6=7, many people complained about it. A TN 6 is equal to a TN 7. There is no possible chance of getting less than a 1 on a die. It is important when adding up modifiers, as many has pointed out, since 6+1 is different from 7+1, which is true, and it is the best argument to support keeping 6=7. I'm not here to discuss the merits of 6=7, rather I intend to propose a novel solution to the problem for those who want it.

One proposed solution was a 1s never count system, but that only created a 7=8 problem. But what if, instead, we subtract 1 from the value of all sides of out dice, so that 1 becomes 0, 2 becomes 1, and so on, and we subtract 1 from all TNs so that the minimum TN becomes 1. This gives us a 0-5 die with average with the same probabilities on the first 5 TNs. The Rule of 6 becomes the Rule of 5. A TN of 6 (Formerly 7) is reached if one rolls a 5 (Formerly 6) and a 1 (Formerly 2), thus removing the 6=7. Shifting all values back one also means that a TN of 7 (Formerly one must roll a 5 and a 2 (Formerly 3), and so on. The 7=8 problem is avoided.

I don't know why this would be particularly useful, I just thought of it and felt like posting it.

Or you could just accept 7, and every 6th number after that one, being a "gimmie" in order to make what would be utterly brutally punishing utterly brutally punishing, but with a small bone thrown to the player.

Or just accept that 6=7, for sufficiently large values of 6.

you can greatly reduce the 6=7 issue by, as i recall, rerolling on a 5 or greater and subtracting 1 from the reroll. so, for instance, let's say you roll a 5; that entitles you to a reroll, which comes up 6; that entitles you to a further reroll, which comes up 4. your total roll for that die is 5 +6 +4 -2 (for two rerolls) = 13.

you can reduce it even more by rerolling on 4+ and subtracting two.

That's been proposed before—the real problem with it is that it either makes a bunch of tests significantly harder (basically everything TN 11 and greater, though you notice it on 7-10 as well), or makes you go back and redo a gigantic swath of TNs—which are substantially at nice, visually appealing points right now.

Edit: Eyeless discusses some solutions here.

~J

that's surprising, to me. i wouldn't think the -1 per reroll would mean more, statistically, than doubling your chances of getting 6+ per die and reroll. regardless, it seems like it'd be easy enough to skew the system such that the issue is pushed off to TNs of, say, 18+, at which point it generally becomes academic.

Whoops. No, I was talking about the reroll-6 add-5 solution. I don't remember offhand how big a chance your suggestion was.

~J

i'm pretty sure i remember seeing charts of it, compared to charts of 6=7, and thinking "that's not too far off."

if you really hate the 6 =7, whenever they roll a 1 following an exploding die, give them a 50% chance of going to either 6 or 7. no fancy math, and keep it simple without modifying to much of the mechanics. just delay the game for a moment.

Really I've learned to deal with it the 6 = 7 and feel it works just fine. besides those mooks have the same advantage, along with anything you throw at the players.

Hum hum, there is a problem with the 0-5 and -1 to all TNs though. While it is no different for TNs of 2-6, which will still have the same odds, it quickly becomes a problem as it makes higher TNs even harder to get to. Lets see an example. As rules are, a TN of 15 requires a roll of 6, then another roll of 6, followed by a roll of 3 or higher. Now, at your proposed system, the TN would change to a 14, and you would now need a roll of a 6 on the die, then another 6, followed by a 5. Obviously the test has become more difficult now, and there is no real way to fix it. If you say 'well, let's just add another -1 at some point' then you run into some TNs being adjusted to be exactly the same, which was the original problem.

Don't know about other solutions, I'd have to look at charts for the reroll on 5+ thing to know how that works out.

It's a feature, not a bug!

No complaining about chicago´s local wildlife please. The white man has done enough already to damage nature, no need to endorse people who want to nuke natures best...

--
Our prefered solution was the reroll 5+6, but only add 4 solution. The reduction in variance is also neat...

My solution: (Take this with a giant grain of salt as I have played very little SR3)

7 does not exist. TNs of 1+6n where n is a positive integer do not exist. Thus a TN of 6 with a +1 to TN penalty becomes a TN of 8. Any TN assigned in the books that would be eliminated (I'm pretty sure that there aren't any) is reevaluated to a multiple of 6, for a result of the exact same probability. I think this is in line with the dev's thinking. If you look at the TNs for weapon range table in SR3, the TNs go 4,5,6,8 (IIRC) indicating that there shouldn't be a TN of 7. This keeps the math minimal as long as you don't have bonuses or penalties in excess of 5. I would think this wouldn't occur too often.

that just shifts the problem one place. instead of 6=7, you end up with 7=8.

4, 5, 6, 9, weapon ranges. (Here nor there, although I used to use the Vision Mag rating as a reduction of the TN rather than range category, so a Extreme Range shot with a Mag3 scope was a TN6, rather than 4. Mag didn't reduce the TN under the base for the shot though, so no Mag3 at Short range-- it was still TN4.)

4,5,8,9 for some weapons.

~J

I wasn't aware of a variant Long range, but I'm mildly curious. Which weapons got the 8?

And to at least pretend this isn't off-topic, the 6=7 thing never bothered me. My first group added the next result to 5 instead of 6, but after that group disbanded I never saw that houserule again. If you had a TN6 and something took it to 7, it was basically a little "Merry Christmas" moment; you could have been getting screwed harder by the system but you weren't.

@Telion: I also considered the 50-50 roll for 1's on the re-roll, but I came down on the side that the one thing SR didn't need was more rolling.

Grenades (thrown and launched), mortars, rockets, missiles, and, curiously, target designators.

I was about to make a quip about how, while the designers obviously didn't care enough to fix 6=7, they did care enough to not give anything 7 as a fixed, unmodified TN, but the Tiffani Needler just had to come along and prove me wrong. Still, it's the only TN I can find offhand in {6n+1 : n ∈ N} that's specified as such (that is, not part of a sliding TN or the result of modifiers).

~J

the reason 6=7 is unwanted is that there's a skip in the progression of difficulty. you have a fixed chance of succeeding at TN 5 with X dice. that chance goes down when you roll against TN 6. it should go down again when you roll against TN 7, but it doesn't. and at TN 8, there's a big jump in difficuty. that's the issue that these fixes are attempting to resolve.

with your solution, the same problem exists, it's just bumped up by +1 TN. you have the same fixed chance of succeeding at TN 6, but if the TN is 7 (say, base TN 4 and +3 for some situational modifier), you treat it as an 8--you have to roll a 6, and then reroll a 2+, to succeed. if the TN is actually 8 (base TN 5 and a +3 modifier), you of course treat it like the 8 it is. you've got that same lack of progression, and the same big jump in difficulty.

Actually, at TN 8 there's a very small jump in difficulty (6/36 -> 5/36), except insofar as the last "jump" was zero so it's infinitely larger than the last jump. The closer you get to a multiple of 6, the steeper the probability curve. Your other criticisms of that plan stand.

~J

Is it really that big a problem, unless you're the kind of person who worships probabilities and orgasms over statistics?

that doesn't match my experience, or what i've come to understand from discussing TN probabilities with others. i could be misremembering, but from what i recall, succeeding at TN 8 is roughly twice as difficult as succeeding at TN 6/7.

Nope, it's at the 5/6 or {n-1/n : 6|n} boundary that things are twice as hard. Odds on a single die go:

psh, yeah, if you trust numbers and science.

Transitive relations may lie to you, and the power set of multisets may deceive you, but the probability mass function is always faithful.

(Guess what I've been doing for the last sixteen hours solid!)

~J

Something from which you ought to have a long break?

I had about an hour-long nap before the combinatorics started again, does that count?

~{J : J ∈ {dead_people}}

My solution was always to not care. It worked well.

Math make head hurt.

You know, there really -isn't- a solution to the 6=7 'problem' that doesn't involve affecting the difficulty of other TNs in some manner one way or the other, just accept it as a limitation on the very nature of using dice to make random numbers. The one solution out there would be to figure out the exact % chance of getting any particular number and having a computer generate a random number. Then you would tweek the TNs of 7, 13, 19 and the others to be a % somewhere inbetween 6 and 8. (well, 13 between 12 and 14 etc. of course.)

Actually I take back my first statement, there is a solution, and that is when you get a six, then a one, you reroll the one with needing a 4, 5, or 6. But this is just more dice rolling, does the game really need that?

I'm too tired to be sure, but I think your suggestion is just a slightly obfuscated version of reroll-6, add-5.

~J

Ditto. I always liked 6=7, in fact, because it gave you a little more incentive to try and work your combat options (using cover, take aim actions, stacking the right bonuses and stuff) in order to hit that "sweet spot."

I think you are over thinking it.

As I understand it a 7 is not 'useless' because of maybe modifiers. for example sure a '7' for a target number shooting is a dead cert if you have only that, but a +2 for a smart link means you have to roll a 5 or better to hit (1 in 3 chance) but if it is a 6 normally and you have the smart link you need to only beat 4,(1 in 2 chance). Or if you are a bit further out and the target number to shoot is a 9, there is your incentive to close the range until it is only a 7. etc

I just like 6=7 because it's a little, as was said earlier, "Merry Christmas" moment.

The system's way of self-correcting for it's extremely steep difficulty. It's like saying "We're using bell-curvacious dice on a linear difficulty scale that gets logarythmically more difficult as you go higher. So as a way of giving something back, we'll just make every multiple of six just as hard as every multiple of six +1, or make every multiple of six +1 just as easy as the multiple of six. Therefor, you can eke out that little bit of extra chance without more risk than would be incurred on a six.

It's hard to tell, but it looks like the effective TN shift magnitude gets increased by one when going past a multiple-of-6 boundary (henceforth referred to as "the 6 boundary", even if it's 12<->13, 18<->19, etc.). What happens when going downwards, or when both positive and negative TN modifiers make it cross both ways, or when crossing due to multiplication, isn't clear and may make TN summing non-associative and non-commutative.

~J

so if the TN is higher than 6, you add 1 to the TN; when it's higher than 12, you add 2 (or, rather, another +1 for a total of +2), and so on? that would make a lot of tests a whole lot harder. someone else will have to figure out how such a scheme would work out, as far as smooth progressions go.

It was uncommented, so again:

Reroll 5+6, add 4. 5 is unchanged, 6 achieved at 2/6*5/6=10/36, 7 at 2/6*4/6=8/36. Both used to be 6/36. 8 is 2/6*3/6=6/36 now. And it is smoother because it uses more balanced hit/miss propabilities for rerolling.

smooth compared to what? gravel street you're being dragged on? O.o

Got that one from DS back in the day - one of the few houserule suggestions that was suggested once and accepted without discussion.

So you do find it workable in day-to-day play? I've always sorta looked askance at it. I'll have to slap together a chart when I get the chance.

~J

Under the current system: TN 8, +3 TN = TN 3, +8 TN = TN 11

Under Narse's proposed system: TN 8, +3 TN = TN 11, TN 3, +8 TN = TN (3 + 3=6) + 5 TN = TN (6+1)+4 TN = TN (6+1+1)+4 TN =TN 8+4 TN = TN 12

In other words, base TNs and modifiers stop being commutative. Never mind the probability distribution, that's a deal-breaker for me right there.

~J

yar, that's what i meant by the progression of difficulty. probably shouldn't be using the same term to refer both to this and to the array of chances to succeed at TN X with Y dice, but whateva.