Quote from: Lunamancer;895805I understand that part of what you're doing. I just don't think it flies. You don't get the sense of simulating it without actually playing it out round by round and making the choices along the way.

...

The point is, you cannot sum up the experience of playing the thing out into a single die roll. The experience of playing influences choice. And choice is one of the variables in your single-roll redux. You're bumping up against a conceptual impossibility.

Sure, I'm not looking to substitute either. Nor do I seek to reduce the whole adventure to a single MTTD roll that determines success or failure. It's similar to doing a mathematical transformation on the model of a car's suspension and analyze the performance that way. Sure, every road is different, every road will have different potholes, turns, and features in general. But it's the overall performance and not any particular instance I'm more interested in analyzing.

In the particular case of game combat it's already been observed the rather irrelevant nature of initiative. So spending a lot of points on initiative bonuses while having low hit points or chance to hit is not good, although this may seem the best course of action at first glance. Hit first! Hit first! Well, turns out not so important after all.

Quote from: Manzanaro;895730I honestly think many "old school" GM's fudged like fucking mad to avoid TPKs, or at the very least were extremely liberal in allowing PCs to escape death via GM fiat. People get mad when you say that, but it is what it is.

So here's something I've been meaning to toss at you, and now is a good time as any...

I've got this, I guess you could call it "mega dungeon" Gygax wrote in his later years, The Hall of Many Panes, which is definitely the most fun adventure module I've ever come across. It has an intricate backstory, the relevant part of which is that the precipitating events that lead to the PCs involvement in the adventure is a squabbling among gods. Good vs Evil. But one in particular, a neutral trickster god--think a super-powerful leprechaun--has tweaked things a bit for his own amusement.

Now one of the instructions in the module is to allow a one-time "Get out of TPK free" card where the trickster god rescues the party. He mocks them for their lack of brains but says they're giving him too many laughs to let them die just yet, so he's going to save them just this one time. He lets them know next time he'll just let them rot in their graves and find another group of mortals to entertain him.

Nothing about this contradicts my expectations considering we're talking about a fantasy world with active deities. But the question is, is this fiat, or is this just part of how the world works?

And let me follow up with this. Wishes. These are powers that D&D characters can happen upon by-the-book. Hell, they can even learn Wish as a spell and use it to their heart's content. They are part of the rules-as-written, and in fact the rules do have some proscribed limitations for wishes. But aside from that? Within those bounds, the players just get to make up whole cloth the effect of the power. And mind you, the PCs certainly can and do use wishes for things far, far less subtle than a suave GM fudging things to save the life of a PC (something within the bounds of what a wish can accomplish).

Wish does all sorts of things--usually of even greater power and breadth--normally allowable by meta-game resources like fate points. And let's not get hung up on the player decides the wish vs the DM decides divine intervention. The DMG 1st Ed actually does have rules and a formula for determining the probability for divine intervention. So DI, in a sense, is just a wish that the GM gets to word.

So, how about them wishes? What are they? Fiat? Meta-game devices? Narrativist rules? Always bearing in mind, they are grounded in things specifically kosher to the game world and explicitly BtB/RAW.

Quote from: Saurondor;895820In the particular case of game combat it's already been observed the rather irrelevant nature of initiative. So spending a lot of points on initiative bonuses while having low hit points or chance to hit is not good, although this may seem the best course of action at first glance. Hit first! Hit first! Well, turns out not so important after all.

Well, that's going to depend a lot on the system.

D&D--which is usually the poster child of excessive hit points that lead to the long battles of attrition you describe--is in reality full of one-hit kills. A level 1 fighter with 18/00 strength wielding a two-handed sword can potentially do up to 16 damage against small- and man-sized creatures, or up to 24 against larger than man-sized creatures. 88% of ogres--which are on the tougher side for a 1st level group--will drop to a max damage attack like that. Higher level fighters against less than one HD creatures also get to sweep attack, essentially a chance to hit a number of creatures equal to the fighter's level. Of course, anything less than one hit-die is a candidate for a one-hit kill. The potential of a 5th level fighter taking out 5 goblins before the goblins even get a counter-strike is pretty huge.

In Lejendary Adventure, Speed modifies initiative as an optional rule. I don't even use that option. However, the parry option is only allowed if you win initiative. Unless you have sufficient speed and skill to allow otherwise (speed 13+ AND either Minstrelsy or Swashbuckling skill of 31+). Skills advance faster than attributes, so the extra damage of highly skilled characters accumulates at a faster rate than you gain Health points. That first hit, then, becomes more and more important as you move to higher levels of play. Since initiative is in essence a glorified coin toss, the ability to parry becomes vital to having your skill win out against the coin toss. And thus investing in Speed, Minstrelsy, and Swashbuckling has more and more of a pay-off as you move to higher levels of play.

At super high levels of play, it gets even more complex. A combination of high Speed and high Weapons skill allows for multiple attacks. When taking multiple attacks, though, you sacrifice some of the damage bonus due to high skill. Damage is a de facto degree of success. A successful full parry must beat the degree of success of the attack. A partial parry still allows some harm through. A parry with a degree of success substantially greater than that of the attack also allows for a free counter-attack. So then the question becomes how many total actions to take, 1, 2 or 3, and of them, how many will be attacks, how many will be parries. It always comes down to a guessing game to some degree. But the key is to manage your risks. You don't want to leave yourself exposed to a one-hit-kill. And that means if your opponent gets in that first strike on you, you have that much less wiggle room in formulating your own strategy. And so that severely impacts your ability to effectively hurt your opponent because the fewer options you have, the better your opponent knows how to optimize his actions against you.

LA's dodge maneuver is also both initiative- and speed-dependent. If you win initiative, your odds of dodging are 100 minus your opponent's speed. If it's a tied initiative, your odds are 4 times your Speed. And if you lose, your odds are equal to your Speed, with a bonus if your speed is greater than your opponent's. Consider average human speed is about 10. So you're talking the difference between 90%, 40%, or 10% of success, all based on initiative.

Anyway....

I also have my own initiative procedure I port into whatever game I play, made to work with that game's rules. Again, I don't have Speed automatically modify initiative. When closing to striking distance, weapon length determines initiative. Weapon speed (different from character speed) breaks the tie on tied initiative. However, one option is to hold an action for use later in the round. Winning initiative obviously gives the greatest amount of choice in doing this. You can attempt to use it to interrupt someone--this is my way of giving turn-based combat a more interactive and real time feel--when attempting an interrupt, a speed check comes into play, if successful, you get to act before the person your interrupting. If not, then instead your action just comes as a quick reaction. It depends on the nature of the action, but interrupts, when successful, often do have the effect of negating the action they're interrupting. So again, initiative can be potentially big, and being faster has distinct advantages.

[ATTACH=CONFIG]23[/ATTACH]

This is the graph for the 90hp 10% (grey) , 10hp 90%  (orange) and 10% 10 hp (yellow) runs against a character with 20hp. X axis is rounds survived and Y axis is accumulated percentage.

[ATTACH=CONFIG]24[/ATTACH]

Here's an interesting graph of what happens when I make the decision to disengage upon taking 50% hit point damage or more. The 90% chance of taking 10 hp is impacted significantly as show by the orange line. The 10% chance of taking 90 hp is unaffected as there's never any decision point, I'm either alive or dead after taking 90hp. Yet the 10% chance of taking 10 hp behaves just like the 10% chance of taking 90 hp when I set a disengage threshold of 50%. I disengage because I'm either dead or 50% down (took 10hp = 50% of 20hp). So two apparently distinct situations behave quite similarly when run over and over again one million times.

Quote from: Saurondor;895878[ATTACH=CONFIG]23[/ATTACH]
This is the graph for the 90hp 10% , 10hp 90% and 10% 10 hp runs against a character with 20hp. X axis is rounds survived and Y axis is accumulated percentage.

Either I'm not following your graph or the rounds survived are incorrect. The chance of surviving N round against the low probability, high damage guy is .9 raised to the Nth power.

[ATTACH=CONFIG]25[/ATTACH]

Also against an opponent with only 20hp the odds of survival for an attacker doing 90hp with a 10% chance to hit is the same as a graph for an attacker doing 20hp with a 10% chance to hit. Everything over 20 is just overkill.

Yup, that's about right. The following table shows those odds as derived from the graph data which shows the odds of not surviving (perishing) in said round. It's 1 minus your stated odds.

1 minus the odd's on the graph in turn returns your odds:

rounds   10-90
1   90%
2   81%
3   73%
4   66%
5   59%
6   53%
7   48%
8   43%
9   39%
10   35%

Quote from: Lunamancer;895814The point that I made 80 pages ago is that it's because of the lack of transparency of the author's mind that if the author wants to produce certain effects--like maintaining suspension of disbelief--he MUST use certain narrative conventions.

Which is one reason for the saying “truth is stranger than fiction.” Some things that happen in reality are so improbable as to jar the reader. But we accept the truth, because the data available to show those things, though improbable, really happened (e.g. surviving a fall from a great height without a parachute).

But as you point out, the situation can be different in an RPG. In a game where die rolls and rules are transparent, a player is  more likely to believe that the opponent with a 10% chance to hit did so three times in a row because of an improbable string of rolls, because the player saw the dice rolled. Whereas an author who just tossed in a 1 in 1/10,000 fluke may be met with great incredulity, most especially if such a fluke violates the reader’s expectations for the story.

Quote from: Saurondor;895897Yup, that's about right.

But the top line (labeled 10-90) on the graph shows 100% damage, i.e. death, after 4 rounds. I'm confused.

Quote from: Bren;895902But the top line (labeled 10-90) on the graph shows 100% damage, i.e. death, after 4 rounds. I'm confused.

I believe you refer to the orange line, right? That's the 90-10, 90% - 10hp of the time. You're right it was quite confusing the way it was presented.

I edited the post to indicate colors

Quote from: Saurondor;895907I edited the post to indicate colors

Thanks for clarifying.


On an unrelated thought, when I think of a foe with 10% chance to hit doing 90 pts of damage I think of something like the Cyclops that Odysseus blinded. Could not hit for shit, but if it did hit you SPLAT! I would expect that in anything that wasn't single combat to the death, that it is the companions of the guy who just went SPLAT! who may be able to decide to retreat based on that new information. That's another reason actual combat can be complex. I don't have to make decisions solely on what happens to my PC, but also what happens to the other characters around me.

Quote from: Manzanaro;895797For the example of saving a pet NPC? That is exactly what I am saying I would never do: override the simulation results with an authored outcome.

As far as your last remarks go, I think that is much more nebulous territory. For example, when I talked about how I would create a rival circus with a sinister ringmaster and etc. you remarked that it sounded like an overly dramatic setup. I interpret that to mean that my setup seemed based in overly familiar narrative tropes.

But here is the thing, I am a big fan of overly familiar narrative tropes in the context of RPGs, AS LONG as these tropes are not tied to an overly familiar and preordained plot structure or string of outcomes.

This is what I was talking about in earlier remarks when I referred to the principle of literary deconstruction. We take tropes that are at least partially familiar but than we decouple these elements from the typically associated narrative causality, or in other words, we play to find out what happens rather than force our way to some preconceived result.

Ok, then I misunderstood you. I thought you mean with "inside the simulation results" picking manually from the whole range of theoretical results the simulation system has to offer, thus staying in this result range (and thus each single result indistinguisable from a roll) but dropping relative chances for other criteria.

The thing with seeding - especially post start seeding but not only - is, that there are also chances and simulation focus based assumptions involved, that will heavily influence game and gaming decissions. For example such a big hotpot of dynamic events liek this rival circus will (probably aggressively) radiate out to its environment and thus become the dominating element of the game. It is not anymore about managing your own circus with some events strewn in like a "Normal" rival circus with perhaps one special effect but about one freaking monster circus running around and the characters having to do something about it with having to manage your own circus that probabyl is best delegated to your henchmen, because the chars are probably the best suited ones to deal with the monster circus or it is mandatory for soem kind of self preservation.

Quote from: Bren;895911Thanks for clarifying.


On an unrelated thought, when I think of a foe with 10% chance to hit doing 90 pts of damage I think of something like the Cyclops that Odysseus blinded. Could not hit for shit, but if it did hit you SPLAT! I would expect that in anything that wasn't single combat to the death, that it is the companions of the guy who just went SPLAT! who may be able to decide to retreat based on that new information. That's another reason actual combat can be complex. I don't have to make decisions solely on what happens to my PC, but also what happens to the other characters around me.

... Or even what you anticipate will happen based on nothing but your sense of things. Do you really need to see the cyclops man-handle the men like rag dolls, cracking their skulls against the ground to eat their brains in order to realize the cyclops is a big, strong dude that you don't want trying to hurt you? The GM's description should be enough. And this is where narration can have a very tangible impact on the game without cutesy little "narrativist" rules. To a degree, virtually all GMs do this automatically. They match the narrative, not necessarily to what is, but to the player's perceptions, e.g. "The room appears empty," when it is in fact not. But it's possible to communicate more subtle ideas through the narrative which can cause players to choose differently than they otherwise would have given a different narrative.

Quote from: Lunamancer;895955Do you really need to see the cyclops man-handle the men like rag dolls, cracking their skulls against the ground to eat their brains in order to realize the cyclops is a big, strong dude that you don't want trying to hurt you?

Given the varying levels of realism in different RPGs I need may need to know a bit more than just hey that guy is big. The differences between a giant in D&D Greyhawk and Runequest Glorantha are enormous. I've certainly run D&D characters who could easily survive one or more  hits by a cyclops or giant. Whereas I've never seen a Runequest character who could survive the damage done by even a modest sized giant. That match up is close to the 90 pts damage vs. the 20 hit point character that was graphed out.

But I don't disagree with your basic point. The GM should be providing some clues that the blinded cyclops isn't very accurate, but is very mighty.

Quote from: Bren;895960.... The GM should be providing some clues that the blinded cyclops isn't very accurate, but is very mighty.

Exactly, in my view characters would have an inherent feel for how dangerous a foe appears, especially experienced ones.  Yet people can misjudge, as a GM I base how much info I give on how well known the creature is, or legends of it.  I use a skill based system so have a broad skill called lore or myth that covers such things, along with a lot else.  If it is unknown it's based off the "level" of the character, more experienced characters getting a more accurate clues.  Like if the blinded Cyclops could compensate with hearing or smell, how fast and accurate it seems to be etc.

No need for a narrative or someone to die, but nothing settles the theoretical question of a creature's capabilities like death.