Changing some things...looking for opinions and answers

Hi everyone (i’m new here),

I have been reading this forum with great interest and browsed through the ACKS (a friend of mine bought it and is a big fan).

I have some ideas (aka preferences) for new/changed rules and i was wondering if some of you are interested in commenting on them and, hopefully, can give some informed feedback what i’d have to change to get the “same” results (or better: feel) with the changed rules.

So here goes:

  1. Randomizer:
    I am not a big fan of linear distribution. Unfortunately all the d20 games are based around just such a distribution…rolling a d20. I’d like to exchange the iconic d20 roll with a roll of 3d6. The mean-value is exactly the same (10.5). The range is a bit more narrow (3-18 instead of 1-20) but that’s ok. I like the more “realistic” distribution of a bellcurve but i think i will have to fiddle with modifiers to get similar results (aka not shifting success-chances too much in either direction).

  2. the role of armor:
    I never liked the idea that armor makes you harder to hit…imo, it makes you harder to hurt but most of the time easier to hit. So with that in mind, i’d like to change that armor adds to the AC and instead want armor to generate a damage-reduction. Not sure how much hp per armor-value. Maybe you have an idea?

Now with these two changes in mind…how would i have to change modifiers etc. etc. so that the chances to be successful at skills and combat remain more or less the same (as i said, the feel of the game should not change dramatically, or it would be more than just some rules-tweak).

Anyone any ideas on the topic? Any professional gamedesigners around who have crunched mechanical numbers?

Anyway, thanks for reading and thanks for any ideas, opinions and pointers you can share.


Welcome Loki,

to your questions/changes

  1. what about using 2d10 bell curve without the hazzle to adjust numbers?

  2. think about it: a good starting point for cons and pros is perhaps here:

Focused effort to be helpful:

1 = 3 100.00 (100%)
1 = 4 99.54 (100%)
1 = 5 98.15 (98%)
2 = 6 95.37 (95%)
3 = 7 90.74 (91%)
4 = 8 83.80 (84%)
6 = 9 74.07 (74%)
8 = 10 62.50 (63%)
11 = 11 50.00 (50%)
14 = 12 37.50 (38%)
16 = 13 825.93 (26%)
18 = 14 16.20 (16%)
19 = 15 09.26 (9%)
20 = 16 04.63 (5%)
20 = 17 01.85 (2%)
20 = 18 00.46 (0%)

That should give you similar results and “feel”.

Giving in to being unhelpful:
Having said that, if the odds are the same, what’s the point of having a bell curve? And if the odds aren’t the same, you’re certainly not going to get the same results or “feel”.

Armor is messier. I have some ideas, but they’re only half-formed.

I’m not sure that I have much advice, but perhaps I can add some perspective.

Linear distribution vs. bell curve doesn’t really make much difference when hitting specific numbers isn’t important. What’s important is the ration at which you hit. We’re seeking out a certain percentage of success. And for that, a linear distribution is just a lot easier and sensible to work with. As another post says, if you work out the math so that your bell curve provides a similar liklihood to hit than why even bother using a curved distribution? You’re essentially negating the effects of the curve anyway.

Armor as DR
Technically, it already is. I think you may be getting hung up on what “hit” means. Hit doesn’t mean physical contact. It means you’ve decreased an enemies capacity to fight. Hit actually DOES mean damage (or fatigue or whatever). So a miss is effectively a pure reduction of capacity to fight. AC and HP are abstractions. So if you want armor to be damage reduction, it’s already factored in by the fact that better AC increases the number of “misses” or rather it decreases the amount of damage accumulation.

I agree with what others have said.

1d20 and 3d6 are two different animals… they inherently give different feels according to the results. For instance, 1d20 is more random and favors weaker combatants, whereas 3d6 is more predictable and favors stronger combatants. In terms of “realism”, considering the “fog of war”, I’m not convinced that one is more realistic than the other.

Armor already is a form of damage reduction, even when it simply improves your AC target value. As an example, take a character with a Dex of 13:

Unarmored AC = 11

Now he dons banded armor:
Armored AC = 16

When wearing the banded armor, any attack with a result in the range of 11 through 15 may be considered to “hit” the character, but the “hit” isn’t enough to penetrate the banded armor, thus it’s a form of damage reduction.

This doesn’t answer the “easier to hit, but harder to penetrate” part of course.

HERO uses 3d6 to hit and armor as DR. I believe it does provide an answer to being “easier to hit, but harder to penetrate” as well. It may be worth taking a look at, to see if you can retro-fit any ideas from it into ACKS. It’s very rules intensive compared to ACKS, however.

Having said all that above, I’ve toyed with giving characters and monsters some sort of defense value and assigning armor a damage reduction die as a counter to a weapon’s damage die. For instance:

Cloth: 1d2
Leather: 1d3
Scale: 1d4
Chain: 1d6

Combatant A has a dagger; combatant B is wearing chain mail.

  1. A attacks B and hits
  2. A rolls 1d4 and deals 3 points of damage
  3. B rolls 1d6 and gets a 2… he takes 1 point of damage.

I never fleshed it out because it adds more steps to resolving combat… but it’s just an idea.

I honestly believe the “easier to hit, but harder to penetrate” part is purely flavor and is part of the abstraction. I generally consider “misses” as simply “attacks that deal no damage”. The flavor does not require that I describe a “whiff” or “airball” attack. I consider a “miss” to be many things. Glancing blows, parries, dodges, contact without damage, etc. I would assume that even in a strict AC environment, one wearing heavy plate would be much easier to “hit” or make contact with, but that the contact would less frequently result in actual damage or HP loss. This is part of the abstraction of AC and HP. A low roll may actually make contact but the contact was not substantial enough to justify a damage roll.

I have a friend who uses 2d10 for AD&D 2E instead of 1d20. We switched over to it at the start of a gritty megadungeon game where some of the PCs were able to begin play with plate armour. They became nigh untouchable by low HD creatures and it gave a fantastic ‘Knight vs Peasants’ feel to combats. Those of us not so lucky were still in fear for our lives… And armoured opponents were likewise very scary.

Changing the dice makes things less swingy, and has a huge impact on the ‘feel’ of combat - at least for us it did.

I can't recommend changing armor to be damage-reduction.

First off, it's not really historically more accurate than armor-as-hit-chance. The recent book "From Sumer to Rome" showed, with quite comprehensive physical modeling, that most ancient weapons were not capable of piercing or cutting through ancient armor under combat situations. They conclude that dealing casualties required hitting the foe where the armor wasn't - stabbing in a weak point between shoulder and cuirass, neck and helm, and so on. This is better simulated with AC-as-hit-chance than AC-as-damage-reduction.

Second, it requires completely re-balancing the game. If you want to try, the most detailed discussion of the mechanics involved can be found in 3e's Unearthed Arcana, now available at

Thanks a lot, everyone.

Some comments:

  1. Well, you guys are right, if i shot for exactly the same chances with the new dice, it kind of wouldn’t matter (although i am sure you would never be able to get exactly the same chances anyway). I was not very clear as it seems. I just don’t want too widely different results. As an example: If you have a +4 to hit and you try to hit AC 18, you have to roll 14 or above to hit. With a d20 you have a chance of 35% to roll a number of 14 or higher…with 3d6 you only have around 16% … that is quite a lot. So what i was trying to ask was: “how can i adjust that a bit”. I am not that familiar with ACKS yet. So i am not sure how all the modifiers (for AC and to-hit) are handled and how extensive they are.
    Possible tweaks: - lower AC a bit
  • increase to hit modifiers a bit.
    The same problems turn up with skill TN.
  1. Ok. Alex and others make a good case for AC as it is. I never claimed it really IS more realistic…it’s just that in my mind it always made more sense. Because in an armor worthy of its name you are slower and more “clumsy” so you should be easier to hit. Anyway. I will read up on all the posted links on the topic…thanks for that.


If you’re interested Loki, here’s a spreadsheet I put together to show the probability and target number differences between 2d10 and 1d20 and uploaded to Google.

I might have done 3d6 as well, somewhere…

Thanks James. Very interesting. The changes to probabilities are all over the place (or at least it looks that way without doing an indepth mathematical analysis :wink: ). Hmm…not good.
Well, i guess i have to try it out and see how the feel is.
Or i could just keep everything as written :smiley: (But that is the easy way out…the path of the dark side :wink: ).

Randomizer. I’m actually quite fond of 3d6 - the majority of my GMing experience is in GURPS. If you want a similar probability to the D&D stuff, but with less swinginess, I would recommend halving all 1d20 modifiers.

Because of the very different distributions, the average bonus difference between the two is +0.8 on 3d6 ~ +1 on 1d20, but the average across all starting points does not describe the typical game. The typical game - especially with ACKS, which makes a point of shooting for this - has most of the starting points around 10-11. And at that point in the curve, +0.4 on 3d6 ~ +1 on 1d20. Nudging it slightly to 0.5 ~ 1 yields something that “feels” right across most of the spectrum that is actually used, but also gives more solid results in general.

Note that this brings up a number of specific issues that need addressing:

  1. All bonuses on 1d20 need to be halved. Proficiencies. Woundwort. Blessings. Everything. Anything that isn’t halved is suddenly far more effective. Those proficiencies which grant a +1 on attack or AC are probably too powerful and should be removed and replaced with something else - swashbuckling and the blade dancer’s ability, in particular, are hugely powerful on 3d6.

  2. Magic items. A +1 bonus on a magic item is roughly twice as powerful. You can either halve the probability of finding it, or halve the bonus, or just accept that magic items are going to be more awesome.

  3. Attack throws and saves should improve half as fast. That is, a fighter’s attack throw reduces at the rate of a mage’s; and a mage’s attack throw against AC 0 at level 9 will be 9+. This gives substantially less “feel” of improvement, but has similar probabilities.

  4. I would recommend setting an automatic hit at an unmodified roll of 17+ and an automatic miss at 4-. This achieves similar likelihood as 20 and 1, p. 102.

  5. Armor should be categorized as light, medium and heavy, granting AC 1, 2, and 3 respectively. Light armor can be worn by thieves. Shields can still grant +1 AC; they’re a bit underpowered as is.

  6. Saves should be re-aligned, since they are the only item that isn’t targeted around the 10-11 range. Basically, if you leave saves at their current starting point, you are making save-or-die effects about three times as scary, or chance of survival one-third as high. I’m not going to do the math because really, the whole save system needs to be re-numbered!

Awesome post, Thomas. Thanks a lot!
And don’t worry about Saving Throws…i guess i will just keep the d20 for this.

You know, i always wanted to play GURPS, but alas, as of now, i never have. I guess my groups are not into “testing new games”. They always say “why learn a new rules system? We already do know d20/AD&D…that’s enough…”.
Oh well… :-/