couldn't "objective" DCs still allow for gonzo play (something similar to how Mutants and Masterminds, a superhero RPG, gives you reliable abilities that are definitely more gonzo in nature)?
And, secondly, can't you have a good chart for "stunts" or "improvised actions" that use the "objective" DCs as a guideline for the rest of play?
4e does have a chart for improvised actions, though it's based on scaled DCs and "genre logic" rather than on "objective" DCs.
I'm not familiar with M&M other than by reputation, but I have been given the impression by some posts I've seen about it that a lot of the gonzo is introduced via its Hero/Action(?) Point mechanic, which seems to be more about "genre logic" than objective DCs.
Whether or not that impression is accurate, my feeling - and it's based on experience, not a priori conceptual reasoning - is that one you are using objective DCs adjudicating the gonzo elements of D&D-ish high fantasy becomes harder, and the overall pressure is towards conservatism, because reality is the baseline for the setting of objective DCs.
For example, how easy is it for someone to shove their hands into a forge and hold a red-hot artefact still as magical (as well as mundane) energy builds, and dwarven artificers try and grasp it with their tongs? My feeling is that in the real world that's impossible, or very close to. And then the pressure is on to look for abilities (or spells) that give magical protection against heat, and the like - and then we're out of the realm of improvisational gonzo and into the realm of operational play and the management of magical resources.
Another example that came up on these boards early in 4e's life. Someone was complaining about their mage getting swallowed in a purple worm encounter - the worm struck with surprise, got to go twice before the mage, and then gulped the wizard down. So the player never got to act (under the ruels at the time, only melee basic attacks were permitted in a worm's gullet) and the PC died. [MENTION=386]LostSoul[/MENTION] asked something like "Why didn't the Rogue dive into the worm's gullet and pull the mage back out?", with reference to the DC for a basic Acrobatic stunt as being 15 (per the 4e PHB - in the post-Essentials environment, where some of the notation for DCs has been standardised, I think we'd be talking about a Hard level-appropriate DC.)
This suggestion was met with a lot of scepticism - how can a rogue dive into a purple worm's gullet and pull someone to safety? The objective DC for that is "ludicrous and impossible". But I think LostSoul's question was posed within a presupposed framework of scaling DCs and genre logic. And I do think that this is the framework that tends to open the door to gonzo play, rather than tending to bring it back down to earth, and into the realm of operational management of magical resources.
Reliable DCs allow players to reliably gauge their skill against the DC, and make informed decisions on whether or not they're able to shape the story reliably.
Can I ask how the skill system that is essentially decided by DM fiat emphasizes player agency more than, say, BW's approach?
Most importantly, the DC of things being defined. That's by far the most important thing to get done, in my mind.
<snip>
That's the main thing I want to see in the skill section in 5e. It's important, and gives much more "player agency" than what pemerton described in his reforging of Whelm, in my opinion.
4e (and HeroQuest revised, to which I've compared it) don't have DCs set by GM fiat. They have DCs read of a chart which has entries specified in metagame terms (in HQ, by reference to the pass/fail cycle; in 4e, by reference to level).
In this situation, the player is still saying "can I do this?" and you, as DM, get to say yes or no (this doesn't strike me as strong player agency). Then, you set the DC (again, he doesn't know what this will be until the DM decides, so this doesn't strike me as strong player agency either).
I don't see any difference - as far as this particular issue is concerened - between setting DCs by reference to level, and setting them by reference to "objective" factors.
And I don't see any difference between combat and non-combat, either.
So in classic D&D, or in BW, the GM assigns a monster AC and hp (or, in BW, an armour rating and injury thresholds) based on "objective" factors. When the players (via their PCs) come to deal with the monster, they can either try to infer to the numbers from the GM's description, or they can muck in, have a go and learn by trial and error. (Perhaps there are also "monster knowledge" mechanics that mediate between a PC knowing the fictional state of affairs and a player knowing its mechanical expression.)
In 4e , the GM assigns a monster AC and hp based on level (in HQ, rather than AC and hp it is ability bonuses that are assigned in this way, and on the basis of the pass/fail cycle). When the players (via their PCs) come to deal with the monster, they can either try to infer the numbers from the GM's characterisation of the situation ("You see a fearsome beast", "That's going to be hard!", or convesely "You don't think that will cause you much trouble"), or they can muck in, have a go and learn by trial and error. (Again, there can be monster knowledge mechanics to mediate between a PC knowing the fictional state of affairs, and a player knowing its mechanical expression. 4e has these. HQ revised doesn't, but its system is simple enough that it doesn't really need them.)
If we turn away from monsters to (say) jumping a pit or climbing a wall, you can describe it to the players with enough precision that they can identify the objective factors that will set a DC (this is how BW approaches it - and it is taken for granted that it is the GM who has ultimately authority over DC-setting, though players are entitled to ask for one advantage die if they can point to some relevant advantage in the way they have narrated their PC's approach to the task; the player isn't actually told the DC until s/he has committed his/her PC to the action, at which point it is too late to pull out). Or, you can tell the players "that looks easy" or "that looks hard" - or even just state the DC - and again they can infer to the difficulty of the task for their PCs.
For me, player agency is not about transparency of DCs as such - I use a range of approaches to DC disclosure, sometime relying on the players to infer them, sometimes stating them outright, my general goal being to keep the excitement of the game alive. (It's a bit like a creature's immunities - generally I leave the players to discover them, or to have their PCs ascertain them ingame via a knowledge check, but occasionally I will state them outright because I think it will heighten the tension.)
For me, it's about the players being confident (i) that if they engage the action resolution mechanics they have a meaningful chance (and, for me, this is what level-based scaling achieves - in BW, the alternative approach is taken of always making failure a viable option), and (ii) that once the action resolution mechanics are engaged, the consequences will be binding on everyone at the table. There are no fudgings, rewrites, "Roads to Rome" etc. Particularly in respect of (ii), I think this is straight-down-the-line Forge-ist hostility to the GM suspending the mechanics "in the interests of the story". It's about the GM maintaining authority over scene-framing (and to a significant degree over backstory), but not asserting authority over plot.
Essentials, for me, took one major backwards step when it stated (RC, p 9) that one of the roles of the GM is "[to guide] the story. . . At times, the DM might alter or even ignore the result of a die roll if doing so benefits the story." This replaced the earlier characterisation, in the PHB (p 8) of the GM "adjudicat[ing] the story" in those situations "[w]hen it's not clear what ought to happen next". I think the PHB formulations is a little elliptical, but rougly right (relative to my interests and desires). I think the RC formulation is a recipe for suspending the action resolution mechanics, for railroading, and for the overriding of player agency.