Complex actions...

Considering my new dice-pool system a bit more, and I realise that again, for the sake of expediency, the results of complex actions can be determined by a single roll of multiple dice.

For example, a player wishes to climb a cliff.  The cliff is craggy and covered in vegetation making it reasonably easy to climb.  But it is 200 feet high.  So climbing will take some time.

Because of the ease of climb, the player only needs to roll a four or more to succeed.  They have four ranks in climb, meaning they get to roll four dice in their pool.  But as the cliff is 200 feet high, the player needs to get at least three fours.

In this situation the game is now making use of the possibility of multiple successes to determine a complex action.  This robust mechanic also provides the GM with a lot of information from which to interpret success or failure.  Imagine the player rolls only two fours, but rolled a tweak die of 19.  I'd interpret that as the PC made it up the first two thirds of the cliff (two of three successes obtained) and then slipped and fell.  However, the tweak roll of 19 suggests they manage to grab hold of a tangled root on the way down.  They strain their shoulder a little (as this should be incorporated into further actions involving that shoulder - i.e. combat, or climbing or swimming will now be slightly harder for them - they need a five rather than a four).

I'm liking this new system.  =D

But in the example above, what would an epic fail on the tweak die conclude?  Well, say the player made all of the rolls needed to climb the cliff, but rolled a tweak of 1.  The tweak cannot indicate a fall, as the player succeeded in their rolls.  Remember, the dice pool indicates success or failure.  And the tweak die tweaks those results.  In this situation, I'd look at the bigger picture.  For example, lets says that the player was climbing the cliff in order to catch someone.  You could thus say that despite them climbing the cliff, they crest the top just in time to see their quarry escape by horse.  Or be rescued by masked riders.  Or indeed, you could have them merely waiting at the top of the cliff to strike the player when they are at their most prone.  Or another creature, incidental, is waiting for them.  So despite having made the climb, the epic fail on the tweak die illustrates that something bad has happened, that the bigger goal is now in jeopardy.

Dire Peril

I'm not sure if I like this idea or not.  Too early to tell, but the fact that I'm on the fence suggests I ought to play-test it.  I just came across this notion whilst watching a YouTube video of John Wick being interviewed.  I was pleased to hear John talk about combat in a similar way to how I view it (watch from 5:10 onwards to avoid the guff).  But he also brings up this concept of Dire Peril.

And it seems to tie-in with what I was discussing in my last post about occassionally overtly rolling dice in front of players to determine the outcome of the combat.  Wicky talks about having a rule where his players cannot die, unless he begins a scene by saying: Dire Peril.  The players then know that the safety is off and they can die in this scene.  I kind of like the idea, as I can see how it would make players cautious and nervous - and thus create tension.  But I'm unhappy about the 'at all other times the players are safe' vibe.

As said, I'll have to play-test this and see how it works.  I'd prefer to use Dire Peril on the fly, perhaps saying it as I reach for my dice and roll them overtly before the players.  And of course virtually any action can involve Dire Peril, not just combat.  Climbing a cliff can be hazardous, after all.

Dire Peril!

It has a nice ring to it.  Although it somehow conjures images of schlock-horror 50s B-movie sci-fi flicks.  There is definitely something to it, though.

I shall report back.

Narrative over mechanics - Revenge of the Number

Players that have had some involvement in my games will probably tell you that I am forever striving for a decent combat system.  A combat system which does not slow-up the game, but which is yet authentic, realistic and dangerous - whilst also being flexible.

Can this be achieved?

My current take on it - a take that has been developing over the last ten years or so, is that combat should be intrepetive, like all game rules, IMHO.  Looking up tables, adding and subtracting a variety of bonuses etc. etc. all slows down the game into a number-crunching nightmare that steals the passion and excitement from the scene.  Yet I still believe it is important to have a final, random, dangerous element - so anyone entering into combat cannot be sure they won't die, or lose a limb or whatever.  Part of the enjoyment of roleplaying is not quite knowing where the story is going, and this is almost as true for the GM as it is for the players.  For the players, they are exploring a world, a narrative, and they really don't (or shouldn't) know what's in store for them.  But for the GM, the players, by their actions which of course the GM does not dictate, also modify the story.  But the dice do this also.  Through the success or failure of the players' actions, and to a certain extent, the actions of NPCs too - the story can shift and change dramatically.

As a GM I tend to use dice rolls in a variety of ways.  Firstly, in order to propel the story forward, the majority of incidental rolls (players finding stuff, overhearing stuff etc.) if its required to happen to drive the story forward and keep the pace going I employ the 'Gygax Rule' and roll the dice just for the sound of it.  Effectively ignoring the result.  Sometimes I need a little inspiration myself though, and if its not critical to drive the narrative forward, and I determine that a little randomness can provide for an interesting divergence or outcome, I will roll the dice and allow that randomness to tweak the direction things are going in.  How an NPC reacts to the players' behaviour for example (if its not critical to the story).  Barmaids fancying the dashing paladin, or the quirky mage, that kind of thing.  And then, at some critical tense scenes, I deliberately go against the grain, and roll the dice right in front of the players so there is no hiding it.  Because contrary to what John Wick says, drama can be derived from randomness, and I have witnessed many a heroic death brought about by the dice.

So as I strive to make combat effective, thrilling and yet dangerous -this is how I achieve it:

For the sake of expediency, combat rolls are heavily interpreted, I don't use table or charts and I pad out the dice results with description.  As I believe I mentioned in an earlier post, I use a 'tweak die' to generically influence how successful something was.  So the basic rolls determine success or failure, and the tweak dice determine the extent of the success or failure.  As combat is a fluid thing, one decent success, or epic failure can lead you into an advantageous position, or a leave you prone to further mishaps.  I use the dice rolls to embellish and describe the combat events that occur.  And this necessarily leads on to dramatic fights.  A series of fails can eventually lead a player to the point where the next blow from the enemy could be devastating.  And this is where the drama and story-telling come in.  After a series of four failed rolls, or poor tweaks the player is on his back, having been disarmed and his attacker looms over him, sword raised to finish him off - and it is that point that I will roll directly in front of the other players - but it is also at that point that another player can intervene, turn, and hurl their own axe at the looming attacker.

Such descriptions cannot be achieved when counting hit points.  I don't think.  Hit points take into account all that drama but distil it down into a mere integer.  What I'm suggesting is removing the numbers and interpreting the results of the dice, descriptively.

Play-testing awaits.

Messiah scenario idea...

I've been toying with this idea for a number of years.  When I say toying, I mean considering vaguely from time to time, but not doing anything about it - including writing notes, so it is only a vague-formed notion.

The idea is that the players are tasked with protecting a young girl, maybe 7-8 years old, who is allegedly the earthly form of a god, or perhaps the offspring of a god, on her journey to a distant place where she may fulfil her birthright.  The girl's protectors, in order to try and keep the girl safe have hired, with the promise of great reward and arcane treasures, a band of experienced adventurers from a distant land.  The distant land bit is important, because the hirelings have to be of a sufficiently different culture as to have little understanding of the religion the men adhere to.  Nor any prophecies concerning the girl.  The reason for this, is that the girl is actually evil.  The men appear to be good.  Very gracious and in all manners and appearances benevolent. And the girl too, appears good.  Blonde hair, flowing white dress, beatific smile - the classic hallmarks of benevolence and corruptability.

The game starts with the PCs waiting on the roadside, on an out-of-the-way road on the border between two lands.  The girl's escort are to bring her here covertly, under cover of darkness, so she can be whisked away into foreign realms and kept safe, until the location of the distant place she is to be brought to has been revealed - scholars are investigating that element concurrently.  The escorts suggest to the PCs that they keep moving, as far away as possible, because the girl has many dark and eldritch enemies (actually good and wholesome enemies).

They provide the party with an artefact which will enable communication at a distance.  But the artefact should only be used occassionally.  It also requires a roaring fire.

I use a skill called 'arcana' in my fantasy game world, which is essentially lore regarding enchanted artefacts, because every magick item has a history, was created by someone, used by people and potentially in various important events.  The magick communication device (I'm considering a statuette one has to place in embers, and which then speaks in a deep booming voice - not the voice of the communicator on 'the other end' - similar statue), if a player wishes to, and rolls a high enough score against their arcana skill, they know that such a device was created by X for Y but lost many years ago, suppossedly fallen into the hands of darkness.  This is a clue to the fact that the people who have hired them may not be precisely whom they appear to be, and not necessarily benign.

To create tension, whilst this initial exchange is taking place, two dozen riders come storming in to take (or kill) the girl.  The riders must be in sufficient number to cause the party concern as to whether they should stay and fight.  To build this tension up, have a couple of the escorts fall early on in the conflict, and the leader instruct the party to go.  Thus the game kicks off early on with a chase scene.  And as the PCs look back they see the escorts battle with some of the riders, but most begin to pursue the girl (and the PCs) indicating what their true goal is.  The players are experienced though, and should be able to dispatch and evade - though not easily - their pursuers.

After putting some distance between them and their pursuers, the party will eventually make camp against a developing storm.  Storms in my fantasy games are also an indication that people are using magick - the more powerful the magick, the more violent the storm.  This knowledge is common to any spellcasters and rumours/omens to non-spellcasters, but such information should be relayed to players in their background notes, and thus not drawn attention to overtly within the game.  This makes it a slightly more subtle clue than if one was to explicitly state it at the time.

During the night a winged beast attacks the camp.  It has been summoned by the Order the riders belong to, and its task is to slay the child.  The PCs must obviously protect her. Describe the roaring winds and the beating wings and the foul breath of the beast.

The game continues in such a way with the various hostile encounters in various forms and guises, all with the objective of killing the child.  These are intended to be the 'action scenes' used to surprise and alarm the players.  But each, must have some sort of clue as to what they are really dealing with.

Clues as to the true nature of the child must also come in the form of incidental scenes.  A mad beggar woman on the outskirts of the town (with the gift of trueseeing) recoils in horror when she sees the child.  A very close likeness of the child must be present in an ancient text in a church they stop and shelter at along the way.  The church itself the child says is 'creepy'. As I consider this scenario idea more, I will concentrate initially on two things:

1)Scenes of Action - how are the forces of light realistically tracking the child, and sending forces against her?
2)Clues - what incidental occurrences can be employed to hint at the true nature of the child?

In public areas, out of sight of the PCs, the child will also attempt to illustrate her tendencies.  Essentially, bad stuff happens around her.  Dogs bark, cats hiss and flee.  Ravens alight in her vicinity.  But such reactions must be handled with subtlety.

If the players realise who she is, they may attempt to kill her, but killing the physical embodiment of a god is no easy task, and in fact takes either special weapons, or special places, or special rituals.  I feel there should be multiple options to kill her, but all take some planning and involvement, and will be dogged by her supporters on the one hand - and also by naive do-gooders on the other.  What I mean by this is as soon as the players realise the child is evil and attempt to do her in, some passers-by will only see grown men doing something bad to a child and attempt to intervene.

If the players never cotton on to the fact that the child is evil.  Then they will eventually be tasked to take her to some ancient place, at a specific time (conjunction of the planets etc.) where they are to perform a ritual of 'raising' or some such thing, where she 'matures' and takes on all sorts of additional powers.  At this point the players are asked by her if they wish to bow down and worship her.  Those that do gain benefits but become tainted and evil.  Those that don't have to flee or be slain.

Thats the rough idea for now.  Any suggestions are welcome!

Bag-space in MMOs...

Right, okay, fine.  So this is one of my occassional divergent posts into the realms of videogame RPGs, but I've recently being playing Lord of the Rings Online (basically WoW with better graphics) in irregular doses.  And again they've screwed up how bag-space is handled (as does WoW, as does every other videogame RPG that I can recall off the top of my head).

And it is quite simple to get right:

Firstly, from a usability point of view, the player needs only ONE window to view the items they carry.  We should not be forced to remember which of possibly several bags we have put something in. And we should not have to open all of them to look for it.  We should be able to open one window, and SORT by a variety of different features: alphabetically, date/time added, item type (weapon, armour, quest, consumable etc.) and so forth.  And the window should include a SEARCH box, so if worse comes to worse and you can't see what you're looking for at a glance, you can search for it.  Basically, if you're going to make an RPG run on a computer, learn from what good applications and websites have evolved into and utilise those tried and tested features.  One window for all your pack/bag space added together, in a list.

This does not, and should not, disrupt features of game mechanics involving offering the player a variety of upgradeable bag options.  Imagine in a fantasy setting.  As a designer you may wish the player to start out with a small pack (say 12 'units') and one small pouch (say 4 units).  You allow a player to carry upto 4 pouches, two on each hip; one pack and pockets on some leg clothing, and pockets on some chest clothing.

As they progress through the game they have the option of acquiring and upgrading these various items which add bag-space.  All bag-space is, in effect, a single integer relating to the amount of equipment a character can carry.  A unit may be considered a combination of size and weight.  So big, light items may take up just as many units as small heavy ones.

As mentioned previously, all bag-space is cumulative and accessible via one single window, which a player can access with a click of a button and can sort through and arrange in different views.

For the sake of facilitating a player in combining some potentially disparate items in their pack.  You may wish to provide a secondary 'manipulate' or 'handle' window.  Allowing a player to drag items from their bag-space into a window, perform a secondary search, drag something else into the 'manipulate' window and so on, where they can be combined etc.

If you want to be clever, once one item has been dragged into the manipulate window, you may wish to automatically search on the contents of their bag-space to find items they may wish to combine with the first item.  Or even cleverer - automatically search only for items which the character knows (via their skills) can be combined with the first - so you're not revealing secret recipes in the game, unless the character has the relevant skill/training/knowledge whatever.

Secondly, we should not have to play 'pack tetris' by having to shuffle items around so they fit in the space provided (not a problem for WoW or LOTRO, but is for others games, tho their names for the moment escape me - Fallen Earth?). Any item should just be a line in the bag-space window.  You can still tag it with a pretty icon as a visual reminder of what it is, if you feel the need.

The method I describe here, I humbly put to you, is FAR SUPERIOR to anything in the games mentioned above, and AFAIK, in any computer-based RPG I have played.  So when are RPG developers (Blizzard, Turbine, whomever) going to cotton-on and start providing a pack system which is actually effective and presents decent usability?

Also, and something I have yet to overtly witness in such games are the inclusion of enchanted packs etc, which dramatically increase bag-space.  Such RPGs are typically unrealistic in that they allow players to walk around with a dozen broadswords, a dozen shields and three full suits of plate mail in their packs.  I'm suggesting being a little more realistic to begin with, and then provide increasingly larger enchanted packs and pouches that effectively at ultimate levels do away with the need to sort out your bag-space ever again (within reason).  For the sake of databases becoming unwieldy and unmanageable I expect a finite limit should be included somewhere.

The most important thing tho, is one window, that is both sortable and searchable.  Enabling players to find what they want in their bag-space, quickly and efficiently, allowing them to play the game more, and fiddle with the interface less.

peas out.

The Frankenstein game...

Just quickly while I think of it.  Another interesting game I'm looking forward to running is what I have jotted down in my notebook as 'The Frankenstein game'.

I came up with the idea after reading Gustav Meyrink's classic: The Golem.  Having been written in 1914, and its subject matter, lends itself very nicely to a vaguely off-piste Cthulhu offering.

I haven't decided yet whether to set the game in and around Lovecraft's fictional Miskatonic University. Or make use of some old maps of London and have them at Imperial College or perhaps Guy's Hospital as that way I could set some of the game around spooky old post-Ripper Whitechapel, an area I know well as I used to live one street away from the site of the Ripper's second murder.

The basic idea is this: The PCs are all medical students.   One Jewish PC discovers a sheaf of papers when helping his local community to tidy up an old synagogue damaged in a vicious storm.  The papers, incased in the stone altar that was cracked in two by a falling tree reveal the Jewish legend of the Golem, but a variant form of this myth which talks about constructing a golem from flesh, not clay.

And so the game sees the players locate and hire suitable premises for their experiments, and then they begin grave robbing in order to obtain body parts to assemble into the shape of a monstrous man.  Requiring fresher parts, they begin raiding the hospital morgue and eventually turn to murder in the smoggy gas-lit streets at midnight.

This is admittedly the bare bones of an idea for a game, but I think it has legs. :-)
Especially if you allow the players to create their golems with multiple limbs, beast claws from london zoo and generally go more Reanimator/Cthulhu with it.

A game in which the players are, arguably, the bad guys, can commit grave-robbing, body-snatching, and murder and create a monster they can control(?) sounds like a lot of fun to me!

Player training...

O and the arrogance continues!  So something else than transpired in my recent dialogue with other roleplayers over at RPG.net was the backgrounds of an intended bunch of PCs in a cyberpunk game - which you'll have a vague idea about if you read my last post.

In this scenario the players are all supposed to be police officers.  Nothing wrong with that.  Quite fitting for a cyberpunk game.  But it got me to thinking.  As mentioned in my last post, making the PCs police provides a number of advantages in terms of controlling the game, on the part of the GM.  But in order for it to work, seem realistic and not frustrate players, players have to know how to act like police.  And this is where it's important to work with your players, pre-game and prepare them for their roles.

When I run a game, I generally spend a considerable amount of time working with the players in developing characters and backgrounds that are engaging, intriguing, and as rich and detailed as I hope the game enviroment is.  Again, much of this comes down to that all important essential - preparation.  As a GM your job does not stop at crafting the adventure, or rather, taking an active role in helping each player craft their character is part of crafting the adventure.

And I don't just mean, provide the players with interesting backgrounds, so they have an idea of who they are, where they come from, and what some of their drivers, motives and aspirations are.  Of course you have to do all that.  But depending on what they are, you may have to offer some guidance there too.  And this might be something of an education for both of you (player and GM).

Let's take the above example of PCs as police officers.  Now police officers have training.  And just as you'd expect in any decent police movie, a degree of realism - and this is typically generated by the scriptwriters and director having some knowledge of police procedures etc. to make the narrative authentic. It will also require the actors to do some research to get into character.  The lessons that can be learnt here is that if you're running a game and want your players to be police (or any other specific role) do enough research on that role so that it becomes authentic.  And also, where necessary educate your players.

A couple of hours researching crime-scene procedure online, and maybe watching a couple of classic cop films and taking notes will mean that not only do your NPC police officers act with coherence, you can also remind the players (prior to the game) that they should be doing things like: dusting for prints; cross-referencing images obtained from security footage with the FBI database; looking for bullet trajectories to ascertain where the shooter was; sealing off a crime scene to preserve evidence; so and so forth.

Preparation is nine tenths of the battle.  And it makes your game more interesting, more gripping... more enjoyable.

Collaborative post on RPG.net

I've recently begun posting on RPG.net, collaborating with a few others on the generic design of a cyberpunk scenario.  Some interesting points have been made, and if you can drag yourself away from fraternizing with ladies whose surnames are all .jpg, I recommend checking RPG.net in general, and this thread specifically.

I made a post earlier today which loosely discussed the importance for a GM to have a reason for things happening in their scenarios, and having subsequently considered it some more, realised that what I'm actually talking about is preparation.  Preparation for a games master is everything.  It means your players can go off piste without too much worry, because you're game environment is so detailed, everything seems potentially a part of the narrative.  Preparation gives depths to your stories, your environments, your histories and your characters.  Preparation means you can have four interwoven storylines emerging rather than one - so as to avoid the need to overtly railroad your players.

In the extract below (from RPG.net) I'm talking about the need for having a reason why a particular city (the setting for a proposed cyberpunk genre game) is experiencing high degrees of civil unrest.  I appreciate I'm probably commiting some narcissistic cardinal sin by quoting myself, but I'm really only doing so in order to segue deftly onto a whole piece on preparation.  And besides, before we get blogged down in a meta-argument about the pseudo-intellectual self-mastubatory nature of blogging in general, I'm just going to lay on like MacDuff:

"Firstly, for the sake of atmosphere / intrigue, cohesiveness and making things seem as real as possible, there should be a reason why this civil unrest is occuring.

And I would look to the real world for inspiration. Governments (especially the US) create civil unrest all the time - predominantly in third world nations. They call it 'destabilization' and it gives them new excuses to keep the war machine going, and reasons for passing serious laws like the (otherwise unlawful) detainment-without-proof of suspected terrorists. In a cyberpunk setting, where often the fractures between populaces are not on an inter-country basis, but an intra-country basis, its easy to transplant the exact same agenda. Especially in a genre which typically evinces a shift from government dominance to corporate dominance. The underlying reason behind the civil unrest could be because the local government wants to destabilize the city, so it can pass more stringent laws and put in place a draconian military rule; or a mega-corp wants to destabilize to illustrate just how weak the government is so it can take over the city and turn it into 'Mega-Corp Metro' (not an actual name suggestion ). Or indeed the military could be doing it, so they themselves can become the ruling power and start to dominate everything.

I have an issue in general with 3-4 stories [in a single scenario], and that is: in a single scenario, the GM should strive (IMHO) to tell one story (with the help of the players), else they are in danger of not having any cohesive plot and will be wandering around virtually aimlessly between one potential storyline and another. This is okay in a campaign where you can ultimately tie multiple stories together, or have the party wander off and complete a side mission and then pull back to the main thread. But in a single scenario, its best to keep them [the players] heading in a single direction so you have more control in meting out the drama, suspense, tension, denouement. Furthermore it is more in keeping with the way a police department works [the PCs are all police]: you guys investigate the trouble at the old mill; you guys investigate the explosion that derailed the metro train; and you guys investigate the murder at the steelworks. You know? Police teams are given one assignment at a time. The other GREAT narrative device provided by having the players as police, is that it makes it very easy for the GM to keep them on track. Any time they are stuck, advise them to check back in with their superiors who can advise them on how to proceed. When doing this, ALWAYS give the players at least two options (with each potentially furthering the story), thus keeping them heading in the right direction, though on different branches of resolution. Utilising such a technique is a godsend to the GM as it is realistic and thus nicely sidesteps being seen as railroady.

But what you can do here, and again IMHO should do, is present a backdrop rich enough that the players understand there are more things going on. Have them know that whilst they're investigating the metro train disaster, something at the old mill that their colleagues are investigating, links up. The same man spotted in both scenes. This is a great way for them to compare notes with some NPCs and have the NPCs realistically impart some useful information to the PCs - which can drive the plot forward, and which they may have missed. This is what I call a 'narrative safety loop'.

Also, present characters in this single scenario which don't really amount to anything, but may crop up again in a later game. This adds great depth to your game environment. Someone whom appears to be a nobody in this first game, is actually a significant player in scenario number two."

Again, apologies for the shameless self-quoting. 

Do check out RPG.net, some great stuff there.

I will be going into more detail about some techniques I use to generate narrative and detail when planning a game.  Perhaps someone, somewhere will find it useful.  But I'm not holding my breath.

peas out

Intervention

In my last post I loosely outlined some new mechanics I have been considering for a new game system.  In writing it I started pondering what I feel is another interesting concept.

I talked briefly about GM intervention, and what a no-no it is, IMHO. However, nothing kills a game faster than killing off all the PCs.  As illustrated in: How not to handle death.

So what can a GM (or game mechanics) do when it looks like a PC is going to die?  Well, as I have said before, sometimes you just got to go with it, and a PC death leads to a memorable scene in a game.  But I have been considering an alternative mechanic to sidestep some PC death situations in a viable manner.

I like bonus dice and I like bonus points.  So how about a very rare type of bonus. An intervention point.  This can be considered an intervention by the gods in most fantasy based games.  Intervention by some higher intelligences in some more futuristic / realistic based games.  Or just sheer luck.  Bestow them on players very rarely.  Never more than one at a time.  And perhaps bestow them on the party rather than an individual player.  That way the party has to decide when to use them.  Also, considering John Wick's 'let the players' decide the outcome' mechanic.  Which I personally feel is fraught with danger, but still kinda like the idea of if meted out in ultra-small doses.  If the party has earned an intervention point, a prophecy point, an alternative universe point, whatever you want to call it, if a situation results in the death of one or more party members, then the party can decide to spend that point and they decide how the deaths are avoided.  With GM approval of course.

I plan to test this.

A new game system in the making...

So I've been thinking of a totally new game system.  Yes, that means ditching my beloved, overhauled, Homebrew version of rolemaster and going with something new.  Something rules-light (actually my Homebrew system is very rules-light), and as story-focused as I have a penchant to be.  It'll be interpretive, because that's the direction I've always been heading (and largely encompassed).  And its going to feature:

d10s! - cos I have a bag of 'em.  Plus they just smack of RPG better than a d6.

It's going to be a dice pool game.  Because I think picking up a fistful of dice, and rolling them all is fun.  Plus that gives a lot of freedom in terms of throwing in extra dice for particular reasons... luck points, fate points, bonus dice, blah blah blah.  Don't worry, I'll expand on all this in another, more lucid post.

It's going to have bonuses to those dice.  So the player will end up rolling xd10+y... i.e. say you have three ranks in a skill, thats 3d10. A basic success chance is a 6 on any of those d10.  The harder the action - and several elements come into play there - the higher than success number is.  But a player can modify their rolls in a number of ways, and add dice to the pool in a number of ways.  So another example might be: try and get 12 or more on 6d10+3.  i.e. a player need only get a 9 or more (with the +3) to score 12 and succeed in that action.

On top of that, I want to include a d20. For no other reason than a d20 for me (probably because I cut my teeth on D&D, back in the day), signifies roleplaying.  Despite the fact that I have not used a d20 in any of my games for years... decades even.  When that 20 comes up, its a special moment, and one I want to include.

So the d20 will work like this:  You roll a d20 as a modifier to any action.  It signifies critical success... or failure.  It's the 'tweak' die.  The higher or lower the number indicates a tweak to the success or failure of the dice pool.  Furthermore a critical is obtained when a 20 is rolled (or a 1 in the sake of critical failure).  At the GMs discretion these numbers may be modified - i.e. a critical is afforded on a roll of 19 or 20, failure on a 1-3 etc. Regardless of a critical, if the tweak die is higher than twice the number required to succeed, the player earns a bonus point from the action which they may use in a future roll.

If a 20 is obtained and the dice pool has been a success - then the action is deemed to be a critical success. In this case the result is extremely beneficial to the player and they get a bonus die and bonus point (which can be added to a future dice pool), i.e. a 1d10+1.  The bonus die and the bonus point are considered separate entities and do not need to be used together.  These bonuses derived from critical success are referred to as being 'on a roll'.

If the dice pool (xd10+y) roll was a success, but the tweak (d20) roll was a 1 indicating critical failure, then the action still succeeds, but the player is left in an awkward position.

For example, a PC attempts to make a jump between rooftops.  The player has 4 ranks in jump so gets to roll 4d10 by default.  As a standard manouevre a roll of 6 or more on any of the dice would indicate success.  However the distance is quite far, so the GM states a roll of 7 is requried. The GM also determines because they are being chased (+1), are wounded from a previous encounter (+1) and its raining (+1); the action is complicated by a further +3, resulting in a 10 being the success marker.  The player determines that this is an essential action for their character, and from their bonus pool adds two more dice and two more points to the equation.  So they're intending to roll a 6d10+2.  The +2 means the result of any dice is raised by 2. So now the player has 6 chances to roll and 8 or more, with 8+2 = 10 = success.  They roll and get and 8 and a 9 (as well as some lower numbers) and so success has been achieved!  However, the d20 tweak roll they make at the same time comes up 1.  This means that despite them having made the jump, something bad has happened.  To determine how bad, the GM looks at the lowest number they rolled - in this case a 2 - and interprets that as them having taken some damage from a poor landing.  A 3 or 4 may have indicated some non-essential piece of equipment was damaged from a poor landing.  If a 1 was present, then maybe something important was dropped into the street below.

In this way perfect flexibility can be brought to bear on interpeting the outcome.  Such rules-light games are fast to play, not bogged down by multiple table checking, and the mechanics hinge upon the imagination of the GM and the players - which is what makes roleplaying so special.  Such rules-light systems harness what is special about Pen & Paper, and what makes it a unique form of entertainment -  i.e. a collaboration of imaginations.

On with the example: if the outcome of such an action fails - nothing 8 or above is rolled, but the tweak die comes up as a 20, it means the action failed, but something nevertheless beneficial happens.  In this example, the PC does not make the jump, misses the other rooftop and plummets down, but lands in a passing haycart.  Again the GM can take a look at the dice pool rolls for some inspiration or guidance.  Because this was a critical success the GM looks at how high the dice were ( though clearly not enough to warrant success in the desired action).  In this example we can say the player had terrible rolls, nothing above a 3, so the GM decides that the passing cart was actually transporting manure, not hay, and the PC is now rather stinky!

In the case of very poor dice pool rolls resulting in a failure of the attempted action, and a critical failure (1) generated by the tweak die, the PC is in a rather tricky situation:  They didn't make the jump and suffered a critical failure to boot, which could be translated as landing badly in the street below.  The intrepetive scale here, I would suggest, is based on what the 'scene' was trying to deliver.  In our example the PC was being chased across rooftops, so the ultimate failure here would be capture.  So this is essentially what is delivered by a critically failed action.  However, the GM does have a requirement to present an interesting, atmospheric and thrilling story.  So whilst they can suggest the player falls to their death, I would suggest avoiding that outcome in all but the most dramatic moments.  I.e. if they 100s of feet up and fell, it might be difficult to realistically avoid death.  And now we come to a potential cardinal sin of GMing.  Despite the fact that IMHO PC deaths should be rare and handled with care, they MUST be delivered if the situation requires it.  One cannot shy away from these moments.  Because nothing ruins atmosphere and verisimilitude faster than GM intervention.  By that I mean when a GM unrealistically pulls a PC out from a certain death situation.  If players feel they cannot die, it removes risk and tension, which results in a lack of excitement and a lack of atmosphere.

In the above example of a critical failure, the interpretive scale has PC death at one end, and something along the lines of a sprained ankle on the other, with broken bones, unconsciousness and capture somewhere in between.  As a GM, one should strive to deliver the next thrilling scene, which in this case might just be the PC hobbling off down a dark alley, having evaded the pursuers (as they're on the rooftops above) but now the PC is wounded and in a bad neighbourhood.

These, as basic mechanics, I think will be a lot of fun. They are of course intended to work for non-combative actions.