The big advantage of D20 is that it's easy to design for. It has a linear distribution where each step creates an equal, exactly 5% increment to your chance of success, so you can design a D20 game with very little advanced game design knowledge and it'll work.

This means that the world is flooded with D20 content.

Of course, in a post-5E universe, all this goes out the window. The D&D Advantage mechanic works by squaring the probability of failure (your chance of success is always a number between 0 and 1, and when you square a non-negative number smaller than 1, you get an even smaller number, so squaring the chance of failure increases your chance of success.)

You'll note this is an exponential mechanic, not a linear one.

The future of RPG game design is logarithmic and exponential, not linear. This suits players well because most of your senses and intuitions secretly work on logarithmic scales, and logarithmic scales handle extreme power disparities much better than linear ones. If you think that RPG mechanics don't scale well, that's because the linear systems descended from games in the 70s to 2010s don't; we have been banging up against the limits of linear game design for some time. However, this is also the death of the complete amateur game designer. Logarithmic and exponential game design is no longer stuff any old highschool graduate can do in their heads.

That's funny that you mention this... as my nascent "house system" I'm working on is exactly this. Logarithmic scaling for the purposes of heavy lifting for a variety of genres.

The settings will dictate those genre parameters specific for the rules.

(sorry I'm short on replies, been super busy - you guys made a lot of good posts I wanna respond to: Steve, Pulp, and migo specifically). I'll get back here asap.

Quote from: jhkim on November 28, 2023, 05:44:34 PM
I understand you're a fan of Savage Worlds, and I've been liking it too in my recent experiment with it. However, I suspect that Savage Worlds Pathfinder will still be a niche within a niche. The appeal of Pathfinder isn't in the setting of Golarion, but in the rules set and modules.

Savage World Pathfinder does actually borrow some Pathfinderesque mechanics, which are D&D-isms.  A big example is, it has a rough equivalent of classes from its Edges.  As someone who won't touch a classless system, that's a pretty big deal, and that's purely mechanical.

Quote from: tenbones on December 04, 2023, 01:48:01 PM
That's funny that you mention this... as my nascent "house system" I'm working on is exactly this. Logarithmic scaling for the purposes of heavy lifting for a variety of genres.

The settings will dictate those genre parameters specific for the rules.

(sorry I'm short on replies, been super busy - you guys made a lot of good posts I wanna respond to: Steve, Pulp, and migo specifically). I'll get back here asap.

Not to be pedantic, but are you sure you mean logarithmic?

Logarithmic growth increases rapidly at the low end, but then slows towards (but never quite reaches) 0 the higher it goes... i.e. there's a bigger difference between 1 and 2 than there is between 2 and 3.

In a logarithmic growth a stat of 2 might be twice that of 1, but a stat of 10 might only be 1% greater than a 9. That seams less than useful.

By contrast, for handling large scales I'd think you'd want an exponential curve where it starts slow (the difference between 1 and 2 might be 10%) but then increase towards infinite (a 10 might be 100 times the value of a 9).

Or alternately just a straight doubling (2 is twice 1, 3 is twice 2, etc.).

Quote from: Chris24601 on December 05, 2023, 01:37:10 PM

Quote from: tenbones on December 04, 2023, 01:48:01 PM
That's funny that you mention this... as my nascent "house system" I'm working on is exactly this. Logarithmic scaling for the purposes of heavy lifting for a variety of genres.

The settings will dictate those genre parameters specific for the rules.

(sorry I'm short on replies, been super busy - you guys made a lot of good posts I wanna respond to: Steve, Pulp, and migo specifically). I'll get back here asap.

Not to be pedantic, but are you sure you mean logarithmic?

Logarithmic growth increases rapidly at the low end, but then slows towards (but never quite reaches) 0 the higher it goes... i.e. there's a bigger difference between 1 and 2 than there is between 2 and 3.

In a logarithmic growth a stat of 2 might be twice that of 1, but a stat of 10 might only be 1% greater than a 9. That seams less than useful.

By contrast, for handling large scales I'd think you'd want an exponential curve where it starts slow (the difference between 1 and 2 might be 10%) but then increase towards infinite (a 10 might be 100 times the value of a 9).

Or alternately just a straight doubling (2 is twice 1, 3 is twice 2, etc.).

Not tenbones, but I did introduce the word "logarithmic" to this conversation.

There's more to RPGs than character advancement, but it's worth noting that classic Skinner Box advancement is, in fact, logarithmic exactly as you described. It's just that instead of decreasing the amount you improve when you level up (which would feel bad to most players and most systems don't do particularly well) the system itself slows how quickly the player character improves. This is verbatim why most RPG systems with character advancement have you level up at a slowly decreasing rate.

But the real kicker IMO is setting check difficulty, because difficulty tends to work better if it is logarithmic, especially for bell curve systems.

At the end of the day, I think that the exponential vs logarithmic discussion is an interesting application of relative viewpoints. Most of the time RPGs have two subsystems interfacing onto a single mechanic. The core mechanic is usually a dice mechanic set opposite to a difficulty setting mechanic. As such, whether or not you describe a mechanic as exponential or linear can become a point of view issue. The mathematical formulae are inverses, so if you are looking at the difficulty mechanic from the point of view of the dice mechanic, it's logarithmic. If you are looking at the die mechanic from the point of view of the difficulty mechanic, it's exponential.

I tend to use the word logarithmic to describe this because the die mechanic tends to not change much, but the GM is constantly changing the difficulty. The GM will constantly need to look from the dice mechanic and solve for difficulty, so the logarithmic function is  more relevant to the game.

That said, I don't think that the RPG scene is developed enough that most of this math stuff matters beyond raw concepts. At least not yet, and perhaps not ever. I have never actually solved a log function to set a difficulty mechanic because precision just isn't that important for RPGs and game feel is. However, if you don't understand the rough shapes of these functions, the mechanics are quite likely to fail when actually put to high level play.

I've thought this exact line of reasoning myself.  Does OGL have to be based on D&D style rules? 

My thinking it does.  Yes, it very much does. 

Specifically, the classic six attributes of Str, Dex, Con, Int, Wiz, & Cha (usually these are BS as it is the modifier a high or low attribute is what REALLY matters, but that's another argument), armor class, and ever increasing hit points as you level.  After that it's just a variation on how leveling up works. 

D&D copycats is another term we can use.  Or D&D heartbreaker, since it breaks your heart you wrote a better game, and nobody even knows it exists, let alone buys it.  One of my favorite OSR games, Olde Swords Reign, is a heartbreaker.  It used to be owned by the original writer and sold for decent price tag, but the heartbreak set in.  Now it's owned by the guy's friend and playtester who gives it away for free. 

I would also argue that another favorite of mine is somewhat like the OP commented.  I consider Dungeons & Delvers Dice Pool Edition to be an OSR game.  It's got the same races, classes, and uses armor class to hit people.  But.  And it's a big, Kardashian sized but, it doesn't use the d20.  It's replaced the attributes and skills with dice sizes.  Weakest is a D4, best is the D12.  Roll one for an attribute, one for a skill, and probably a few more because of race and class talents, pick the best two, and that's your dice roll. 

I love it as it fixes most of what I don't like about D&D, such as rolling a "bad" character is impossible, hit points don't inflate out of control, and classes, especially fighting types are just as customizable as a wizard. 

Quote from: PulpHerb on November 30, 2023, 11:13:01 AM
...
HO is 1:87.1. What kind of scale ratio is that. It makes sense neither in Imperial or Metric measurements. It replaced O, 1:48 in the US (O gage in the UK is a bit larger for the same reason as OO vs. HO). Meanwhile S scale: 1:64 or 3/8 inch to the inch languishes. TT at 1:120 or 1/10 inch to 1 inch is dead in the US (although being revived in the UK and popular in Eastern Europe).

Looks over at all my new Tillig TT stuff...

Liar! My single anecdotal experience says modeling European TT in the US is alive and healthy!!!

LOL no; it's dead – I'm a lone voice in the wilderness on this one...

Tillig or Piko will need start making US prototypes if TT is going to get any traction here. Because US model rail roaders in general are tied to an 'Operations Uber Alles' mindset that is very different from the European side of the hobby. I've seen abject refusal to model anything but US trains...

Oh, and yeah, HO is the biggest because it established massive market share earliest.

Quote from: tenbones on November 30, 2023, 03:36:24 AM
...I'm wondering what is the attraction to d20 as a system other than familiarity?

It's not just familiarity. Along with that familiarity comes a massive network effect.

With the exception of the 4e debacle, D&D has ridden a "good enough" system to massive success due to that familiarity/network effect.

Quote from: tenbones on November 30, 2023, 03:36:24 AM
While I confess I do possess the strain of hubris that believes I can make a better d20 mousetrap... I always feel this energy could/should take it into better abstractions that make it qualitatively less d20. There is *definitely* a sweet spot where d20 sings, no question.

I suspect the following: I like running big big sweeping sandbox campaigns, where PC's start wherever we decide (normal people, exceptional people, whatever) but I want them to have unlimited scaling potential to grow within the constraints of the setting, *not* the rules. d20 *really* starts to strain post 10th level. Which is FINE. There is scant support for GM's to learn how to run games post-10th, much less on a chassis that was never intended to go beyond 10th. But I maintain it *could* if it were re-designed correctly.
...

I have always wondered why no one has designed for that lvl3-8 "sweet spot" and just extend it over 20 levels...  But, no the  "level based system must go from zero to superhero." paradigm continues to dominate unchallenged.

The Basic d20 system is really just Stat Mod + Skill or attack mod + d20. Which is No different from most skill based games...

In my opinion: There is a big design space for fixed HP d20 games:

Six ability scores, -5 to +5
HP - fixed.
Saves
Skill list: 1-8 scores
AC =  Ac could be divided like Conan d20 into Defense and Evasion. Damage reduction for armor could be introduced as well like in the d20 Conan RPG.

Bam, done: You have a "d20" RPG that functions like any other skill-based game. "Level" is just an advancement method – or even just get rid of levels and have the PC's buy advances with XP, like any other skill based system.

You could virtually recreate games like CP2020 with such a scheme relatively easily. And it would still be largely d20/5e "compatible"... i.e. familiar enough that it does not seem like a "new" system to players coming from D&D.


Once OSR/d20 designers break themselves of the "HP must go up with level" mindset that they all seem to be locked into; a whole new d20 design paradigm will open up.

But I'm not holding my breath on this one. For some reason the OSR/d20 design sphere seems locked into the "HP must go up with level." it must. No exceptions. Like, Ever.

Quote from: Jaeger on December 07, 2023, 08:15:34 PM
I have always wondered why no one has designed for that lvl3-8 "sweet spot" and just extend it over 20 levels...  But, no the  "level based system must go from zero to superhero." paradigm continues to dominate unchallenged.
...

Once OSR/d20 designers break themselves of the "HP must go up with level" mindset that they all seem to be locked into; a whole new d20 design paradigm will open up.

But I'm not holding my breath on this one. For some reason the OSR/d20 design sphere seems locked into the "HP must go up with level." it must. No exceptions. Like, Ever.

Because that's not as popular as the other way.  That's why I built my own--I never seem to fully enjoy the more popular options.  Though in fairness, I'm not really in the OSR sphere, either.

My own system is built around something very close to what you are saying here.  I've got 24 levels, and levels from about 4-17 are in the sweet spot.  So not exactly your idea, but a lot closer to it than most d20 games.  My equivalent of hit points are much scaled back, and don't go up every level (or only go up a point or two on some levels).   Though my levels 1-3 are introductory, and the higher ones are just to handle a few edge cases.  The intent is that people wanting sweet spot just start at level 4 and go.

The other thing though, is you can't just do the "ability mods from -5 to +5" thing and make a few other changes to really make that work.  The whole game needs to be designed around the plan of the bulk of the game is incremental advancement in the sweet spot. 

The problem with true skills-based games is that there are no "gated" abilities other than what the skill costs put on them.  Or what the group agrees to do or GM imposes, of course. The problem with class-based games is being locked into a silo.  Tacking skills onto classes has its own set of problems.  What I'm trying (jury still out if it works well or not) is to build a mostly skills-based game that only uses its classes for some limited gating and to handle the advancement sequence.

There's nothing inherently wrong with having a d20 mechanic and building a game around that.  There are a lot of bad ways to do things, and WotC has managed to explore quite a few of them.   

Quote from: weirdguy564 on December 06, 2023, 07:48:53 PM
Does OGL have to be based on D&D style rules? 

OGL, the Open Gaming License, cover a lot more than just the d20 system.  One very notable system that is covered under the OGL 1.0a is West End Games D6 system of Star Wars RPG fame.

If you mean OSR, then the above doesn't apply but I still disagree.  However, in order for it to be OSR it would need to be backward compatible to the old TSR D&D.  If you're making a new rules set (you can) then you should still be able to use all the modules without any conversion.  If you're making a module, then you should be able to run it with AD&D 1e or BECMI.

Quote from: weirdguy564 on December 06, 2023, 07:48:53 PM
I've thought this exact line of reasoning myself.  Does OGL have to be based on D&D style rules? 

My thinking it does.  Yes, it very much does. 

This has already been answered over a decade ago. It does not. FATE, for instance, is OGL.