Mechanics question:

[StormBringer]

Hello, everyone!

I just had a quick question about some mechanics I was mulling over.  Nothing too elaborate or detailed, just a starter to get my brain going:

I wanted to use a mechanic similar to Alternity among others, in which higher scores used larger dice.  I also wanted to give the opportunity to improve these, as for skill use with experience.  The base mechanic I was thinking of was to have the die type rolled, add in modifiers, and compare to a difficulty rating.  Not terribly different than many other systems.

As far as the improvement of the ratings, I was going to make it expensive to raise the die type, as from d6 to d8, but make it cheap to increase a modifier to the die, as from d6+1, to d6+2.  Also, increasing the die type would cancel out the modifiers so far increased, similar to WEG's d6 as used in Star Wars.  So, d6+2 would increase to just d8.  I was thinking of putting a cap of about 20 on the modifier, but that isn't graven in stone.  I'm still debating that.  Of course, it depends on the structure of the difficulty ratings, and the other situational modifiers that I decide upon.  I also considered allowing improvements to multiple die of each type but that would seem to discourage increasing to the next die type.

My first question is, does this seem balanced as a mechanic?  Without going through every permutation of the possibilities, I wanted to make sure it didn't break down at higher levels.  If anyone has experience with these kinds of systems, I would appreciate input as to how well these work.  Handling time isn't really critical, but if there are some suggestions for improvements to that, I would appreciate those as well.

Thanks for all the tips, in advance!

[Mike Holmes]

So, d6+2 would increase to just d8

Hmm. Interesting. The +1 added raises the average to the same as the next die (4.5 in the d6 to d8 example), and the max to just one under the max (7 in the example). Wheras the next increase only raises the max by one.

If these cost the same, you have a sort of breakpoint here. The +1 levels are more efficient.

Of course going from D12 to D20 (assuming you have em) would be a whole different deal.

Mike

[StormBringer]

So, d6+2 would increase to just d8

Hmm. Interesting. The +1 added raises the average to the same as the next die (4.5 in the d6 to d8 example), and the max to just one under the max (7 in the example). Wheras the next increase only raises the max by one.

If these cost the same, you have a sort of breakpoint here. The +1 levels are more efficient.

Of course going from D12 to D20 (assuming you have em) would be a whole different deal.

Mike

I haven't hammered out the exact numbers, but that is one of the things I was concerned with.  I was hoping to mitigate this by using a mechanic like this:

Intelligence: d8+1
Decipher Skill: 8
Appropriate workspace: +3
Computer&software: +2
(attrib d8+1)+(skill +8)+(situ +3)+(situ +2)=d8+15
Or, there would be a range of difficulty ratings from 16-23 that would be attainable.

So, raising the mod for Int to +2 would raise the range to 17-24, while raising the die type to d10 would increase the range to 16-25.  Of course, these would only have meaning when placed against the backdrop of the whole task resolution system, but I haven't worked that out yet.

The first instance, where the range as a whole in raised, there are obvious applications for making higher difficulty ratings possible.  The second instance does the same thing, in essence, except that the range itself is also increased.  I guess without having a result for difficulty ratings lower than the minimum of the range, it would be hard to judge which increase is better.  Unfortunately, I won't have those kinds of details for a while yet.

So, with a cap in place on how many levels of mod one could get, it would also eventually be desireable to buy up the die type itself, since a wider range of results could also be deisriable, mechanically.  So there is some degree of control over which the weighting of the usefulness of each method.

Hopefully, that adds to the explanation a bit.  Otherwise, I would be happy to clarify.

[Andrew Martin]

I wanted to use a mechanic similar to Alternity among others, in which higher scores used larger dice.  I also wanted to give the opportunity to improve these, as for skill use with experience.  The base mechanic I was thinking of was to have the die type rolled, add in modifiers, and compare to a difficulty rating.  Not terribly different than many other systems.
...
Intelligence: d8+1
Decipher Skill: 8
Appropriate workspace: +3
Computer&software: +2
(attrib d8+1)+(skill +8)+(situ +3)+(situ +2)=d8+15

Sounds a bit complicated to have some character descriptors as dice types, and other character descriptors as modifiers. Why not express all attributes as dice types and roll all appropriate descriptor dice at once? Very quick and easy to read.

[StormBringer]

Sounds a bit complicated to have some character descriptors as dice types, and other character descriptors as modifiers. Why not express all attributes as dice types and roll all appropriate descriptor dice at once? Very quick and easy to read.

Yeah, there was that, but I didn't want to add to the confusion and limit the range of possibilities.   Using just die types for everything is a bit more grainy than I was looking for, although I wouldn't be adverse to expanding what uses die types.  Perhaps attributes and skills can use die types, then other situational modifiers would be straight adds.

I was also thinking that die type plus mod is pretty standard stuff, and it allows players to take a bit of uncertainty out of the task resolutions.  Plus, buying up to d6+2, and adding that in would require a bit of extra description.  Otherwise, players would be confused as to whether d6+2 combined with a d8 is supposed to be (d6+2)+(d8), or (d6+d8)+2.  The latter is essentially getting a free modifier, and the former can get confusing in the middle of play, as one would have to dig out the specific die types, and add the individual modifier to them, then add them all together.

However, expanding the die type progression to other things isn't an idea I am against.  Perhaps weapon damage, or as a way to determine random difficulty ratings.  For instance, a level 5 difficulty could be randomly determined on a d6+1 and added to a base difficulty rating or something, if the GM doesn't have all the facts to make a determination, or perhaps they are in an area of the planned scenario the GM hadn't prepared for.

I think the core mechanic would work the way I have described, but it is very very rough in it's current incarnation.  I was hoping to polish this part up a bit, but of course, without anything to really measure it against right now, that is not something I am looking to do right away.  I understand this is something that will evolve as I get a better idea of what I want to accomplish, and what scope I want to encompass.

Thanks for the input, however.  I will consider what else would be good to have as a die type progression.  Can you think of anything that would work well having this kind of progression?

[M. J. Young]

If I'm understanding the progression here, a character goes from die plus zero to die plus one to die plus two, then upgrades to the next die type. I'm assuming that this does, or could, go from d4 to d6, d8, d10, d12; the jump to d20, as someone has already observed, puts a major kink in the process. However, I think that there's a misunderstanding of the progression even as it is. These are the numbers I see; since I apparently can't use HTML to create a table here, you'll have to do the best you can with the layout I've used:

Die+Mod--Min--Max--Average
d4+0-------1-----4-----2.5
d4+1-------2-----5-----3.5
d4+2-------3-----6-----4.5
d4+3-------4-----7-----5.5
d6+0-------1-----6-----3.5
d6+1-------2-----7-----4.5
d6+2-------3-----8-----5.5
d6+3-------4-----9-----6.5
d8+0-------1-----8-----4.5
d8+1-------2-----9-----5.5

The difference between any die plus one and the next higher die can be described thus:

the maximum roll is one point lower for the d+1
the minimum roll is one point higher for the d+1the average roll is identical.[/list:u]

Thus it appears that the difference between d4+1 and d6 is that you can roll one point higher, but also one point lower, with the d6, and otherwise the chance of success will be the same. The d6 gives you one chance in six of making that roll that needs a six, but also gives you one chance in six of not making the roll that needs at least a 2, which for d4+1 is a guaranteed minimum. Advancing from d4+1 to d6 is not a statistical advance; it is only a psychological one.

The difference between any die plus two and the next higher die can be described thus:
the maximum roll is identical
the minimum roll is two points higher for the d+2
the average roll is one point higher for the d+2[/list:u]
And, missing from this picture, you've got one chance in 4 of getting the six on d4+2, and only one in six of getting it on d6.

Obviously, going from any die plus two to the next higher die unmodified is in every way a step down.

You can get away with the pretense that die plus one is an intervening step because of the psychological advantage of the possibility that you can roll one number higher, because most gamers won't recognize that it also includes the risk that you will roll one number lower. Die plus two is not a valid step.

Hope that helps.

--M. J. Young

[StormBringer]

You can get away with the pretense that die plus one is an intervening step because of the psychological advantage of the possibility that you can roll one number higher, because most gamers won't recognize that it also includes the risk that you will roll one number lower. Die plus two is not a valid step.

Hope that helps.

--M. J. Young

I dig the numbers, those are pretty much the way WEG laid it out, too.  I was actually thinking of letting the modifier go as high as 20, or something.  Substantially higher than 2, at any rate, as 20 may be a bit high.

Along with the situational modifiers, and skill or other such character descriptors, I was going to test against a difficulty rating.  Perhaps as high as 30 or more, for difficult tasks.

Therefore, the die would represent the random nature of the character attribute.  For instance, no one can keep track of exactly what a character may or may not pick up in various studies, so Intellegence would represent some bit of recalled trivia, perhaps, or a sudden flash of insight from a past experience relating to the current one.  The modifier would be one method of advancing the character, perhaps representing additional overall studies or experience, in the case of Int, while the skill ratings would represent static increases in body of knowledge.  Perhaps the GM would even allow direct increases in the modifier, as a reward, instead of having the player pay the cost to increase it, since I was going to make that relatively cheap compared to increasing the die cost.

So what I was shooting for was a subtle interaction between skill increases, and ability increases, with ability increases themselves being another subtle interaction between adding to the modifier, and increasing the die type.  If the cap for the modifier were to be set higher than about half the die type, however, then raising the die type would have to include the modifier already purchased, or much of it anyway.  Otherwise, it would be pointless to raise the die type.  There would be a certain point where the die type would have to be raised, if the player wanted to raise that particular skill as much as possible, but they also shouldn't be penalized by losing the entire modifier, especially when they have raised the modifier to something like 8 or 10 already.

Now that I think about it, letting them keep the modifier when they raise the die type wouldn't actually be so bad.  There is still the choice there, and someone who only raises their die type could still be outstripped by someone who only raised their modifier.

Also, thanks for pointing out the large gap between d12 and d20.  Perhaps the highest die type could be d12, or possibly, go from d10, to d12, to 2d8, to d20.  The averages would then progress as 5.5, 6.5, 9, and 10.5  There would be a rather odd jump of 2.5, then 1.5, however.  But the idea of multiples for die type increases could also be viable.  From d6, to 2d6 for instance.  Perhaps with that mechanic, the modifier cap could be a good deal lower, too.

I guess my ideas are still rather disorganized about this.  :)  Thanks for the input, tho, it is helping me get some things sorted out.  Like my goof on the math in a previous post.  :)

Thanks again to all, I will see about organizing some of these ideas a little better.

[Christoffer Lernö]

2d8 and so on have quite different spread from even distributions of Dn dice. It's quite a difference.

That aside, I'd like to throw out a warning here. I've been playing Earthdawn. ED uses a variant of this scheme. Unfortunately it ends up being one of the most frustrating mechanics to play in because of the varying results.

I played a warrior which occasionally could kill almost anything. That was on a lucky roll. Bad luck put me with the fumbling idiots. The problem was that ED didn't only use a success/fail mechanic, but also implemented a "fail/succeed by" mechanic. That screwed things up most terribly.

For example I had friggin horrible sword. I could do something like D20+D8+D6 for damage. Compare this to others having something like 1D12. Still, a lot of times I could perform a lot worse than them on the rolls. It just felt so... unreliable.

Besides, the system was designed so that unless you had a certain margin of success your damage was sucked up by armour and things. Basically you sat there and hoped for the high end results on all your dice no matter how high your rating. It didn't matter, you always had to roll high, because there was always a good chance of failure.

Of course, the standard target number being around 9 and an average human having 1D6 didn't exactly help things either.

Anyway, all I'm saying is that you might take note of the possible whiff effect of this mechanic. Die+Mod is more predictable... in a good way.

[Andrew Martin]

2d8 and so on have quite different spread from even distributions of Dn dice. It's quite a difference.

That aside, I'd like to throw out a warning here. I've been playing Earthdawn. ED uses a variant of this scheme. Unfortunately it ends up being one of the most frustrating mechanics to play in because of the varying results.

I played a warrior which occasionally could kill almost anything. That was on a lucky roll. Bad luck put me with the fumbling idiots. The problem was that ED didn't only use a success/fail mechanic, but also implemented a "fail/succeed by" mechanic. That screwed things up most terribly.

For example I had friggin horrible sword. I could do something like D20+D8+D6 for damage. Compare this to others having something like 1D12. Still, a lot of times I could perform a lot worse than them on the rolls. It just felt so... unreliable.

Besides, the system was designed so that unless you had a certain margin of success your damage was sucked up by armour and things. Basically you sat there and hoped for the high end results on all your dice no matter how high your rating. It didn't matter, you always had to roll high, because there was always a good chance of failure.

Of course, the standard target number being around 9 and an average human having 1D6 didn't exactly help things either.

Anyway, all I'm saying is that you might take note of the possible whiff effect of this mechanic. Die+Mod is more predictable... in a good way.

I have to agree with Pale Fire. In my S system, I intially tried using target numbers, but quickly discovered in play test that it gave similar effects to the above. It was best to use a opposing dice as the "difficulty" level. Then even when the player rolled a "1" (lowest and worst result), they could still have a small chance of success, if the opposition rolls a "1" as well.

One other significant problem I had was players not rolling their dice at the same time. If the opposition rolled first with a high dice type and got a high result, a player wouldn't bother to do "anything" as their character would automatically fail. I got around this by making players re-roll all dice if they weren't rolled at the same time, and if players did nothing, their totals were zero and things got worse for them.

[M. J. Young]

Therefore, the die would represent the random nature of the character attribute.

This may sound entirely counter-intuitive, but have you given any thought of advancement by  decreasing die types?

Hear me out on this. Let us suppose that the ordinary character starts with d12+0. That means he's got an average roll of 6.5, a maximum of 12 and a minimum of 1. If we go to d12+1, we increase these each by a point, and at d12+2 we increase them by two points.

But what if the big increase was to drop to d10 and add two to the modifier? A character going from d12+0 to d10+2 keeps the same maximum, raises the minimum from 1 to 3, and so boosts the average to 7.5. It could happen at any point, as long as the die drops two pips and the mod increases by two simultaneously.

In this construction, part of the concept of advancement is that your abilities become more reliable as you get higher.  A d12+8 and a d4+16 have the same maximum roll, but the latter will make it 25% of the time while the former will only make it 8.33% of the time; and the latter will never roll below 17, while the former might roll as low as 9.

I've read too many threads, so I'm not completely certain at what you are aiming; but this might be something to consider.

--M. J. Young

[Jeremy Cole]

As a thought, what about going backwards, D20 to D12 ... D4.

It seems to me that as people improve they get more consistant (less deviation from one attempt to the next), rather than less consistant.  If a brilliant character has a D4 his variation is small, while a clueless (or maybe just untrained) fellow will wander all over the place on his D20.

If the target numbers range from 1 (uber-hard) to 20, then anyone has a chance of doing anything, which I think is fun.  The chance of success ranges from 5% to 25%, perhaps not enough variation.

As another thought,
Perhaps you could have an attribute (D20 is best) and subtract from that a skill dice (D4 is best), needing to beat a number in the 0 (very easy) to 19 (near impossible range).  So an average build fellow (D8 attribute) is brilliantly trained in melee (D4 skill) and would roll a D8, subtract a D4 from that, giving an average success of 4.5 - 2.5, 2.

Any thoughts?

[Andrew Martin]

As a thought, what about going backwards, D20 to D12 ... D4.

It seems to me that as people improve they get more consistant (less deviation from one attempt to the next), rather than less consistant.  If a brilliant character has a D4 his variation is small, while a clueless (or maybe just untrained) fellow will wander all over the place on his D20.

If the target numbers range from 1 (uber-hard) to 20, then anyone has a chance of doing anything, which I think is fun.  The chance of success ranges from 5% to 25%, perhaps not enough variation.

As another thought,
Perhaps you could have an attribute (D20 is best) and subtract from that a skill dice (D4 is best), needing to beat a number in the 0 (very easy) to 19 (near impossible range).  So an average build fellow (D8 attribute) is brilliantly trained in melee (D4 skill) and would roll a D8, subtract a D4 from that, giving an average success of 4.5 - 2.5, 2.

Any thoughts?

I did some thinking about this on RPG.net a while ago, in this thread: http://forum.rpg.net/showthread.php?s=&threadid=13232

[Jeremy Cole]

Interesting reading.

I really like the D100 to D2 in eight steps proposed there, opposed by a D100 to D2 difficulty.  I like the idea that something that is normally quite simple can just go wrong and be utterly impossible sometimes, even from a D100 or D20 position.

I guess you could have level 1-4 in an attribute (say 2), and then 0-4 in skill (say 1), giving you your skill level (in this case 3, meaning a D12).  Or a flat attribute level modifies the dice roll, or no there is no attribute, or like I said a dice roll for the attribute and one for the skill (but then you have 3 dice rolls for one simple action).

Is anybody thinking of using this mechanic in a system?

[StormBringer]

Excellent ideas.  I really dig the idea about using smaller dice as the skill increases.  That is the kind of thing I think I was heading for, but didn't realize it.  After I mull these ideas around a bit, I will post some more.  Thanks for the great ideas, I will see about hacking around with them for a bit.

[Andrew Martin]

Interesting reading.
...
Is anybody thinking of using this mechanic in a system?

I thinking of using it as an alternative dice system in my S system for players that want less variance for higher skilled characters. I haven't yet figured out how I'd handle players using both systems (could just ask one player or side to translate their dice?).