* Fate: It took me awhile to "get" what it was asking me to do and how it worked, but after revisiting it after half a year, it finally clicked. Afterwards, the difficult part was convincing my group. So it may not have been so much of an anti-heartbreaker for me, but it definitely exceeded expectations for my players at least. Now they are open to trying Cortex Prime, which has some similar design principles.
The results are far less random, particularly relative to the size of the modifiers (which are obviously much smaller with 2d6, but still relatively more significant), so investment seems more rewarding, and "just chancing it" is less likely to succeed. In D&D, even if you are really going hard on a skill, i.e. you have Proficiency, and a large stat bonus (which is all you can possibly have without Feats, being a Rogue, or a handful of specific subclasses), because you're rolling a d20, the results are extremely random. You'll routinely and constantly fail checks that aren't particularly hard, or roll so high your investment was immaterial, and it's very easy to see streaks of failure or success. There's no curve (technically it's not a "curve" with 2d6 but let's not get into semantics!). With 2d6, there are a lot more results that are going to result in numbers in the middle, so results are more predictable and situations where someone fails four or even six rolls in a row despite having invested in a skill become vanishingly rare, where they're routine in 5E. It also means that totally unskilled characters are a lot less likely to succeed at stuff just randomly. The situation you see constantly in D&D, if the DM allows it (and sometimes you can't stop it), is that the specialist with heavy investment fails, possibly repeatedly, and some chancer with no modifier just rolls a 19 or whatever. This isn't rare or unlikely - it's routine.
There may be some genre or type of game that benefits from that, but I'd say it's not Heroic Fantasy Adventure (which is what D&D is), or even adventure as a genre at all. I actually can't really think of a genre where "lol so random that 8 CHA Barb succeeded at both these Persuade checks when the Bard failed!!!" seems it would be fitting. I mean, obviously it's not impossible to work with - we all do - but I am consistently not impressed with it. I've never felt like any game I've ever played rewarded me less for investing in a skill than 5E does (except Perception/Insight/Athletics). Which is saying something, especially as the relative cost of skills in D&D is high, given most PCs will only ever have 4.
Modern AGE and the other AGE system RPGs have given me more games sessions than I hoped for. I've GMed this system more times than any other system in the last 40 years, except D&D. It is now my default RPG system.
There's no curve (technically it's not a "curve" with 2d6 but let's not get into semantics!). With 2d6, there are a lot more results that are going to result in numbers in the middle, so results are more predictable and situations where someone fails four or even six rolls in a row despite having invested in a skill become vanishingly rare, where they're routine in 5E.
Technically, no dice are generating true curves, since the results are always quantized into integer-like states, but that's a tiny quibble.
It's only important if modifiers or changes can be applied.
In games like most PBTA, the roll is only ever modified by ability scores. Therefore, the value of dice types is irrelevant provided the modifier is calibrated to the experience set.
In games like D&D 3E, that the modifiers are flat shifts does matter; changing the game from 1d20 to 2d10, aside from the moving of nat 1 to nat 2 and the average from 10.5 to 11, if you have a +5 vs a "standard" TN of 15, the 2d10 is going to get diminishing increase from mods, If you have a peak level 1 of +8, the 2d10 is pretty much swamped, but the d20 isn't; the 7+ needed in the 2d10 is 85/100 and on 1d20 is 70%; adding a +1 bonus makes it 90/100 and 75/100 respectively; but a +2 is 96/100 and 80/100.
If, however, you're using a dump stat unskilled... that 15+ is 15% on 2d10, and 25% on 1d20; with the +1, 21% and 30%, and with +2, 28% and 35%, +3 is 36% and 40%, +4 being 45% on 2d10 vs 45%.
The actual dice only matter when modifiers are used.
Not in many modern games. Most things PBTA have no modifier other than the relevant ability score.
I don't like lacking at least a difficulty system, but it's become a huge subset of non-trad games.
To give an example, if one wants to switch MASHed from 2d6 to 1d100, one can do so with trivial difference in odds.
Strong Result: 10+ =6/36 so becomes 83-100
Weak Result : 7-9 is =15/36 so 42% so 41-83.
Trouble is everything else.
So, we can make a table to replace the +0 to +4 modifiers.
(My rounding is a little inconsistent in the above.) this eliminates modifiers and is otherwise super close to the probabilities in MASHed. One could put those numbers next to the attributes on a grid, and move to 1d100, with no changes to experience mechanics.
Fantasy Flight Games' narrative dice system / Edge of the Empire.
Which is not to say that I didn't have at least some expectation that it would work when I tried it, but the two sessions we got in before the pandemic broke up our group in March 2020 were excellent. I was expecting it to be fun, but was also expecting a lot more choppiness, slowdown, and overall "kludginess" than we experienced.
So overall it very much exceeded my expectations. Once we finish our Tiny D6 / Tiny Frontiers experiment for the next 4-6 weeks, I'm fully expecting to go back to it.
Not really convinced by your case here mate. Particularly not that you appear to be basing it on the M*A*S*H RPG I once picked up by accident as part of a bundle lol. Wouldn't be like, my top candidate for convincing people of really anything except "there's nothing so absurd you can't make an RPG out of it!".
Reverting to d100-based tables to simulate what you could just do with 2d6 way faster and easier is magnificently "1985 in gaming" though.