The New School Manifesto

[erik]

Yeah, I wasn't really certain what I was gonna find on Wolfram Alpha and was very pleasantly surprised. I tried a few wordier searches before attempting XdY notation. It still didn't give me all I wanted, but more than I was expecting. Unfortunately it isn't giving me all the probabilities I want (i.e. chance of multiple doubles/triples/etc.). That is where it starts to get more complicated and really not conducive to having printed out tables to reference.

Also apparently once I go to 11d10 and higher amounts of dice on wolfram alpha, it no longer shows the probabilities for dice. *big sad face*
(edit: okay, maybe that was just 11d10... weird. 12d10, 13d10, 14d10 and 15d10 are okay, but again not that helpful since they don't give probability of additional doubles/triples/etc.)


I've thought more and decided I was crazy for thinking more dice is going to equal more backlashes, at 10d10 every additional die is essentially going to equal at least another success. Of course I don't know how big the dice pools are (it sounds like 8 or so is normal tho) or if they are even using d10's.

[Lokathor]

http://topps.diku.dk/torbenm/troll.msp

http://www.diku.dk/hjemmesider/ansatte/ ... ckRef.html

Does just about any dice method you could come up with, and then some. Only trouble is that converting it into the proper code snippet can be a little fiddly sometimes. But once you've done that it'll print you out the exact probabilities.

[Kemper Boyd]

Hi!

I thought I would just make a quick post despite being at work right now.

The system uses only d10's, using huge amounts of different dice would be a hassle and make testing the system really awkward.

You practically never roll more than 10 dice (stats go up to five, skills go up to five). If you have a relevant trait, you get a dice you don't roll: instead you roll the other dice and then declare what's the trait dices value for that roll, which ensures you always get at least one matching set.

The magical backlash thing is kinda cool, since it only appears towards the higher-end pools and doesn't have much of an effect if your magical skills are weak. And like the blogpost says, you can mitigate effects of backlash by planning ahead.

Tell you what, I'll put up a new post about the subject soonish since it's apparently a thing that raises lots of questions.

[PoliteNewb]

The system uses only d10's, using huge amounts of different dice would be a hassle and make testing the system really awkward.

You practically never roll more than 10 dice (stats go up to five, skills go up to five). If you have a relevant trait, you get a dice you don't roll: instead you roll the other dice and then declare what's the trait dices value for that roll, which ensures you always get at least one matching set.

The magical backlash thing is kinda cool, since it only appears towards the higher-end pools and doesn't have much of an effect if your magical skills are weak. And like the blogpost says, you can mitigate effects of backlash by planning ahead.

I don't know if you understand how stupid this whole backlash thing is. Allow me to demonstrate.

I rolled 8d10 ten times, to simulate a fairly powerful wizard casting a spell ten times. The results were:


2 pairs, 4 unmatched
1 triple, 1 pair, 3 unmatched
1 pair, 6 unmatched
2 pair, 4 unmatched
1 quad, 4 unmatched
2 pairs, 4 unmatched
1 pair, 6 unmatched
3 pairs, 2 unmatched
2 pairs, 4 unmatched
3 pairs, 2 unmatched

So I'm seeing an average of around 4 unmatched dice each time, which according to your blog post is "problematic". Twice I'm seeing 6 backlash dice, which I'm guessing would be pretty damn bad. So this wizard is fucking himself over one time in five. Minimum. He is getting more than trivial backlash 7 times in 10.

Now: there are 2 possibilities:

1.) You don't need 8 dice to cast most effects...in which case, why invest in that many dice? The normal, intuitive design is that as you get more dice, you get better (less chance of fucking up), not worse (as is the case here).

2.) You DO need 8 dice to cast important effects...which means those effects are really NOT available to PCs, because the danger inherent in trying to produce them is really freaking dangerous.

If you tell me "casting X spell at Y power level entails a chance of your character dying immediately" (as your living storm example), you have just told me a spell my character will never ever cast. So why does it exist?

[Kemper Boyd]

I managed to make a post yesterday about the subject: http://lionsofthenorth.wordpress.com/20 ... ery-stuff/

Essentially, yes, it's really bad to roll a big pool. However, it only applies when you have no way to mitigate fro the unmatched dice, which you do mostly via prep (and a few other things not yet detailed).

[Username17]

Essentially, yes, it's really bad to roll a big pool.

That is stupid. There is no way you can possibly polish that turd, it's still a turd. I know you think it's "neat" or some shit, but it's shit. You are talking about a resolution system that takes a relatively long time to complete, and which has a failure rate that rises and then falls as you go up in skill in a totally non-intuitive fashion.

There is absolutely nothing good about your proposed resolution system. It is one of the worst I've ever seen bandied about. And a jackass was literally suggesting a roll under system with degrees of success being calculated on the fly by dividing skill ratings by various different prime numbers just a few days ago. His system was laughably terrible, yours is in all ways inferior to that.

Have you looked at Framewerk? It is a shitty system with no redeeming features. Yours is very similar, but manages to be even worse in every imaginable way. At least with Framewerk dice you only have to count the matches instead of having to count both the matches and the unmatched separately.

-Username17

[Kemper Boyd]

That is stupid. There is no way you can possibly polish that turd, it's still a turd. I know you think it's "neat" or some shit, but it's shit. You are talking about a resolution system that takes a relatively long time to complete, and which has a failure rate that rises and then falls as you go up in skill in a totally non-intuitive fashion.

Well, you're never rolling more than 10 dice, so it doesn't take that long to check how many matched dice you got. For all tasks except magic, that's really it. One step. Two, if you count rolling them I guess.

Counting unmatched dice only comes up with something like the top tiers of spells, and if you do the preparations of casting (detailed in the rules), you don't even have to care about unmatched dice that much.

It all comes back to the idea that magic isn't the universal and best tool for everything. It's powerful, but the sorcerer is always bound to the fact that he or she works with something beyond human understanding. In general, directing the flow of something (like binding the wind or making it rain) isn't a big deal, it's taking actual control of sorcerous power that's risky. In Lions of the North, this idea is more or less explicit.

[Username17]

That is stupid. There is no way you can possibly polish that turd, it's still a turd. I know you think it's "neat" or some shit, but it's shit. You are talking about a resolution system that takes a relatively long time to complete, and which has a failure rate that rises and then falls as you go up in skill in a totally non-intuitive fashion.

Well, you're never rolling more than 10 dice, so it doesn't take that long to check how many matched dice you got. For all tasks except magic, that's really it. One step. Two, if you count rolling them I guess.

It's no more than forty five comparison operations, you stupid twat!

Shittiest resolution system. Ever. You're seriously competing with FATAL for "worst system", and I think you might actually win. If you can't accept that this is an incredibly shitty idea without a single redeeming feature,there is no hope for you or anything you ever write at any point in the future.

-Username17

[Piell]

That is stupid. There is no way you can possibly polish that turd, it's still a turd. I know you think it's "neat" or some shit, but it's shit. You are talking about a resolution system that takes a relatively long time to complete, and which has a failure rate that rises and then falls as you go up in skill in a totally non-intuitive fashion.

Well, you're never rolling more than 10 dice, so it doesn't take that long to check how many matched dice you got. For all tasks except magic, that's really it. One step. Two, if you count rolling them I guess.

It's no more than forty five comparison operations, you stupid twat!

Shittiest resolution system. Ever. You're seriously competing with FATAL for "worst system", and I think you might actually win. If you can't accept that this is an incredibly shitty idea without a single redeeming feature,there is no hope for you or anything you ever write at any point in the future.

-Username17

It's the exact same thing REIGN does, you drooling incompetent, and it's fast and easy. REIGN is a pretty great system in itself, as well.

[Kemper Boyd]

It's no more than forty five comparison operations, you stupid twat!

Shittiest resolution system. Ever. You're seriously competing with FATAL for "worst system", and I think you might actually win. If you can't accept that this is an incredibly shitty idea without a single redeeming feature,there is no hope for you or anything you ever write at any point in the future.

-Username17

Hey, no need to get rude, we're just talking about game rules here

The thing that makes this sort of dice mechanics fast is that we as humans are really good at pattern recognition, such as "look at dice to see what dice have matching numbers."

And REIGN is a big source of inspiration for me.

[Username17]

So... you violated your own design "manifesto" in order to copy shitty mechanics from Greg Stolze? What the fuck is wrong with you?

-Username17

[Kemper Boyd]

So... you violated your own design "manifesto" in order to copy shitty mechanics from Greg Stolze? What the fuck is wrong with you?

Sorry, what part of the manifesto do you think I violated by adapting a good basic dice mechanic? I got confused a bit

[erik]

It's not a good basic dice mechanic. You get worse as you get more skilled. At a certain point calculating probabilities is not intuitive or easy. That's a horrible mechanic.

[Kemper Boyd]

It's not a good basic dice mechanic. You get worse as you get more skilled. At a certain point calculating probabilities is not intuitive or easy. That's a horrible mechanic.

Forgot if I said it here or elsewhere, but it's just the sorcery stuff that has the potential for backlash. No other concepts you'd roll for have to deal with that.

[Username17]

So... you violated your own design "manifesto" in order to copy shitty mechanics from Greg Stolze? What the fuck is wrong with you?

Sorry, what part of the manifesto do you think I violated by adapting a good basic dice mechanic? I got confused a bit

Well, that resolution system isn't good. It's the worst resolution system anyone has ever made. It's slower than any other resolution system, the probabilities it outputs are inane, and figuring out whether a result is "likely" or not is ludicrously obtuse. Your particular flair on it - namely that of making unmatched dice go up against spell-specific critical failure charts is a work of failure that even Rolemaster never quite achieved.

It is garbage design. And if you think it's good or even "not incredibly shitty" then you suck as a designer. If you can't tell me - quickly - what the likely results for a five die test or a seven die test are - and you fucking can't - then your resolution system is not something that you are qualified to design things for. If you're not qualified to design things for your own design, you're incompetent and unworthy of consideration.

But to add to the shit sandwich, your own stupid "manifesto", in the few times it was trying to tell me that rejecting the aesthetics of 1996 made it "new school" in two thousand and fucking twelve told me that player empowerment and light hearted adventure were the things you were aspiring to. You know, exactly the fucking opposite of telling players that they should be super careful when trying to use their abilities because each power was also a land mine where it would randomly trigger things off of secret magical mishaps charts that could kill their character.

So I think it's established: your design is stupid, you're stupid, and your game design manifesto is stupid. I think we're done with this thread. And with you.

-Username17

[Kemper Boyd]

Lets see, I don't have my probabilities chart at hand so let's use Wolfram Alpha.

The chance to roll a match with five dice is one in 1.6.

The chance to roll a match with seven is one in 1.5.

[Username17]

Lets see, I don't have my probabilities chart at hand so let's use Wolfram Alpha.

The chance to roll a match with five dice is one in 1.6.

The chance to roll a match with seven is one in 1.5.

You didn't answer the question at all. You were asked to tell us what the likely results were. And you didn't, because you don't know the answer and don't know how to get the answer.

-Username17

[Kemper Boyd]

You didn't answer the question at all. You were asked to tell us what the likely results were. And you didn't, because you don't know the answer and don't know how to get the answer.

Gotta say, I'm kind of stumped for what the question even is.

For me, designwise, all that really matters is that both 5 dice and 7 dice give you a pretty good chance of rolling a match. It's something that's pretty intuitive.

[Username17]

OK, I'll walk you through this. A single die cannot get a match, because it's a single die. The next die has a 9/10 chance of not coming up a match. The chances of getting at least one match are always 1-(chance of getting no matches). So on one die, you have 0% chance of getting a match, on two dice you have a 10% chance.

This continues, when you roll your third die, there are 8 possible numbers it could roll that don't match either of the other dice. With your fourth die, it's 7, with your fifth die it's 6, the sixth die has 5, and the seventh die has only 4 numbers it could possibly roll that don't produce a match. The chances of getting a match on 7 dice are thus:

1 - (10/10 * 9/10 * 8/10 * 7/10 * 6/10 * 5/10 * 4/10) = 93.952%

That is, by the way, absolutely nothing like 1:1.5, I don't even know what the fuck you put into Wolfram Alpha to get that wholly incorrect number. That you couldn't look at that result and just eyeball that it was totally out of the ballpark means that you don't even have a minimal feel for the kinds of numbers your system outputs. That makes you completely unqualified to write for it.

And that's well before we get into any of the stuff you are intending to do with the matched numbers or the unmatched dice or whatever the fuck. Just the first part of "what are the odds of getting a match?" you were unable to get even to the most generous tolerances when you had unlimited time and access to the internet and a calculator.

Are you beginning to understand why your proposal for a system is garbage?

-Username17

[Stormgale]

Isn't the whole point of doing probability calculations that you can design a system without the results necessarily being obvious? You just do that calculation once for each die pool and then you can design perfectly fine?

Using your math it comes out as follows
Diepool Probability of at least 1 successs
1 1 0
2 0.9 0.1
3 0.8 0.28
4 0.7 0.496
5 0.6 0.6976
6 0.5 0.8488
7 0.4 0.93952
8 0.3 0.981856
9 0.2 0.9963712
10 0.1 0.99963712

That isnt that hard to intuit, more dice equals A higher probability, which slowly lessens in growth at a higher level

[Kemper Boyd]

Whoops, I accidentally used the wrong numbers from the chart! Sorry about that.

Wolfram Alpha's built in dice tool is really cool:
http://www.wolframalpha.com/input/?i=roll+7d10

[erik]

Just a note, Kemper. Frank's tone is harsh, because, well, it is. But don't let that blind you to the arguments that a d10 matching dice system is bad and you should know that this is so.

If you're using wolfram alpha then it should give you the exact probability of not getting a match if you just plug in 7d10 (6.048% just as Frank's math showed).

You ought know your basic chances of success for your system by heart. There's only 10, and the top and bottoms are incredibly easy to remember.

10d10 = 99.9964%
9d10 = 99.64%
8d10 = 98.19%
7d10 = 93.95%
6d10 = 84.88%
5d10 = 69.76%
4d10 = 49.6%
3d10 = 28%
2d10 = 10%
1d10 = 0%

Now ignoring tragedy of design in that the best spellcasters are the least skilled since they at least have little chance of exploding their heads, you still have a problem in the diversity of probabilities that your die mechanic offers.

You really only are likely to use the skills from 4 to 10 dice, and of those, the top 3 are nearly identical. So you basically have the option of having 50%, 70%, 85%, 94%, and 98/99%. Not a diverse spread at all.

[Stormgale]

That of course is assuming you only need a single success, if Kemperboyd is copying of reign (which is amazing) then the width of your dice pool can matter, thus allowing for two tiers of probability, 1: Succeeding at all and 2: succeeding well?

Also This math assumes all the die are different apart from the pair... If I understand it correctly?

[angelfromanotherpin]

It's a big step towards user-friendliness if the players at the table can estimate their odds of success reasonably quickly and accurately. It empowers them to know whether they can expect to successfully swing on that chandelier or whatever. Sure, you could look up the exact odds on a table before deciding, but that slows shit down and more importantly makes the players feel less in control than a system they can calculate in their head.

Gygax put the to-hit tables in the DMG because he was a control freak. Obfuscating the odds is a disempowerment maneuver, and there are games and tones it works for. But if you want a game which is accessible and light-hearted, what you want is a more conventional die-pool or rng+mods vs TN model.

[Kemper Boyd]

Personally, I think player empowerment relates more to giving them some narrative control and making it clear that the player characters are straight away cool and competent people, instead of being guys that die of one hit, like 1st level D&D wizards.