What do you think of the die mechanic?

[momosnyx]

Each time I think there's no new way to roll dice, a rpg appears and uses a new die mechanic...

If I understand the OVA rules right, you roll 2 dice and an additional die for each +1 you have in an ability that can be applied to the situation. Nothing new so far.
The new thing is how to calculate the "result" of the roll:
For each of the numbers from 1 to 6 add up the rolls of the respective number. Take the highest sum as the result of the roll.
I haven't looked into the probability distribution of the rolls, sorting the dice, then adding them up, comparing the results to the other sets doesn't seem too elegant to me. (I haven't used the system in play, maybe I'm exaggerating.)
Any ideas why this might be a better system then let's say adding 2d6 and the boni from your abilities? (That's no snipe on behalf of BESM, just an example for an - IMHO - easier die machanic...)
Comments would be highly appreciated

[Clay]

Obviously, I'm SLIGHTLY biased, but I thought you might want to hear some of the inside scoop on the mechanic just the same.

The probability distribution is pretty solid. I've run tests with the base die-roll compared to the difficulties and it seemed to pan out.

Although certain combinations appear more than others (there's far more ways to end up with 12 than 15) the actual end result, when compared to the DN chart, is fairly structured. In this situation, there are few, if any, combinations between 12 and 15. This is why 15+ is in a difficulty class by itself. It includes any total that requires at least 3 dice a like. (15, 16, 18, etc.)

The reason behind the whole mechanic is so the little guy ALWAYS has a chance against the big guy. When you have a traditional addition mechanic, especially one with flat skill bonuses as opposed to additional dice, there's very little room for variation. This may work for a realistic game, but I feel it goes against a very crucial vibe in anime.

And finally, the die mechanic may seem complex, but it's actually pretty easy to pick out the correct dice. Unless you're one of those characters chugging 7 dice at once, there's only going to be a few duplicates. I've never had a complaint in the playtest. (Hey, at least I didn't go for Risk-style comparison like Ironclaw. Interesting, but...takes some time.)

[migo]

The probability distribution is pretty solid. I've run tests with the base die-roll compared to the difficulties and it seemed to pan out.

Can you share some of those results?

What I've seen so far is that with an average ability (rolling 2d6) the probability looks like this.

1 100%
2 97% (-3%)
3 89% (-8%)
4 77% (-12%)
5 62% (-15%)
6 39% (-23%)
7 8.3% (-31%)
8 8.3% (-0%)
9 5.5% (-3%)
10 5.5% (-0%)
11 2.7% (-3%)
12 2.7% (-0%)
13+ 0%

That just looks like a weird probability table to me. The difference in chance of success between two neighboring numbers is between 0% and 31%, the difference in differences ranges from 3% - 8% but that doesn't scale smoothly. Am I missing something here?[/quote]

[Clay]

You have to see the Difficulty Number chart to properly understand it.

Let me restructure your chart with this in mind.


2 97% Easy
4 77% (-20%) Moderate
6 39% (-38%) Challenging
8 8.3% (-30%) Difficult
10 5.5% (-3%)) Very Difficult
12 2.7% (-3%) Extremely Difficult
15 0% (-3%) Nigh Impossible

The drops in percentages are around the 30% mark from 2 to 8. Once you go past that, there is no longer room to drop. You've already reached a point where it is practically impossible to succeed at these difficulties. As you go up in dice, the scale goes more smoothly down the entire chain of rolls. It's not perfectly smooth, but no RPG can claim this once you factor in things like bonuses and skills and whatnot.

I think the lack of the difficulty table was the biggest contributor to your confusion. Hopefully this is more acceptable.

[Andrea from Italy]

Clay,
would you consider posting here the tables with percentages / difficulty for 3, 4 etc. dice?
Regards,
Andrea

[momosnyx]

Obviously, I'm SLIGHTLY biased, but I thought you might want to hear some of the inside scoop on the mechanic just the same.

Yes, I had hoped so

And finally, the die mechanic may seem complex, but it's actually pretty easy to pick out the correct dice. Unless you're one of those characters chugging 7 dice at once, there's only going to be a few duplicates. I've never had a complaint in the playtest. (Hey, at least I didn't go for Risk-style comparison like Ironclaw. Interesting, but...takes some time.)

I may be overestimating that as I didn't try it in real.
The example on p. 78 (pdf numbering) uses 12 dice. Reminds me of my Shadowrun games a couple of years ago

[momosnyx]

I think the lack of the difficulty table was the biggest contributor to your confusion.

I didn't see that connection... It seems odd that 7/8, 9/10, 11/12 have the same probability of coming up, but with the difficulty table, it makes more sense. Have to think a little more about it & calculate the distribution for dice numbers > 2.

[Clay]

Despite a long time searching, I could not find my old probability tables.

Because the "combine like dice" has no easy corellation I can find in math, the only way I could do it was manually calculating all the probabilities one by one. An arduous and annoying task, and doesn't work above 3 dice. 4 dice has far too many possibilities to write out by hand.

I went ahead and went through the process again, and here are the results.


2 Dice

2 100%
4 80% (-20)
6 37% (-43)
8 9% (-28)
10 6% (-3)
12 3% (-3)

3 Dice

2 100%
4 93% (-7)
6 59% (-34)
8 23% (-36)
10 15% (-8)
12 7% (-8)
15 1% (-6)


With two dice, the average person will succeed at an easy task all the time, a moderate task most of the time, a challenging task less than half the time, and difficult tasks only rarely. Nigh impossible tasks turn out to be plain impossible.

A moderately skilled person (+1) however, will succeed at both easy and moderate tasks almost all the time, challenging tasks more than half the time, and even have success at difficult tasks roughly 1/4 of the time. From there, the probability grows gradually less.


When you look at the numbers in a purely mathematical way, the percentage progression may not seem to make sense. But when you take the verbal descriptors into play, it becomes a lot more appropriate. There are a degree of difficulties the person is good at, a few he's okay at, and then it drops sharply for tasks that are out of his skill range. However, these tasks are NOT impossible, just unlikely. In most addition based systems, more difficult tasks are simply not possible under any circumstances. The OVA mechanic always gives the underdog a chance, and always lets a degree of luck play a part. How many times have you seen the most unlikely characters do something successfully in anime?

[Clay]

Oh, forgot to comment about your "too-many dice" concerns, momo.

The example on page 76 (PDF page 78) is an extreme example in order to show the mechanic at its most complex. 12 dice is actually nearly the maximum possible in the game. Few occassions will warrant more than +10 bonus

The most dice you'll be chugging realistically is 7. Most of the skilled combat characters have a +5 bonus either to attack or defense. That's not to say more dice will never happen. I'm just saying 12 dice is the exception, not the norm. More reasonable examples can be found on 77's opposed roll explanation and the examples detailed under Combat.

[Andrea from Italy]

Thanks for the table. I guess of one wants to calculate the % for more dice, the only chance is writing a pc program. I might do it... I am sick at home with flu. We'll see

[Andrea from Italy]

>With two dice, the average person will succeed at an easy task all the time

Wait a moment. Isn't a double-1 a failure at an easy task ('cos 1s are never summed?)

[Clay]

Yes, you're right. At one point in OVA's production, the "don't add ones" rule was optional. I did my statistics without it, forgetting I had since made it a built-in rule. Oops. ^^; Either way, it's an extremely rare occasion, and not likely to change my tables very much.

Sorry to hear about your flu though

[Andrea from Italy]

It's not a big deal, I'm already getting better. 'sides, it's more free time to desing adventures for my doomed group to play in
I don't have the time to refresh my Pascal enough to write the % calculator now, but maybe in the future sometime.
Anyway, I am not concerned with % too much, knowing that you have given some thought to the mechanics is enough for me, I am not a believer in simulationism in RPGs.
Regards,
Andrea

[Andrea from Italy]

Clay,
please note that this thread at the Forge: http://www.indie-rpgs.com/viewtopic.php?t=14285 has people discussing dice roll stats.
One of them (grand commander something) mentions a java program he has made for calculating %, maybe it might be worth your time to check with him if he can adapt it for OVA.
Regards,
Andrea

[Clay]

Thanks for the tip, Andrea! I'll have to look into it.