"2d8 as an option?" Topic

I am trying to look at the advantages/disadvantages of using one of 3 types of dice 2d6, 2d8, or 2d10 for dice rolling in a skirmish game.

The basic mechanic the sum of Nd6 + attribute. This is then modified by bonuses or penalties. Dice rolls are opposed and the player who rolls highest succeeds, the other fails. On rare occasion, the player makes an unopposed test roll. In this case, the person rolls 2d6 + attribute and has to beat a target number.

The question is, are there particular advantages or disadvantages to using 6-, 8-, or 10-sided dice?

Would there be a preference to one or another?

Thanks!

Dice with more sides let you "fine tune" modifiers. With 1D6 and mod is 16%. With a D10 each modifier is 10%, and so on. Now with 2 dice you get a bell curve, but even so it lets you have modifiers up to +4 or +5 so a lot more fine level of detail is possible…

d6 are more common than d10 are more common than d8. Therefore that is my order of preference for them.

Using 2d6 enables you to discriminate to a degree so fine that I think it frankly unlikely that any historical data you might have on anything will be good to a greater precision.

Using two dice does not get you a bell curve, never has, never will. The triangular distribution is never called a "bell curve" by anyone who has even a modest command of both maths and English. Please check your facts before posting nonsense to the internet, as we used to say in alt.religion.kibology.

All the best,

Gokmop the Irrefrangible.

You will still have the same problem only with 2 not likely 9 the most likely and 16 not likely. instead of 2 not likely 7 the most likely and 12 not likely. you need to start with the center and work your way to the ends or it just does not work out right.(I have not see a system that does this) Better yet use one die like a D20 or 2 D10 as 1 to 100.

Considering the MUCH greater availability of polyhedral dice these days, I'm always surprised by how many new games come out, STILL based on the ubiquitous d6. I really wish more games used d10's or even d12's.

If you have two players rolling against each other with modifiers, using more than a D6 creates the problem of widening the variable swing so much, that the modifiers themselves become meaningless. i.e. if I roll a 1 and you roll a 10, then the odd +1 or +2 here or there for important stuff such as cover, are nullified.

Rolling 1D6 keeps things tighter, rolling 2D6 helps group 'typical' rolls towards the 6,7,8 results.

I am surprised that more is not made of Average Dice 2,3,3,4,4,5

2 x D10s give a very nice distribution, if you add them together – most results of 10 – 11, and less frequent ones as you progressively move to the extremes.

nDm does not give a true bell curve, but it does approximate one (with the curve becoming smoother (and "flatter") as n increases) which can be quite useful.

Doug

Which is more important, the die roll or the attribute?
A few d6 or d10's added to a 1-100 attribute places a lot more emphasis on the attribute than the die roll.

Anything that makes you buy Strange Dice is bad in my opinion.
Non Strange dice are D6 and D20. Anything else is just odd.

Nothing rolls as well as d6's and d20's.

John that is too much old school thinking. I think d20's are odd. I like d10 and d8 though. Although I am especially fond of my 3 inch D6's. Yes I have 3 inch dice and they make a resounding thud.

so, would a d8 be compromise between the advantages/disadvantages of a d6 versus d10? With regard to the effects of modifiers, etc.?

@SidtheSingh

I suggest thinking about this from the other way

What are the different events/outcomes that you want ? What are the probabilities? Build a table. Then start fitting that to a dice combination

You are likely to find that using more than a d6 results in a very complex system with lots of possible outcomes and lots of modifiers, that is going to be harder to play. This may be exactly what you are looking for, or not :)

I would also add that when using 2 dice the modifiers become non-linear when added together- 3 small modifiers added together can give a huge probability shift.

As an example, if you see 4 possible outcomes (say dead, wounded, forced to flee, attack missed) where the middle two are twice as likely as the others, then an average die (as normsmith says) is all you need.

I am playing around with a single d10 based system, because I want to have lots of different troop modifiers (say fresh veterans with general present, uphill from opponent) and to have them act in a linear way. A d10 also makes it easy to figure the probabilities :)

John

Several points:

1. If you are using opposed die rolls, why roll 2Dx at all?
The outcome of an opposed die roll is binary: one side or the other wins. If both sides use the same die roll, e.g. both sides roll 2D6, both sides have an equal probability of winning (this is making abstraction of the probability that there is a tie, but a tie does not change the fact that both sides have an equal chance).
Hence, it makes no sense to use anything else but a single die in opposed die rolls, unless the number of dice is a modifier in your mechanic.
Note that modifiers on the die roll result are orthogonal to this. Modifiers can be used on a 2Dx roll as well as on a Dx roll.

2. I also strongly object against calling 2Dx a bell curve. It is not a bell curve. It is a triangular distribution.

3. Applying +- modifiers on the die roll result when using more than a single die has strange consequences. Modifiers do not change the probability of outcomes in a linear manner, as a modifier would do on a single die roll.
The clue in understanding this is that if you roll die against a target number, what you really should be looking at is the cumulative distribution curve – which states the probability of scoring a number x or lower.
When rolling a single die (e.g. one single D6), the cumulative distribution is linear: the increase in probability of scoring a <=3 to <=4 is the same as the increase of scoring <=4 going to <=5 (16.6%).
This is not the case when rolling a 2D6. The distribution of 2D6 is triangular, hence the cumulative distribution approximates a quadratic increase (and a decrease in the second half) over the target numbers. A +1 modifier (basically reducing the target number by 1) is much more effective (%-wise) in the middle range of the curve that on outer ranges. With multiples dice (nD6), this effect is increased. The higher n, the more we approximate a true Gaussian bell curve in distribution, and the steeper (relative to other parts of the distribution) the cumulative distribution gets around the expected value of the die roll.

4. People often get confused with opposed die rolls. In essence, an opposed die roll simulates D6-D6, anything over 0 gives a win. Now, D6-D6 is the same triangular distribution as 2D6-7. This is easy to see, since the range of numbers with their probabilities for a single D6 is exactly the same as the distribution for 7-D6.
Hence, subtracting die rolls from each other does not change the shape of the curve, but does shift the curve left or right on the spectrum of outcomes. 5D6-4D6 has exactly the same shape as 9D6, except that 9D6 is a curve centered around the mean of 9*3.5 = 31.5; and 5D6-4D6 is centered around 3.5.
Therefore, numerical modifiers in an opposed die roll system have the same disadvantages as described above. It is much better to alter DICE TYPE (e.g. a character rolls a D8 to shoot, but this gets altered to D6 or D10 or even more depending on circumstances) when using an opposed die roll system. That also has its problems, but smoothes out the unevenness of pure numerical modifiers (the reason being that dx-dy, with x>y, has a middle range of values each having the same probability of occuring).

For simulating and visualizing various distributions, use anydice.com. Great tool!

But, and I really think this is the most important criterium – in the end it is all about what mechanic you think is the most fun to use :-)

In a time were most wargamers scrounge the most obscure corners of the internet in the hunt for exceptional figures, some apparantly still have problems of buying a bunch of D8's?

I have been in gaming for over 30 years, and I cannot count how many dice of all types I have accumulated through over the years – most of them not specifically bought but just included in games of all sorts.

There is, of course, one big point in favor of using polyhedral dice sets for a game: they're just really cool to look at and handle (in my opinion). Despite that, I'm currently really digging the possibilities of nD2 systems.

Doug

One more note about opposed die rolls. They can be binary but not necessarily. If the difference matters, as in DBA, then that will have a big effect on what dice and how many. In this case using single dice may yield too few outcomes, or they may be too extreme.

I agree though: work out the outcomes and build backward t the dice.

Note that in DBA it's not the difference that matters, so much as it is the quotient, which gives a very different dynamic. For example, a +1 to one side is definitely not the same as a -1 to the other.

Doug

Thanks for all the comments and thoughts. This is very much appreciated.

@Phil Dutre … thanks for the detailed explanation regarding modfiers and dice rolls. That is one aspect I was having a hard time wrapping my head around.

I see the advantage of d4 through d12 is you can shift the dice sized rolled as a mechanism.

If you want your game to be attractive to the general public, D6 only.

If you want your game to be attractive to the general public, D6 only.

Nonsense.

One of the most popular games ever published uses a plethora of dice types.

One of the most popular games ever published uses a plethora of dice types

Yes, but not many games have a chance to become "one of the most popular games ever published".

The "less popular games ever published" are more attractive to the general public if they avoid strange dice that average people are not familiar with.

Rarely use opposed die rolls. Takes too much time and players attempt too much mental mojo and body English to get a good roll. I use with impunity whatever die face number suits my needs. For example in my modern naval game I use 8 sided die, but they are blank 8 sided die which I have painted the faces black, yellow, red, or blue for hits on armor (flotation), superstructure, guns, or speed. White faces are misses. quantity of die thrown is a function of power over range.

I use 12 sided die for air to ship combat and vice versa.
Again colored faces, not numbers determine the results. Here it's take the number of air factors firing on a ship. Same as above, but A GREEN face means the air factor is damaged and returns to base or carrier and can't be used again this battle, and a dark green face means the air factor is shot down by AA. Against ships with heavy AA both green faces mean sot down.

Don't need 20 sided die-- damn things never stop rolling!

That's a clever approach to do away with damage allocation rolls and/or tables!

Doug

The problem with rolling 2-dice with a system that allocates +/- points (for high skill, bad weather, etc), is that it tends to screw up the triangular distribution – an extra point can seriously unbalance what were nuanced tables or threshhold numbers.
So I've moved away from 2 dice, whatever their face count, back to 1D12s, 1D10s and even 1D20s…if only they'd roll nicely.
Now when will 1D14s and 1D16s become widely (and cheaply) available?

If we're going to be pedantic, die rolls have multinomial distributions. The triangular, like the "bell curve" are both continuous distributions.

you have two players rolling against each other with modifiers, using more than a D6 creates the problem of widening the variable swing so much, that the modifiers themselves become meaningless. i.e. if I roll a 1 and you roll a 10, then the odd +1 or +2 here or there for important stuff such as cover, are nullified.

Actually, that is evened out by the lower probability of being at the extremes. If you think about it rolling two d10 with modifiers or two d20 with double the modifiers gives you exactly the same success failure distribution. Since the d6, d8, and d10 don't aren't harmonic, there are slight differences in the resultant distributions, but they are as close as you can get at that level of granularity.

The real answer about the difference is already above – it depends on how much fine control you want in your system. The larger the die, the more fine control you have.

I agree with the above sentiment that 2d6 is probably all the control you need. I don't really care about historical data, that's pretty much beyond the ability of the overwhelming majority of players to distinguish on the fly in a game without grinding things to a halt while they analyze.

Two d4 is really probably enough control, but most people don't like the way d4 roll.