Stalker0's Obsidian Skill Challenge System (Update: Version 1.1) Now with PDF!!

[Dr. Strangemonkey]

Well, I finally got a chance to try it out. Here's a brief breakdown on how I think the attempt went, I don't think I have enough data to provide a direct analysis of how well the system itself fit the game.

1.) I don't think I did a very good presenting or running the system. The notes I had weren't sufficient, and it fit into the encounter oddly. I certainly need to review the materials and make fresh notes.

2.) I think the players were a bit confused about what was being asked of them, but that's fair. This was their first encounter with skill challenges at all, and I wanted it to be a sort of blind encounter.

3.) The feedback I got from players was overwhelmingly positive. They didn't know what they were dealing with, but they liked the vague outlines of what they saw.

4.) It's extremely difficult for me to be disciplined with the Challenge system as of yet. Certainly, in the situation I was in it was way too tempting for me to simply use it as a way of 'fixing the rails' for every other plot point I didn't want to deal with directly.

At first I felt ashamed, but now I'm not entirely certain if that wasn't the point the whole time.

 

[clearstream]

De-trivialising the control strategy

Hi Stalker - I'm working on the mechanics, rather than the numbers, and wondered if you could look at something.

Premise Players being the only source of successes and failures, the best control strategy is trivially realised to be control over participation.

Proposal Challenges require N successes per round, or the round outputs a failure. Successes >N output a success (possibly scaled for quality).

N must be < the number of successes players could generate per round. N might therefore be expressed as a set of numbers scaled against participants. 1=0, 2=1, 3=2, 4=3, etc. At the expected number of participants N can be more freely assigned.

So, my question Does this idea hold up mathematically? That is to say, with players needing N successes in a round to avoid counting a failure, what happens to the numbers?

Finally, in the backdrop I am engineering a number of skill challenge powers that will help players gain successes: the basic power reads 'do this, roll again if you fail your next check'.

-vk

 

[Neubert]

Stalker (and others), I was wondering what your views are on the uses of "Bonus" / "Backlash" skills (see the DMG example of the negotiation with the duke). History was a skill that was opened up by the use of another, whereas Intimidate was an automatic failure (I am writing this from memory as I don't have the book with me).
I like the idea of skills opening up to the players. The "Backlash" skills are easily implemented in making the players unable to succeed in that skill, but this needs to be out-weighed by the bonus skill(s) - or the challenge will be harder of course.
My initial thought is to let the players get the same +2 on the bonus skill as for the primary skill, but it might only be usable a set amount of times or to gain a certain amount of successes?

 

[bardolph]

Hi Stalker - I'm working on the mechanics, rather than the numbers, and wondered if you could look at something.

Premise Players being the only source of successes and failures, the best control strategy is trivially realised to be control over participation.

Proposal Challenges require N successes per round, or the round outputs a failure. Successes >N output a success (possibly scaled for quality).

N must be < the number of successes players could generate per round. N might therefore be expressed as a set of numbers scaled against participants. 1=0, 2=1, 3=2, 4=3, etc. At the expected number of participants N can be more freely assigned.

So, my question Does this idea hold up mathematically? That is to say, with players needing N successes in a round to avoid counting a failure, what happens to the numbers?

Finally, in the backdrop I am engineering a number of skill challenge powers that will help players gain successes: the basic power reads 'do this, roll again if you fail your next check'

I see a few issues with this:

• The RAW skill system is pretty broken no matter how you look at it. Adding more failure chances will only make it worse.
• The Errata'd skill system eliminates rounds. The assumption now is that a single player is making all key rolls, with everyone else usually aiding or opting out. Your mechanic doesn't really fit with this system either: if the skill monkey fails his roll, do you count two failures instead of one?
• Obsidian doesn't count failures. It only counts successes. The proposed mechanic also doesn't fit here.

If you're working on an alternate system that combines the original RAW "rounds" model, but with minimum success requirements per round, my first instinct is to say that this will not work nearly as well as either Obsidian or the errata'd DMG system.

Also note that, even with the errata, the DMG challenge Lost in the Wilderness still does not work, since it was written to fit the old "rounds" model. If I were running this challenge, I would simply require every character to make an Endurance check or lose a healing surge every time the leader rolled a failure on his Nature check.

 

[clearstream]

Right, I see you have followed a similar thought process as I did. The problem with pending the ultimate decision on the final segment is the lack of interim tension points. Also, it may sometimes happen that a loss is inevitable by the end of segment two.

I therefore feel it is better to break this down segment by segment. Look for N successes per segement, not overall, and output a success or failure. I like your quality of outcome stuff, but that is easily replicated by considering successes over N within a segment. Ultimately you might need to abandon having just three segments, but I feel that will prove no loss.

-vk

 

[CodexofRome]

I used the Obsidian System in yesterday's game to great effect.

The party consists of a Dwarf fighter, an elf ranger, a tiefling rogue, a half-elf warlock, a human wizard and a human paladin. They're adventuring in a homebrew world. All of them are new to D&D. Two of them were sitting in with the group for the first time. They are all still very much learning the game.

I built several skill challenges into the adventure, exploring various options for the system.

In the first, the party spotted and chased saboteurs who were making their way way through the city. The players picked up on the concept right away. The fighter and ranger concentrated on their endurance. The wizard used her insight to determine where they were headed. The rogue using her stealth and acrobatics and the warlock his streetwise to get ahead of the saboteurs. The party got their 10 victories, stopping the saboteurs before they could open the city gates to an invading army. The players came up with lots of colorful descriptions of their actions.

In the second challenge, the party tried to get their patron, a priestess of the Raven Queen, to disclose information about the true situation they faced. They had a harder time coming up things to do, and only scored a partial success on this challenge, which focused on Religion and Diplomacy.

The third challenge occurred in the caverns below the city's catacombs. The party came up on a very recent, very blood fight. Numerous dead bodies littered the ground. The "Battlefield Detectives" challenge, focused on insight and perception, went a bit better, as the players came up with some excellent descriptions and, although a close call, got their 10 victories. I have the Dwarf a +2 racial bonus, and when they won, handed him this note to read to the party.

"Here a Dwarf of the Streggi Clan fought many of the Low Men, who are sometimes called Orcs. The Dwarf approached from the direction we did, and was ambushed by the Orcs, who he may have been following. He has some sort of spike in his chest, and it is clearly of infernal and magical origin. Not all who fought here died. The tracks of a someone wearing shoes leads off in the direction we are heading."

As I said, the players picked up on the idea of the system right away. The only real problem was their unfamiliarity with their skills and what they could or should do with them. All three challenges ran well technically, provided lots of dramatic interaction for the players, and gave me a chance to introduce some new dynamics to the game.

Thanks Stalker0 for an excellent addition to the game! Since we're both in Atlanta, you'll have to come watch your system in action sometime.

 

[clearstream]

I see a few issues with this:

• The RAW skill system is pretty broken no matter how you look at it. Adding more failure chances will only make it worse.
• The Errata'd skill system eliminates rounds. The assumption now is that a single player is making all key rolls, with everyone else usually aiding or opting out. Your mechanic doesn't really fit with this system either: if the skill monkey fails his roll, do you count two failures instead of one?
• Obsidian doesn't count failures. It only counts successes. The proposed mechanic also doesn't fit here.

If you're working on an alternate system that combines the original RAW "rounds" model, but with minimum success requirements per round, my first instinct is to say that this will not work nearly as well as either Obsidian or the errata'd DMG system.

Hi Bardolph. Great feedback, thank you

To explain: I'm neither trying to fit with RAW or Obsidian. I realise Obsidian counts only successes, and RAW is good and broke. I hate the one-expert assumption, and I can think of examples where it doesn't produce SOD.

Obsidian is a nice system with a great maths grounding, but mechanically it lacks the tension-from-crunch I desire. See my quick reply re: that.

I'll post up my scratchings in full later today!

-vk

 

[grickherder]

Well, I ran a session that included the skill challenge system presented in this thread.

I ran one of the hardest things to do well mechanically-- an investigation!

The players had returned to town after discovering a piece of evidence that the Abominations of Khyber (underworld in the Eberron setting) had a spy in the fraternal order they were part of. Not all members of the order are necessarily adventurers or whatever. Lots of regular people from all walks of life.

There were 12 main possible suspects to which they initially could narrow it down.

Here's how I handled it:

Each roll, success or failure, would get them a piece of information. If they succeeded, the information was air tight. If they failed, it was probably accurate, but they'd have to assess it's merits on their own. Also, they didn't just "get the information" but learned it from NPCs through dialogue. So they could assess the information based on who it was coming from and all that.

I did not start out with any one of the 12 as the traitor. It could have been any of them. It didn't matter to me as a GM. For each of the 12 I figured out a piece of information that would prove their guilt beyond a shadow of a doubt and another piece of information that exonerated them. They two or three pieces of information that might implicate them but not necessarily conclusively. A few particularly devious ones who also suspected a spy had planted false information to implicate their rivals in the order.

I generally kept the information that completely implicated someone for near the end of the 2nd segment and into the 3rd.

I allowed mental and social skills. For the first segment, I assigned insight and diplomacy as the main skills (the ones that get you a +2). A few physical skills were rolled during the challenge as they were appropriate.

It really works for an investigation. The key I think, as the DM, might be to not have the answer set in advance. Whomever the players go after is likely the guilty party. Go with the flow of the story and give information even when they fail their skill roll. Yes, this means lots of prep work to prepare information for each and every roll.

Just want to send a shout out to N, who lurks here and has also ran an investigation with this system with his group out west. Sign up and post about it! Failing that, email me and I'll post your AP from your email last nigh.

 

[Stalker0]

Stalker (and others), I was wondering what your views are on the uses of "Bonus" / "Backlash" skills (see the DMG example of the negotiation with the duke). History was a skill that was opened up by the use of another, whereas Intimidate was an automatic failure (I am writing this from memory as I don't have the book with me).

I'd have to run some math to check on these ideas for sure, but in general one of the advantages of Obsidian is that each roll is more "self-contained". In the normal system, not getting a success is really bad because it also gives you a failure, and that can prematurely end the challenge. In my system, each roll matters less so you can have a larger variance in any single check without having a huge impact on the challenge as a whole.

So bonus and backlash skills would likely work alright.

 

[Stalker0]

Right, I see you have followed a similar thought process as I did. The problem with pending the ultimate decision on the final segment is the lack of interim tension points. Also, it may sometimes happen that a loss is inevitable by the end of segment two.

I recognized that these could be issue points with some people when first making the system. In the long run I chose a cleaner and simpler system instead of one that addressed these issues, though I recognize they can be sticking points for some.

In the long run though, the Obsidian system is a skeleton for the DM's and players imagination, a framework that is durable enough to handle a variety of "hey...let's try this" ideas. So if you want to have success and failure points more within each segment, you can likely do that.

The DM can also make certain segments different by changing the primary skill (an elegant suggestion that was posted earlier) or by adjusting the DC for one segment only. Perhaps the second segment is exceptionally difficult so the party gets a -1 to their skill checks for that one, etc.