to see here.  Basically, I think DCs would range from 5 to 25 for the entire run of the game, with most falling in that 10-20 zone.  I'd leave a little room for DCs as low as 0 and as high as 30 for truly special circumstances. At the high end that starts to get near auto-fail zone even for characters with a +10 bonus (+5 competent, +5 ability score, which is what I've been assuming is about the highest one can get. At the low end even the worst imaginable character with 3 in an ability score and no training has just a -4. (Assuming the 3e/4e method.)  The benefits of expert and master training are already very strong between extra dice and easier access to qualitatively better results, they get no additional skill bonus.That said, there is plenty of wriggle room in this math, and it doesn't really matter how it comes about.  If the skill bonus for Competence is +3 (which has some appeal to me), a 20 ability score is still worth +5, and characters gain +1/5 levels to trained skills for 20 levels, then the largest bonus is +12.  If characters get to increase ability scores, I might just let them use that for any advancement of skills beyond competence training.  Either gives trained people a sense of growth without leaving the untrained average people utterly behind.  If ability mods change from the 3e/4e version accounting for it shouldn't be difficult.  To me the important thing is to keep the spread between the absolute best and absolute worst bonuses somewhat less than 20, and the best and worst typical DCs at about 20.  There is still room for some autosuccess and autofailure at the edges, but nothing like the huge disparities between characters found in 3e/4e.Thanks for reading another long one!  More when I get a chance. <img src="https://cdn.jsdelivr.net/joypixels/assets/8.0/png/unicode/64/1f642.png" class="smilie smilie--emoji" loading="lazy" width="64" height="64" alt=":)" title="Smile    :)"  data-smilie="1"data-shortname=":)" /></blockquote>

Verification