Is it me, or is there a glitch in the DoS mechanic?

Degrees of Success: When a test succeeds, the character making the test automatically gains one degree of success (abbreviated as DoS).He also gains additional degrees of success equal to the tens digit of the target value minus the tens digit of the roll.A character who rolled a 23 when he was testing against a value of 66 would pass the test with 5 degrees of success (1 for succeeding, plus 4 more for the difference between 6 and 2). - p13

OK, so...doesn't that mean that if your skill is bang on a '10' multiplier (10, 20, 30, 40, etc.) then it's almost impossible to score 1 DoS (1% chance). You will almost always score 2+ DoS.

E.g.

Shooting with WS40

I roll 41+ i fail.

I roll 40 i succeed with 1 DoS (1+(4-4))

I roll 30-39 or less i succeed with 2DoS (1+(4-3))

I roll 20-29 or less i succeed with 3DoS (1+(4-2))

I roll 10-19 or less i succeed with 4DoS (1+(4-1))

I roll 0-9 or less i succeed with 5DoS (1+(4-0))

And...doesn't that mean that until you hit a '10-threshold' you're actually getting worse as you increase your chances of only getting 1DoS?

Given that you can now choose your characteristics with 'points build' and can therefore ensure your Characteristcs hit the 'tens', isn't this a bit odd?

okay, if you decide to punch a guy in the face, your agility is 40, your weapon skill is 40, and your strength is 40.

If you roll a 40 or higher, you succeed, for another dos, you need to roll 50 or 60 or 70 or 80

40-49 is your initial hit and makes the attack a success.

if you WS is 44, and you roll a 25 then you succeeded with 1 dos.

So I really don't see how it is glitched.

You're not actually getting worse. Your chances of getting additional degrees of success stay the same, while your chance of getting one degree of success/any success at all increase. Yes, if you only take into account all successful options, your chance of getting 2+ degrees of success decreases if you only account for the successful numbers you can roll, and ignore failed numbers.

WS40(2.5% chance of 1 DoS, 25% of 2 DoS, 25% of 3 DoS, 25% of 4 DoS, 22.5% of 5 DoS) to

WS41(4.88% chance of 1 DoS, 24.39% of 2 DoS, 24.39% of 3 DoS, 24.39% of 4 DoS, 21.95% of 5 DoS)

However, the actual roll you make includes everything from 1-100, and thus the actual numbers would be:

WS40 (1% chance of 1DoS, 10% of 2 DoS, 10% of 3 DoS, 10% of 4 DoS, 9% of 5 DoS.

WS 41 (2% chance of 1 DoS, 10% of 2 DoS, 10% of 3 DoS, 10% of 4 DoS, 9% of 5 DoS.

So no, you don't ever get worse from increasing a characteristic. HOWEVER, you only improve your potential multiple DoS when the 10s digit increases, otherwise you're just increasing your chance of gaining 1DoS.

Also worth noting from this, the increase of the 10s digit from 0 to 1 is actually worse than all the others due to the fact that only 9 results exist in that range to grant the extra DoS, compared to the 10 results for every other 10s range. That's pretty interesting.

d% Engine Failure

http://community.fantasyflightgames.com/index.php?/topic/87219-d-engine-failure/

I don't think there is too much here to hold against the system. Its always been the case that its "best" to worry about just the bonus for any given statistic, and that the ones digit is almost only there to increase the % chance of success.

The current system makes the math a whole lot faster; it removes previous techniques that were often error prone.

Also, with the consideration of how common permenant characteristic reduction is, its actually far more wise to have a stat exist somewhere above having a 0 in the ones digit. As then taking just one point of characteristic damage greatly reduces your maximum DoSs.

Changing the language to count passing a test as a Degree of Success makes sense; the old system was always one of the more awkward mechanics to explain to new players ("It's a success- but not a Degree of Success...").

But the base numerical change is just dumb. It throws linear progression out of whack (BS 30 rolling a '29' is two DoS, but BS 39 rolling '30' is only one DoS, despite passing the test 8% better than the previous guy). The argument that it 'makes the math easier' is absurd- figuring 10% increments of 100 is just about the easiest bit of math one can do- anyone too dense to be able to work out that 43 is 10 less than 53 is too dumb to be able to grasp any RPG mechanics, so who is this change targeting?

This is an example of arbitrarily changing something that isn't really broken- and replacing it with a experimental system which contains worse glitches...

I'm getting my phd and have played with two engineers and someone with a biochemistry degree. We all stumbled on the math of things like "okay what's 57-29...." because the natural inclination is to do the full subtraction rather than using a heuristic to make it easier (subtract the tens digits, if roll ones digit is equal or lower than the target ones digit, add another dos". Even then, that is a lot less intuitive than just subtracting the tens digit. The glitch mentioned at the start of this thread doesn't actually exist. Yes, there is the problem of a roll difference of 1 possibly having an equal effect as a roll difference of 10. This was obviously a conscious decision made by FFG to include this issue, though. I imagine they decided that its worth the added ease and speed for the game. It does effectively, for DoS mean increasing the 10s digit is the only way to increase DoS. Given how much more important DoS are for things like RoF, this change makes some sense. You have to look at it from a more broad level (what is this change doing to rolls as a whole) rather than the individual instance level (its totally unfair that this roll that beat the target by less than mine did got an extra DoS!).

But the base numerical change is just dumb. It throws linear progression out of whack (BS 30 rolling a '29' is two DoS, but BS 39 rolling '30' is only one DoS, despite passing the test 8% better than the previous guy). The argument that it 'makes the math easier' is absurd- figuring 10% increments of 100 is just about the easiest bit of math one can do- anyone too dense to be able to work out that 43 is 10 less than 53 is too dumb to be able to grasp any RPG mechanics, so who is this change targeting?

Wow. There's some hate.

My cousin introduced me to RPGs. He's a comic book artist and a hell of a storyteller. Never met a better role-player in my life. He's not quick with math, though - degrees of success, damage vs. armour vs. penetration, etc. all slows him down a lot. I'd still play with him over anyone else in the world.

The 40kRPG line is already pretty crowded with on-the-spot calculations. Not hard stuff, but there's a ton of it, and people get confused. I'll gladly take any change that reduces that problem.

okay, if you decide to punch a guy in the face, your agility is 40, your weapon skill is 40, and your strength is 40.

If you roll a 40 or higher, you succeed, for another dos, you need to roll 50 or 60 or 70 or 80

40-49 is your initial hit and makes the attack a success.

if you WS is 44, and you roll a 25 then you succeeded with 1 dos.

So I really don't see how it is glitched.

You're talking about DH1, though. In DH2, that would be 3 degrees of success.

(Emphasis mine in both quotes.)

I still think it's a bit weird that the degrees of success depends on the target number. If the target number is 40, rolling a 39 nets you two degrees of success. If the target number is 41, rolling a 40 to hit is one degree of success. That's the kind of inconsistency I don't really see any point in introducing. What was wrong with the old system?

I agree on that point. 1 DoS for every 10 points worked well, especially if you gain an automatic DoS anyways.

... We all stumbled on the math of things like "okay what's 57-29...."

If you are subtracting 29 from 57 to figure out Degrees of Success in DH1.0 , you are doing something horribly, horribly wrong. it's incriments of 10: 57-47-37-27-17-07... that's by far the easiest math option. I'm terrible at math (I constantly had to work out sums on my fingers playing D&D 3.5 ), and I've never had a problem quickly counting DoS in DH1.0 ...

The 40kRPG line is already pretty crowded with on-the-spot calculations. Not hard stuff, but there's a ton of it, and people get confused. I'll gladly take any change that reduces that problem.

So, drop percentages and house-rule your own campaign as d10-based, and don't expect all players to accept glitchy totals to compensate for a 'problem' that only effects a tiny handful of players.

... We all stumbled on the math of things like "okay what's 57-29...."

If you are subtracting 29 from 57 to figure out Degrees of Success in DH1.0 , you are doing something horribly, horribly wrong. it's incriments of 10: 57-47-37-27-17-07... that's by far the easiest math option. I'm terrible at math (I constantly had to work out sums on my fingers playing D&D 3.5 ), and I've never had a problem quickly counting DoS in DH1.0 ...

But thats the thing, my gaming group consists of software developers and engineers, and it all to often seems like we resort to counting on our fingers or saying the increments out loud.

All too often I've said "Alright, 57, so 47, 37, 27" while holding up my hands to come up with total DoSs. Its clunky. As of now I can look at any two numbers and quite easily apply this method. Its a flat 2 steps instead of X steps.

O(2) is better than O(X)

Personally I consdier the exact effects they have on the probabilities to be largely inconsequential. Its like less than +/- 10% on results in the 1DoS/F range.

O(2) is better than O(X)

I dunno, maybe if you're pushing for that agenda. The way I see it, the new system is "okay, so I roll a 27. So that's 2. The target number was what, 63? So that's 6. That gives me four, no wait, five degrees of success." vs. "Okay, I roll a 27 against 63. That gives me a degree at 27, 37, 47 and 57."

The complexity of the math is pretty minuscule in both cases, and arguably slightly less so in the old system (assuming you grew up with a base 10 number system). In return, you get some pretty quirky probabilities. I get that the new system is trying to do the same thing as the old system in an easier way, only it doesn't strike me as any simpler, and it introduces some new inconsistencies.

I don't think it accomplishes what it's trying to do, any better than the old system (especially if it's not supposed to accomplish anything new.)

Now that I think about it, the new ruling does equalize characteristics somewhat. The statement the rule makes is, increasing your characteristic bonus is profitable above and beyond the percentage chance of success it brings

In previous rules, this was categorically true for some stats, situationally true for other stats, and completely irrelevant for others, because only specific rules (some of them not even applying to all characters) ever bothered with specific characteristic bonuses. The only three that really mattered for everyone were Strength (damage and carrying capacity), Toughness (soak and carrying capacity) and Agility (movement rate and initiative), the rest were only important to people with specific skills and talents.

Now, with a very basic and very vital mechanic being tied to characteristic bonuses, the field is somewhat evened out in that regard.

Personally I consdier the exact effects they have on the probabilities to be largely inconsequential. Its like less than +/- 10% on results in the 1DoS/F range.

Then drop the percentiles and use 1d10 in your game. Don't expect everyone else to gladly accept results that are objectively wrong (turns out 9% is more than 1%- who knew?), just because you personally consider anything under 10% to be irrelevant- when that's clearly not the opinion of most gamers.

Personally I consdier the exact effects they have on the probabilities to be largely inconsequential. Its like less than +/- 10% on results in the 1DoS/F range.

Then drop the percentiles and use 1d10 in your game. Don't expect everyone else to gladly accept results that are objectively wrong (turns out 9% is more than 1%- who knew?), just because you personally consider anything under 10% to be irrelevant- when that's clearly not the opinion of most gamers.

First, you and a few people posting here in agreement with you don't actually make "an opinion of most gamers", so I'd ask you to either back your claim up with some solid statistical evidence or rescind that argument.

Second, even if you have hard statistical evidence, it's still an argumentum ad populum, and thus irrelevant to the discussion of whether the rule is good or not.

Third, the rule in question isn't objectively wrong. It's just different and not mathematically intuitive, but "lower roll is better" is in no way an objective fact when discussing percentile roll-under mechanics. You may not like it, but it's as good a method for measuring the totally abstract thing like quality of your imagined success as any other you may devise.

Fourth, even if you were right on any one point you've made, it wouldn't make your tone any less awful.

I still think it's a bit weird that the degrees of success depends on the target number. If the target number is 40, rolling a 39 nets you two degrees of success. If the target number is 41, rolling a 40 to hit is one degree of success. That's the kind of inconsistency I don't really see any point in introducing. What was wrong with the old system?

Aye, that's the glitch in a nutshell.

Simply saying 'every 10pts below the skill number = 1 DoS' removes the threshold issue.

An odd little glitch easily fixed.

Ill say again that going up or down by increments of 10 from the target number has not been the natural inclination of anyone I've gamed with, and that I believe the book itself encourages just subtracting the two whole numbers. It's my opinion that a simple, "start with one degree of success or failure, subtract from the 10s digit to add more" is more intuitive than "start with one degree and add or subtract from the target number until you reach the result closest to above or below your roll, depending on if its a failure or a success."

I don't really think the "glitch" is a glitch so much as there being a change in how DoS are emphasized in characteristics (only the 10s digit matters for DoS, 1s digit matters for chance of any success). The "glitch" is so obvious to notice that I can guarantee its not an accident; the game was designed with this change in mind. I also like the added benefit gained to increasing the 10s digit of stats that don't use their bonus frequently.

I still like the "1 plus tens column of your roll is DoS" method. It doesn't fix the problem that DoS chance is bound to the tens column of the stat, but it does remove the one DoS glitch. It's also the simplest way of calculating DoS.

I still like the "1 plus tens column of your roll is DoS" method. It doesn't fix the problem that DoS chance is bound to the tens column of the stat, but it does remove the one DoS glitch. It's also the simplest way of calculating DoS.

It's also somewhat counterintuitive as it changes the simple concept of "roll under" to "roll under, but not too much under".

I'm just not seeing people shouting "hell yeah, natural sixty three!" around the gaming table

Morangias, that's a fair cop, but it could be easily explained as you succeeding in a situation where lesser skilled characters would not. "Anyone could have made that test with a 01, but only I, a true professional could have handled THIS situation, requiring at least 63"

I call the "use the 10s digit from your dice" the EasyDoS rule, and we still use it for 2nd edition. No maths. (Well... a +1, but that hardly counts)

I don't really think the "glitch" is a glitch so much as there being a change in how DoS are emphasized in characteristics (only the 10s digit matters for DoS, 1s digit matters for chance of any success). The "glitch" is so obvious to notice that I can guarantee its not an accident; the game was designed with this change in mind. I also like the added benefit gained to increasing the 10s digit of stats that don't use their bonus frequently.

If the target number is 40, rolling a 39 nets you two degrees of success.

If the target number is 41, rolling a 40 to hit is one degree of success.

I'd suggest it is a glitch whether you think it or not. DoS chances are skewed around the 'tens' threshold. Better higher skills have a lower chance of 2+DoS. That's a glitch.

Better higher skills do not have a lower chance of 2+ DoS. I already put the math earlier in the thread. If you'd care to refute that, please feel free to post me the numbers/logic behind it.