Forum changes: Editing of posts has been turned off until further notice.

Main Menu

[Delve] fast-forwarding through meaningful decisions?

Started by David Berg, February 05, 2010, 01:48:30 AM

Previous topic - Next topic

Callan S.

David, huh? Your first couple of posts try and solve a problem that isn't the real problem, as far as I can tell. I start getting down to the real roots of what's the problem AFAICT and importantly, I thought you were with me on that. Now you've snapped back on how to fast forward again, or something or other from your first post. To me, you've just regressed.

And yes, responding to Josh can take up too much space, I'm fully taking that on board. But of the prior two posts to yours, mine was addressed soley to you.

It's not some mistake of posters to diverge from an original post with an incorrect assumption. I would prefer you just say your not mistaken and are right about your original two posts than seem to be going with me, then suddenly call me off topic.
Philosopher Gamer

David Berg

I thought all your ideas were great ones to explore, but I think I've failed to get anywhere with them.  We didn't really diverge until your last post, so I'm sorry that my request to change focus came out of the blue.

The facts of "What would the characters do in this situation?" are something I won't compromise.  However, the coverage of such actions is more flexible ("How much play time gets spent on it?").  Perhaps that flexibility is insufficient to make Dan & John happy, and what I really need to do is avoid handing them situations with tons of decision-making and no drama or time-pressure.  I'm still hoping that another solution might be found, though...

Those are all cool things that could come up in Delve.  A few of them already have!  But this sounds to me like an account of situations wherein the players can get what they want when they want it.  I'm not sure how to extrapolate from your ideas to cover when that's not the case.

Maybe giving the characters a safe place to stash stuff should be an early-on GM priority?  But that's not much help when you go on an adventure far away... which is something I don't want to rule out...

Yeah, I wish my friends could play late to get to a good end point!  Damn work schedules.
here's my blog, discussing Delve, my game in development


Quote from: David Berg on March 05, 2010, 12:51:03 AM
But this sounds to me like an account of situations wherein the players can get what they want when they want it.  I'm not sure how to extrapolate from your ideas to cover when that's not the case.

I was mostly trying to hedge out the class of situations associated with this problem, so you keep the game as it is but just tune setting specifically not to hit these buttons.

But on the core of what you were talking about, it seems like there is another thing I hadn't considered until that last post; you currently have a lovely dial that doesn't apply to all situations; how do you normally deal with setting the dial to high speed in problem solving situations? Has it come up before? Because every time that you go to a fastforward and then slow down again, there is that loss of fictional resolution that has to be built up again; so if you go "we fight some dudes, we do really well, wait, zoom in" then it's sort of like scene framing, filling in those gaps that have been created by all the ways you could have gone.

In that sense dealing with this is similar to a problem you will face many times; where an activity done at a low detail level has impacts at a high detail level. If you want to make the fast-forwarding a pure aesthetic choice, then you need to insure that people are not tactically disadvantaged by doing so.

For an example of setting-mandatory dial shifts, say people are rapidly running and weaving their way through the market streets of a town only to be stopped by a series of guards. Now at that point the pace of the story has shifted. Should that dictate where to put the dial, or should they be able to go "ok we bribe the guards and it costs us .... this much"? It seems like your game by nature pushes towards the real-time point on the curve, would it still be a worthwhile game to you if it could all be completed in high speed? Also if games can be played at that speed through high sensitivity decisions, then coming off the other end is much more like scene framing from PTA, so maybe it's natural for players to shift style a little and explain how there characters got there, because it could be part and parcel of shifting the dial down.

David Berg


The title of this thread covers the dial's problem area.  When there aren't meaningful decisions to be made, no one minds skipping ahead.  Resolving what was skipped over is usually a combo of (a) GM asserts whatever outcome he deems most likely and (b) GM rolls randomly among multiple plausible outcomes.

Identifying the class of problematic situations sounds like a good idea to me.  I'll try:

"Lots of leads to follow and questions to ask with no time pressure or danger" is a contender.  Sometimes, though, it can be sped through.  We've played through interrogating every possible source on the haunted castle when that was fun, and jumped ahead to, "So what did we learn?" when it wasn't.

Actually, that latter has gotten slow when I've said, "Tell me what you cover in the fast-forward, so I can tell you the results."  Then they start thinking of specific questions for specific people, etc.  Sometimes I'll offer, "Do you just spend however long it takes to talk to everyone you can reach?" but they'll say no when I describe what that would consist of and they deem it silly.

Maybe I could try leaving more gaps in the fictional data sequence?  Like, it's okay to not know what the characters were doing during "a week of info gathering"?  These thing always seem to matter later, though.  When you try to go incognito, then all of a sudden we need to know whether you went house to house earlier.

I think the problematic situation comes down to "many options, but each with consequences".  But that's also kind of the meat of the game -- that players follow their interests, self-initiate quests, and that what they do matters in the setting.  What I might need is just a way to help them choose in a quick and satisfying matter.

Hmm.  In the real world, how is this done?  What's the most efficient/enjoyable way for 3 or 4 people to form a plan of action?

Ways to help Dan and John agree when enough is enough would go a long way.  A quick reference for "cover your bases" might also help.
here's my blog, discussing Delve, my game in development

Paul T

Some really interesting discussion here.

I'm with Ron, I think:

If you value your fun over the integrity of the fiction, you need a metagame solution, to step back for a second from the game and figure how to make it work.

I think I might have done something like:

"We all want to move on [the next event], right? So, let's skip ahead and then we can start next session as a flashback at the moon rock. This means we have to agree right now that whatever you decide to do with the rock, it won't be helpful to you in [the next event], which we'll play tonight. Cool?"

The next best thing would probably be to engineer some in-fiction excuse for moving on--for instance, why couldn't they take a chunk out of that rock and carry it with them, only to shape it later?

David Berg


Your solution might be perfect for some of the "we're not in the mood for this right now" moments.  Someone would simply propose the flashback and then the group would agree on whether or not "no impact on next activity" was plausible.  If it wasn't, though, we couldn't use it.  My hope is still to achieve fun without ever sacrificing the integrity of the fiction.

As GM advice, "make resources portable" might be one of the options toward "avoid situations of many decisions with no pressure"...
here's my blog, discussing Delve, my game in development


I didn't state it very clearly in the last post, but my brainwave was that with the dial, you imply that the game can handle every situation that will come up (and it's associated GM-player interactions) at 15 different detail levels. So you can pull back from that, and say, "actually we won't do coffee drinking in bullet time!" but in a way, it would be awesome if you could. :)

In terms of making the game work interactively at that the scene summary level, you probably need to set up the basic abstractions at that scale, like (((attack)->fight)->war), so that people can say, "ok I do this" and talk at the appropriate level of abstraction. At the moment the different speeds have cues as to what to ask for, and how to talk IC, but not how to describe your actions. I know I'm just restating the problem here, but hopefully that will help you know where to start.

Hopefully if you get something like that going, it will hopefully still feel like delve in the faster parts, but it will be more like the memories of a busy day; a feeling of the motion, and a few snatches of detail here and there.

The slowdown system makes a lot of sense to me, though obviously it has tactical implications: How we partition the possibility space decides the weighting of a certain type of outcome, you know, "do we win this way this way or this way, or do we lose? roll a d4". You can go further with weighted random switches, or you could specifically watch for that and just sticking duplicate states down. One rule I like is that when dealing with things in the general view, you just turn the situation specific tactics someone would have done into a random factor, so that we know they did something well or badly, just not what it was. It means they can state their strategy from a low detail scale, then roll for extra twists that tactics put into it.

I get a feeling your making things quite hard for yourself by doing that kind of transition though, because although from a system modelling point of view it's cute to automate players responses and your own moment to moment whims with randomisation, it means that you will be filling in a lot of the content that would normally be being created by the players. An alternative? How about you have players suggest what their character might have done when it becomes important, you suggest an alternative where they weren't so on the ball, and they roll to see if their character was lucky enough/had sufficient foresight to predict this situation. In other words unreliable prediction of the future becomes unreliable ability to retroactively use current information. It's a more general version of "your character would have known about this". It would mean that the moon-rock-carving would be spread throughout the later scenes, in lulls or when relevant. How much you could keep reusing a scene would probably need to be kept track of; as they fill up earlier scenes with extra details, the GM would have his priority being to insure that it could plausibly have happened then, pretty much like normal!

The requirement to state in general terms what your guys were up to would also limit this, and the short dialog when deciding it would provide an inspiration for actually coming up with flashback suggestions.

I reckon you won't have to systematise people's coming to agreement, providing the learning curve is smooth enough, and you make it so it's not all meat all the time!

Jeff B

Regarding the comments about John, I myself am unable to judge without hearing the tone of voice and feeling the dynamic of the room, whether or not his attitude is causing things to drag.

The problem as I see it is in the stone itself.  What I"m writing is not intended as a criticism of the GM in this case, but perhaps illustrating how the GM has laid the perfect trap to catch both the players and himself in a quandary.

The trap functions on various levels.  For one, the item is non-portable, so it must be dealt with WHERE IT IS.  The stone is an open-ended exercise in creativity:  It can be used in a potentially huge number of EXCESSIVELY huge number of ways.  There is the sense that using it incorrectly will be a waste of a terrific resource.  There is also a sense that failing to exploit the opportunity now may result in complete waste of the opportunity.  But there is also no single, pressing need for an item so as to present an obvious choice or a productive platform for party debate.

In effect, the GM has placed an item and said, "This treasure is worth between zero and one million gold pieces, depending upon how you use it.  And you need to use it now [because it can't be moved]."  This piques the Gamist instincts of the players, driven to perform as well as they possibly can and reap the greatest possible gain.  But the way is completely unclear.  The truth is, they don't actually need any one object that desperately.  But it's unthinkable to simply waste such an opportunity. 

They are also being faced with colliding views of physics.  This stone is mythic by nature, but the techniques being employed are physics-bound, (e.g. the discussion of making a whole sword out of stone).  If the world is mythic, a fantastic sword should be possible, made entirely of stone.  If the world is not mythic, the stone probably should not exist.  This stone is a collision point between imagination and fact which causes discomfort.

Solution:  1) Make a similar treasure but which is portable, so it becomes an "ace up the sleeve" of the party.  2) Make the location so secret that they have no fear of it being found and can return in the future, when they know what they want.  3)  Give the players a Wish to play with, instead of a physical object.  This stone is essentially an unformed wish, but because of its physical nature, they are almost certain to under-utilize its potential; they sense this and are unwilling to leave the stone, even though they'd rather be adventuring.

I might suggest this stone would be better used as a trap laid by a dragon to ensnare whole groups of adventurers at once, as they sit and fret about the potential uses, gradually losing track of their surroundings, eventually falling asleep there, whence they can be easily eaten.  The existence of the stone is reminiscent of the way Gandalf the Grey tricked some trolls into arguing amongst themselves until sunrise, whence they all turned to stone, in "The Hobbit".


Paul T


I`m curious to hear your thoughts on one of the suggestions above:

How do you feel about mechanics that abstract decision-making on the part of the characters retroactively?

So, for instance, you might say, "let's skip over the interrogation bit, ok, guys?" But if it becomes important much later, you can make a roll based on the character's Interrogation skills or perceptiveness or something, "Well, did you figure out to ask him about the Moon Rock, or were you not smart enough to?" Likewise, various kinds of expertise-related skills might cover for various situations. It could be open to interpretation, too, if that works for your group--like: "John's character is always thinking about covering his back. Let's say he has a very high chance of having kept his eyes on the back door during the banquet in the earlier scene."

C. Edwards

As far as fictional integrity goes, I think it's important to remember that during play we're already skipping over great swathes of activity that we just don't find playworthy or that we don't consider important enough to focus any play time upon. We're playing characters in a fiction, so we have that luxury. If that were not the case we'd be sitting through every bowel movement, every uneventful minute between point A and point B, and so on.

Plus we're dealing with a limited commodity; play time. So we usually skim through even more things that may or may not end up being important or relevant later on down the line. This creates gaps in the fiction that are usually of no consequence. But every so often those gaps are of importance. Short of playing through every minute of a character's life or making players give a vast amount of instruction and detail for the moments we do play out to account for future contingencies, it's almost a necessity to have some way of accounting for what the characters did during those gaps that aren't covered during play.

Most of the time the handling of such situations is completely informal. "You did what back at the stone? Okay, cool." But it doesn't sound like that is really going to sit well for you, Dave. I'm not sure there is an option that is going to fit well with your vision of fictional integrity.

Many games handle physical resources, such as money and food, with a general value or rating. Perhaps such a thing would work to measure the quantity or quality of the characters "gap time"  in Delve? Just a thought. It sounds like Paul is proposing the same sort of idea, but with existing character traits.


David Berg

Hi guys,

I want to reiterate here that the issue I need help with relates to meaningful decisions.  Delve skips over and/or summarizes crap that no one cares about just fine.

Also, lest the term "fictional integrity" lug any baggage along with it, let's just say that making decisions in character, using your character's knowledge, is important in Delve.  If your character can't wait till problems arise, then time travel back to prior decision points to make optimal decisions in anticipation of said problems, well, then, as a player, your retroactivity options are limited.

So, on to the latest round of ideas:

Best practices: enabling and making decisions

I agree completely with Jeff's assessment of the situation.  I am comfortable calling the rock an instance of Bad GMing.  The suggestion about the dragon's trap is right on.  So, how should Delve GMs avoid such situations?  I need some rules of thumb to go along with "don't beg many decisions with no time pressure".  Here's one stab:

the more options for a resource, the more important that it be spendable on demand

Portability, secret locations, and Wish magic all strike me as specific implementations of this principle.

To go with this effort on the GM's part, it might be useful to give the players an Adventuring Company's Best Practices guide, with wisdom like, "If you have one thing you're excited about doing, don't worry about probing other opportunities instead.  If time and choices not made close some opportunities, fear not; others will always arise."

When is a decision too meaningful to skip, and when isn't it?

Paul and Josh are probing how to establish stuff when the players are comfortable skipping over decisions they could have made.  Which isn't going to answer how to proceed when they're not comfortable, but hey, maybe we can find what determines comfort/discomfort and figure out how to expand the comfort zone.

Rendering my quandary inapplicable might be more doable than "solving" it, so I'll do my best.

There was one session of Delve that took place under odd circumstances.  John had to miss the session, but we knew he'd be back for the next one.  So, I teleported Dan and Merlin's characters off somewhere and gave them a mini adventure.  Dan and Merlin knew they had 4 hours to do this adventure and get teleported back home to reunite with John's character.  So, we did a fair amount of play in Summary mode.

From my journal:
QuoteRoll 1 Situation:
Having been given directions to the town of Theravia, Dan and Merlin headed there in search of a way to return home.

Roll 1 Objectives:
They told me, "We want to find this Nikolai Ibanescu guy and maybe learn a bit about him. Failing that, we want to talk to the prophet child. We also want to ask everyone we see whether they recognize the 6-circle design from the door that sent us here."

Roll 1:
I picked up Dan's d30, said to myself, "High is good; 20-30 finds Nikolai, 10-20 finds prophet, 0-9 finds nothing," and rolled a 24. Then I made a second roll for the unrelated goal of glyph info, just to make the players think there was some chance they might get it. There wasn't.

Roll 1 Narration:
I told them, "You learn Nikolai has recently returned from being away. He is not well-liked, and some believe him to be in league with the evil on the island. You get directions to his cabin. That's about it. No luck with the 6 circles." I then felt it would be proper to give them some sense of the color of their investigative experience, so I told them about swampy terrain, scarce fields, meager dwellings, and skinny, depressed villagers. A lot of my description was similar to what I would tell them in chunks as they approached and then entered the village.

This worked well for "keeping things moving" purposes, and the sacrifice in experience didn't matter because the PCs left the area. However, we all agreed that more color would have been needed if they'd stuck around.

The rolls I made were not character-specific.  In Delve, you play your own brainpower, so there are no ratings on that.  "How good are Dan and Merlin at wrangling info out of folks?" is something that vaguely informed my decisions about the odds.

I don't want players to ever have to choose between immersion and effectiveness, so any Persuasion stat you had would need to be clearly less effective than roleplaying some persuasion.

So, given that, my instinct is that "mostly random, but colored by past behavior" is best for die rolls.  I think Josh is on my wavelength about this with his luck/foresight idea.

Josh's next idea, to establish outcome limits via "best plausible outcome players can envision" and "worst plausible outcome GM can envision" is interesting!  During the communication of those stakes, the past becomes a bit more real.  The players' concept is likely to be horribly biased -- it's hard to un-know what you know now, even with a good-faith effort -- but perhaps a spectrum of results will render this not an issue.  Like, if the players' and GM's outcomes are both pretty rare, and somewhere in-between is more likely.

In-betweens would be a bitch to moderate impartially, though.  And negotiation ("Alright, I'll give you covering the back door, but not overturning the tables."  "No!  We'd rather have the tables!") would render the results hideously contrived.

I think what this means is that, rather than resolve multiple questions with one roll, it's better to roll for each question.  The process takes a little longer, but I don't think that ruins anything.  As a matter of fact, this is basically what I was doing in the session with Dan and Merlin.

The next roll:

QuoteRoll 2 Situation:
Having convinced Nikolai to come with them to the place they teleported into, the PCs wanted to pump him for info on the walk.

Roll 2 Objectives:
They wanted to know about his plans to fight Theravia's enemies, his affinity for dark powers, his knowledge of the teleporter, his familiarity with the Brotherhood, anything about Dakmour's creepy Nobility, and where he got his weird clothes. They also wanted to convince him that it was not okay for him to kill the guy they were about to meet.

I figured Nikolai would have objectives of his own, wanting to know about all the PCs' magical experiences, particularly those that led up to the teleportation. I asked the PCs what they'd willingly share, and they volunteered a long enough list that I judged it'd occupy the whole trip (they wisely specified a few things they wouldn't share too). If they'd been less generous, I would have had to pose some, "If you don't talk, neither does he," scenarios and maybe roll dice.

Roll 2:
I decided that Nikolai would happily talk honestly about the Nobility, his clothes, and his ignorance of the Brotherhood, so I didn't roll for those. I rolled a d10 each for the remaining 4 topics of interest, deciding that 10 was maximum candor and 1 was refusal. I didn't think about lying; I probably should have.

I rolled high for dark powers, low for teleporter, and middling for enemy plans and "don't kill" compliance.

Roll 2 Narration:
I just fed them info. Very little color beyond, "He brags about X," and, "He won't tell you Y because it's his bargaining chip." We were on the clock at that point, as I needed to end the session and get back to work.

So, to the point: Why were Dan and Merlin comfortable giving up control over the many decisions they'd normally make in play, instead letting me roll randomly?

A lot of it had to do with the limited investment of an isolated one-session adventure.  Beyond that, I think:

1) There was a certain enthusiasm about getting to the next big scene quickly.  "Will Nikolai try to kill the informant when he realizes this man's been holding Nikolai's child?  Let's find out now!"

2) There wasn't any danger.  The consequences of failure were investigative dead-ends.  And the consequences of those dead-ends were never discussed.  I knew what the consequences were, and maybe Dan and Merlin did too from having played with me:
a) "Now you need to think of another plan," and/or
b) "Now you need to pay for info."  I think I can safely say that payment in the form of knowledge or small amounts of money would have been acceptable.  Paying with weapons, armor, or magic items would have been unacceptable, and prompted them to try a different plan.

Note that this still requires a plan.  And "covering bases during planning" is a frequent cause of session slow-down.

The rock-shaping equivalent would be "GM rolls randomly to see what you made."  But the cost may well be that awesome magic axe that you later wish you had.  In Delve, that's a pretty high cost.

Given the opportunity to just come up with "awesome axe!" and have it so, I'm not sure how often the players will risk "let's do it later, retroactively, with a die roll".  If the answer is "only when they're really bored" then that's good enough.  If it's "not even when they're really bored" then the problem remains.  Perhaps I'll playtest the outcome-limits-defining retroactive resolution and find out.

I already know that Dan will be more willing to use the retroactive system than John will.  Again, arbitrating this is an issue of "compromise" versus "most bored wins".  How disengaged should Dan let himself get before he says "enough is enough"?  I feel like I should give him some guideline on this.


Because goals need establishing before success in them can be resolved, I wonder if it might be necessary to do the following:

When players declare, "We'll skip over this part and establish later what we did," they write down a quick list of general priorities.  Not specifics like "hide behind tables", but broad strokes like "best position".  The group can then discuss whether these priorities are mutually exclusive, or in other ways not kosher.

Then, when resolution time comes, there'll be less argument about "What's the likelihood you guys would have taken cover while doing all that other stuff?" and fewer questions of "Is it possible to do X, Y and Z?"

here's my blog, discussing Delve, my game in development


Have you considered some kind of narrative override resource?  What I mean is, is there a way for players to push a skipped result after the fact.

Here's my thinking, the players want to do something else but they feel they need to resolve the current situation first.  It sounds like what you might want in that situation is way to put that scene on hold and get back to it later.  The problem there is that if the scene sets the direction of the later scenes skipping it can make starting the next scene harder.

What I'm suggesting is giving the player a way to establish a few key details and end the scene.  They're basically saying "Let's get back to this later.  Right now I want to go this way."  The idea is to give the players a limited resource that lets them create their own leads for getting out of these sticking points.

David Berg

Hey there, I'm not sure how that differs from some of the earlier suggestions in this thread.  Care to elaborate?

A good solution to my quandary should be used whenever it can be used; no need to make it a limited resource.

As for establishing facts, remember, players can only do that via character actions!  I dunno if that shoots down your idea or not, because I'm not clear on what concrete applications you were envisioning.
here's my blog, discussing Delve, my game in development


The idea was to let the player's push for a desired follow up scene without needing to resolve the problem scene first.  This may very well involve establishing fact, even it it's just "we found a lead to this last scene".  You want that as a limited ability so a player doesn't just skip with impunity.  Another way of handling that would be with a unanimous vote.

It looks like your setting up restrictions that make this a very hard problem to solve.  Let me see if I've got this right.
1) For certain situations, the next scene will depend on the outcome of the current scene.
2) In some of those situations, the current scene is not something the players want to engage in at the moment.
3) The players have no way of getting an outcome without going through the steps in character.
Is that about right?  If so I have no idea how you intend to fix that without either decoupling the linkage in point 1 or adding some kind of shortcuts to point 3.

David Berg

Yeah, sorry, although your idea sounds like it could be cool in another context, I don't think it's compatible with some of the foundations of Delve I've described.
here's my blog, discussing Delve, my game in development