"Yes, 4E kept you from making stupid decisions in character design because it kept you from making substantial decisions in character design.". - Jonathan Tweet here:

https://www.reddit.com/r/IAmA/comments/3oq1iy/im_rpg_designer_jonathan_tweet_lead_designer_on/?limit=500

Which I thought was interesting and a topic I haven't really tackled before. I think the amount of variation between characters is something largely independent of the core resolution system or most of the basic design design decisions involved in building a game, like class-based vs. skill-based or attribute-based vs. skill-based task resolution. (a couple of mechanics, dicepool or multidie-additive, result in some major divergence in roll results, such as attack rolls, but even that might be controlled for in various ways).

4E is a pretty good example of a game where there's a low amount of divergence between characters - there's a fair degree of similarity between characters in terms of - HP, defenses, attack bonuses, power recovery schedules, # of powers known, damage output - with specific options being designed to have similar outcomes (e.g. dual-wielding vs. "great weapons", or melee vs. magic). 5E, while not quite as extreme, still ends up with a fair degree of similarity in terms of final ACs, HPs, etc.
Both late-4E (Essentials) and 5E attempt to add diversity in terms of character complexity, with simple and complex options, though the intent is for numbers to remain fairly similar regardless.


Some other games with substantial similarity between characters, IMHO, would be Over The Edge (another Tweet game, maybe ironically), Hi/Lo Heroes.

Whereas, other games may have huge divergences in HPs, attacks/round, attack accuracy, etc.
This is sometimes balanced, sometimes not. Something like Rifts has dramatic differences between characters in terms of # attacks, hit points, and so on. Its also interesting in having considerable strategic weaknesses (the Glitter Boy can't fit into the dungeon; the cyborg can't naturally recover lost HP; major factions in the game hate your Mind Melter; the Juicer can't be touched in combat, but dies in seven years game time). But generally Rifts is also the poster child for an unbalanced game.


However, divergence need not necessarily be crazily-unbalanced as long as some combinations that synergize particularly well can be avoided - a game can be fair without everyone having exactly equal to-hit scores, if a low to-hit is balanced with more damage, then characters vary in effectiveness between low-AC/high HP and high-AC/low HP foes, but would average out overall.

Point-based chargen particularly struggles with this, but it might work in a few ways:
*  Simple attribute-driven games like T&T can code 'options' directly into X-vs-Y choices (more CON means linearly more HPs, more POW linearly boosts spell points, more CHA instead gives more social ability). A more complex game might have that + add a layer of class choices that can muddy the situation substantially. TFT shows some an example here in that there are good trade-offs between DEX and STR e.g. in missile weapons - a character with higher DEX can fire more often, but will have lower STR and so would have to use a smaller, less-damaging weapon.

*pre-built archetype systems can be designed to give balanced arrays of abilities somewhat handily e.g. Feng Shui does this a bit (the Big Bruiser has tremendous resistance to damage, but struggles extremely to hit). I haven't played this much, but it did look like there was some attempt to keep abilities that synergized excessively under control.

i just wana say this is very cool and helpful

Thanks!

You need a gold medal for this thread.

:cool: Thanks Snowman!



***

(more recent thread edits: rewrote 'controlling bonus and penalty accumulation' post. Complexity post moved from pg. 17 to pg. 15 [to go with game evolution/design processes [e.g. hybridization]). Assorted minor edits e.g. ranged combat, balancing powerful races, # actions, skill-check based initiative, ranged bursts (combat moves), T&T poison note (damage); Hollow Earth Expedition notes - damage, cutting down excess rolling, classes, Dungeon World initiative 24/11- Maelstrom notes (social, attributes, hit points, damage)
5 Dec - 'step-dice' renamed to changing-dice-type, post split into sections by subtypes; assorted Deadlands references, more notes on when to define something as an 'advantage' vs. an 'attribute'. 17 Dec - world of dungeons note, 3d12 checks against 2 stats, subdivided 'effect' category for action spending. 18 Dec - minor notes on character conversion, DW 'volley' and ammo. 21 Dec - cards extra notes, advantage/disadvantage order, minor edits [chargen posts]. 22nd - to add Martial Arts A-E (FASERIP)?, added Humandyne XdX and G-core notes + Dodge notes; 23rd - minor initiative, disadvantage, str-req edits. 24th - 'philosophy of combat' (end of combat chapter). 25th - extra 'fudgefactor' site notes - damage, initiative, conversions
1 Jan 2016 [Happy New Year...] - fireborn notes - actions and combat moves. 20ish Jan - minor notes on automatic actions, attribute improvement, count-success damage systems.
23 Jan- new topic added (core mechanic and combat manuevers) here; movement edits

Any new additions I should know about?

Hi Daddy Warpig, and thanks!
With 5E D&D out there's the odd tweak pertaining to that. Post #42 is new actually (within this week); this being some speculating on how combat manuever systems might be affected by the core mechanics of games. Other than that, quite a few little tweaks here and there. I combed through a couple of hundred pages of old rpg.net threads at one point to see what I could find that I'd missed (I'm mostly up to date with here) and found a couple of interesting things - Humandyne XdX and Fireborn, both dice pool type things with the latter also getting some notes in the combat manuevers section.

Most significant minor footnotes I guess, than I can currently remember, might be: an extra way of doing armour in the armour section (...I found a game, Supers!, which basically treats armour like a parry roll...), notes in Skills on how attribute modifiers may be set up to benefit either low- or high-rated skills, in post #10 some discussion on one-value vs. two-value systems (e.g. stat-based purely vs. stat-+-skill), and in #11 ("General intro on resolution Mechanics") some more in-depth contemplation of the pros/cons of various mechanics.
Any new additions are up for debate of course, if people want to do so, cheers...

I myself am doing work on combat maneuvers and how Attributes affect skills, for my own little action movie RPG.

I look forward to any feedback you might have, once I am free enough to post again.

NP can do.

Recent minor edits: 31/1/16 recovery actions, independent rolling for effect (minor notes). 20/02/16 - 'Momentum' (safety valves). Minor notes in damage (SW vs. Forgotten Futures), skills (FC vs. 3E untrained limits), alignment languages, list of 2E NWPs. 22/02-Barbarians of the Aftermath note. 25/02 Streetfighter note (combat moves), DC Adventures note (Safety Valves); 26/2 notes on 'decision points' in character building (bottom page 1)

Here's a short URL that's easy to remember, for people who want to find this thread:

http://is.gd/gamedesign

Also, you should strongly consider compiling this, editing it a bit, and printing it as an ebook. It'd be a handy guide.

:hmm:
I'd started once before, only to stop because at that point the data in the thread was changing too rapidly still. But it has settled down a bit now, so, OK, I'll do it...
Thanks again Daddy Warpig.

Ok, update on compiling this...
Currently the thread is compiled as a Word document here, in google docs.


https://docs.google.com/document/d/1DsZzHaK5a-45Rp8SKh4lPVcdvVGuXoKo2NDZiiL9CZ0/pub

This is reordered somewhat (sections are generally where I want them) and it trims out extra conversations and such. Formatting is however extremely shoddy at this point, sorry - and fixing this is painful since the whole thing is 350+ pages, so further changes will probably be awhile.
Thanks again to contributors. If my limited familiarity with google docs is correct, people should be able to copy from here (and do feel free to do so if you want to send things to your friends or print out or whatever) but not edit the original. Let me know if this doesn't work...

Quote from: Bloody Stupid Johnson;889143Ok, update on compiling this...
Currently the thread is compiled as a Word document here, in google docs.

Great stuff man.  Skimmed through the mechanics section and did not see a reference to the dice pool (count success) mechanic used in Atomic Highway.

Come to love this approach and modified my whole home system to a similar one.  I think I saw it in your original summary, this dice pool approach has very much the feel of Chainmail and in particular heroes, etc. getting more dice.  It has also solved the problem I've had with all other systems in the last 30+ years of gaming, how to make "low level" characters mean anything when along with "high level" characters.  This dice pool approach also has solved some perennial issues I've had with multiple attacks, critical hits and misses, combining movement with combat, defensive moves and shields, barter, etc., etc., etc.  

I can comment on more specifics on the mechanic if not already in your summary.

Quote from: Xanther;900428Great stuff man.  Skimmed through the mechanics section and did not see a reference to the dice pool (count success) mechanic used in Atomic Highway.

Come to love this approach and modified my whole home system to a similar one.  I think I saw it in your original summary, this dice pool approach has very much the feel of Chainmail and in particular heroes, etc. getting more dice.  It has also solved the problem I've had with all other systems in the last 30+ years of gaming, how to make "low level" characters mean anything when along with "high level" characters.  This dice pool approach also has solved some perennial issues I've had with multiple attacks, critical hits and misses, combining movement with combat, defensive moves and shields, barter, etc., etc., etc.  

I can comment on more specifics on the mechanic if not already in your summary.

Thanks! I've downloaded Atomic Highway and am having a lookthrough, as time permits currently. Good catch - I haven't seen its exact system (roll dice equal to attribute, add bonus points from skill to individual dice rolls) before.
I'll go back and add notes here. if you want to comment on how the things you mentioned work(multiple attacks, crits, barter, etc.) go ahead by all means and I can integrate it in earlier with attribution.

Solo Adventure Gamebook Design
First actual topic update in awhile. This topic is a little bit antiquated, given that the medium of gamebooks has been replaced largely by computers which can do the crunching involved more easily with better graphics (a few a few Fighting Fantasy gamebooks I believe later made it as phone apps).  Compared to computer games, the main advantage of a gamebook is probably the cheating potential as they depend on the honour system; if you disagree with a book's author as to a resolution, or you want to port your character that did Scorpion Swamp and has a Luck of 14 over to Warlock of Firetop Mountain, no one can stop you but you.  
Its also this aspect that should make GMs leery of solo-generated and -run characters being brought into their campaigns, though in some cases, like the Tunnels and Trolls [T&T] gamebooks, the existing adventures could be used to 'pregenerate' a high-level character that's more organic that what could be constructed via instant creation. (I've tried using this for NPC creation, then converting to the game system I was using; something that largely turned out to not be time-effective given how quickly they were squashed by PCs).
Potentially a solo adventure can be run by a GM as if it were a pre-made GM adventure; the main drawback of this being that players expect more options than may be provided for in the text, such as ability to retrace their steps, forcing the GM to search for appropriate references or 'wing it' and go off the grid. A couple of FF adventures have again been converted into GM'd modules (e.g. Myriador's Caverns of the Snow Witch) with indifferent success in this case IMHO since the basic adventure is fundamentally pretty railroady. (My attempt to run the Myriador Caverns of the Snow Witch adventure involved a centaur PC having to ride a pegasus...). The higher lethality of solos can also come as a shock to players.
A solo adventure can sometimes be included as a training accessory in a normal tabletop RPG as, as in the case of the D&D Red Box or DC Heroes 3E (which has a very short adventure where you're Batman).
Linking the topic back to this thread: Rarely in the past, I've also used solos as a weird sort of 'alpha testing' for my own homebrewed RPGs; by building a character in your new system and then running them through a solo adventure (converting the oncoming monsters, difficulty numbers, NPCs etc. as necessary), gives a very rough way to see basically how the game shakes out before wasting anyone else's time with it.

Game Mechanics for Solo Adventures
In the simplest case ["Choose Your Own Adventure"] there may be no randomization and you win by picking the correct path, but most adventure gamebooks have streamlined RPG rules. Compared to a full tabletop RPG:
*simpler rules generally, being often aimed at younger people with little gaming experience.
*d6s are often preferred, since these are the only dice likely to be owned by non-gamers. An exception was the 'Lone Wolf' book series, which used 1-10 and let a player poke a page randomly to generate the number. I think another series may have printed dice rolls at the top of the page to allow random page turning to generate numbers. Legends of Skyfall used a coin-based system instead.
*there's no social component; player speech can't be considered in rules outputs, so questions of diplomacy/seduction/etc. are entirely system-based; make a Charisma check or whatever. At best gender or race may be considered on social checks, but a given page may not have so many options.
*players can be subjected to riddles or puzzles. These can be somewhat cheat-proofed if a puzzle answer is itself a number i.e. a paragraph reference; the player can't turn to the appropriate page unless they solve the riddle.
*limited world consistency: like a computer game, scenarios are usually assumed to be 'replayable' with different characters e.g. everyone can save the doomed city from the demon or marry the princess.
*there may be fairly little player investment. Lethality can be quite high, meaning that characters essentially work as disposable probes sent in to find the correct route.  Usually the simpler the game system, the more lethal. Specific items may be needed to complete an adventure.
*gamebooks sometimes provide random chargen, sometimes some choices with points/special skills [rare: I think there's a Robin Hood [TV series] gamebook that does that], or they may be designed for a specific character, in a way that is extremely rare in prepackaged GM adventures.  Gamebooks often accomodate just one character at once, so if they are there 'classes' are designed to provide differences in play but won't need to work in a group - there might be e.g. thief special abilities which bypass some dangers vs. being a barbarian with an extra health level [Duelmaster].
*Gamebooks can include options which are OK in the context of a one-shot but would probably break an ongoing game, like FF's Potions of Fortune adding 1 to Luck when drunk. Grailquest #1 gives out a magic duck that improves die rolls which would generate infinite gold combined with the transforming clay in Amulet of the Salkti (don't ask how I know this).

*single-adventure gamebooks have game mechanics written for that specific adventure, and/or a specific character e.g.
-SHAME or HONOUR scores for the PC Theseus in Cretan Chronicles
-'refusal to use magic' rolls for the PC Fire*Wolf in the Demonspawn series.  
-the spell list in Sorcery! is specifically geared to the situations to be encountered there (it also includes an interesting player-skill component, in that a player can't check the 'spell book' in play, and must remember the various material components for spells).
-Blood Sword gives out dual-sword-wielding to Warrior characters at the end of Book 4; this is probably mostly so a character can use the Blood Sword and Sword of Life together in Book 5 (the second-best way to defeat the Magi, short of just killing yourself, no really).
-Citadel of Chaos gives magical powers for free to the PC of that book, at no loss of Skill etc; the later tabletop book (just called 'Fighting Fantasy') uses the same magic rules there, but also gives wizards a Skill reduction.
Most gamebooks will include some way of recovering lost HPs and the like, which could simply be rest (for e.g. superhero adventures in a specific city across multiple days), food (FF Provisions), healing spells, or first aid. 'luck points' can also buffer HP losses. An especially railroady solo can have HP and damage quite closely calculated to have an adventure be dangerous without being necessarily fatal; in a generic system that might include special healing spots and the like.

Gamebooks usually have low replayability once the options are known. Often there's little inherent reason to replay the same scenario; the exception again is Tunnels and Trolls, where an adventures' treasure and other rewards add a profit-motive to doing them again with new characters.  A book can add some variation by adding randomization of what's given out (e.g. random monsters, random treasures; something often used in normal GM-run prepackaged adventures too but IMHO mostly unnecessary in that case).  'Sea of Death' for T&T is a pinnacle here, with random adventure paths that are extremely varied.
A 'patch' suggested in some of the early T&T solos was to not make player decisions on e.g. going left/right, instead leaving this to random roll (random decisions was something also brought in in FF for 'Creature of Havoc', something the character hopefully got better at as they ceased to be a 'creature'. The book is better known, though, for having a typo rendering it unplayable).
A couple of series (e.g. Blood Sword, Duelmaster IIRC) try to slow down ongoing learning by use of 'codewords'. e.g. instead of asking 'do you have the potion of Frost resistance from the wise woman?' the book would ask you to note the codeword FROSTY if you got the potion, and then ask you if you have this at the appropriate point (or would ask you to delete the codeword if all your potions were demagicked by the magic-sucking anteaters, etc.).
Adventures that are meant to be especially hard in one go can include surprise 'pixel-bitching' options - one of the FF books has an option where if you attempt to eat something, you find a key in it.

Other Elaborations
*books may have sequels, allowing transfer of characters between them in order - e.g. the "Sorcery!" series for Fighting Fantasy, Blood Sword. T&T allows different adventures in whatever order, though sometimes with level, race or class ('type') restrictions; a couple can be played repeatedly (Arena of Khazan, City of Terrors).
*Page numbering tricks. A paragraph can have 'hidden' extra options which players can be told about at other paragraghs. 'If you use the book of Skelos, add 20 to the current paragraph number and go there'. If this provides a nonsensical reference, then the book can't be used here.'
Books occasionally label paragraph numbers for this sort of thing, in the hope that this will be forgotten (or not forgotten, but the player feels better about it than if the extra option was explicit) e.g. the Cretan Chronicles has a * marker for 'you can do something creative here'. Book 3 has this on the first paragraph, which otherwise ends in auto-death (very harsh for players who pick this up first..). T&T 'Amulet of the Salkti' has a & for using 'the Skull' (sometimes very useful, sometimes pointless, sometimes fatal to use). The skull's need for explicit paragraphs meant it is designed to self-destruct at the end of the adventure rather than be carried into other adventures as usual as an ongoing headache, though that is avoidable. This brings up another point that treasures/ magic items in solos do have to be able to work without GM adjudication, moreso than tabletop games.
*matrices are sometimes included e.g. many T&T books have a 'Magic Matrix' where common spells can be cross-referenced to paragraphs, to get new spell effects (X damage) or where necessary a new paragraph. T&Ts Amulet of the Salkti has an 'items matrix' as well; players are told which things are 'items' and when they need to use an 'item' but not which item.
*a couple of books were intended for groups of players, including Blood Sword (group of adventurers -different classes are included to differentiate characters, and it has a level system so a group can have several lower-level or one higher-level.
The Duelmaster series involved two players, with each having a separate book and having to reference e.g. when they found each other. The books often included instructions to WAIT, which would increase chance of detection.
*options can be hidden from players (what doing X does) by having a paragraph and then a different paragraph. A specific combat example here is the epic battle at the end of the first 'Way of the Tiger gamebook; notably, the grandmaster you're fighting learns your moves quickly and can't be hit with the same move twice in succession - something you only find out by doing it since multiple paragraphs are involved. 'Arena of Khazan' for T&T let people attempt ranged or magic options first, then went to melee combat.
 *City of Terrors for T&T had 'job openings' where a player leaves their character card for later adventurers to find.  For instance, a character could become a slaver and enslave future characters going through there. Generally that would mean fighting other characters of the same player/book owner, so it could be a way of e.g. concentrating money or items from several characters (including hard to get items) onto one character.
*City of Terrors also showed some bad design (narrative principles) - in that depending on a player's choice, the world is redesigned in order to screw them over. If a characters goes on one adventure in the temple, the god is a 'false' god, while another has them be the 'true' god; the slaving options has some similiar issues (depending on a player's decision, the person encountered is from a different organization).
*A couple of T&T adventures are badly designed from a mathematical basis as well (T&T, while 'lite' by tabletop standards, is still at the 'high end' of complexity as regards gamebooks). The Corgi edition of Amulet of the Salki has 'recommended' limits on combat adds that probably mean a character will be unable to pass most of the Saving Rolls.  'Beyond the Wall of Sleep' has an Saving roll difficulty adjustment system that makes SRs generally too hard unless a character has a really low stat (in which case, barring uber magic weapons, monsters probably get them). A few others have a 'make a saving roll with a level equal to your level' mechanic which aren't necessarily fair to higher-level characters (since stats increase very slightly with level, and mostly are gained from special magic).

Structures & Adventure Formats
Solo adventures follow more or less usual adventure formats.
*Dungeon - the character can wander around. Solo dungeons are often designed to be moved through 'one way' due to this making paragraph structure easier to build; the later Fire*Wolf books among others usually let characters map areas or  record paragraph numbers. Ceiling cave-ins are popular in some of the worse gamebooks.  
*Quest - the PC (or their ship or spaceship) goes on a long voyage. With a story rationale, this tends to work well for one-way-linearity. Some solos have given separate stats for ship combat /starships for this reason.
T&Ts overland adventure 'Caravan to Tiern' is somewhat interesting in that it has a series of either/or adventures - a character can basically go either way at various 'stops' for scouting etc. - neither being necessarily a bad choice, and with both leading back to the same next decision point.
*Wilderness Sandbox - e.g. Scorpion Swamp for FF, where the character can go wherever, though this also has a choice of 3 quests the character is trying to accomplish - two being to reach definite endpoints and the third to kill three of the wizards in the swamp.  Or 'Sea of Death' as noted above. Scorpion Swamp largely works as an open-air dungeon, though it allows retracing of steps and also provides 'you have been here before' options.
*story-based adventures at one site - IIRC FF superhero book 'Appointment with FEAR' works sort of like that, with various encounters occuring in the city over a period of time. I think (this can be contrasted with City of Terrors as an example of bad structure - City of Terrors lives up to the name by working exactly like a dungeon, with street intersections having people who may try to kill you or leading to houses with likewise possibly fatal encounters.
*'teleport gate' adventures [e.g. T&Ts Deathtrap Equalizer, Beyond the Silvered Pane]; the character randomly goes suddenly with magic into various scenarios (e.g. suddenly underwater, in a magic item store, meeting or seducing dangerous weird NPCs, fighting monsters, encountering strange magical effects, etc.).  These are both plot-less and so the main motivation is character improvement. A more sophisticated version of this is FF's Dr- Who-inspired 'Spectral Stalkers', where a magical item called the 'Aleph' transports the PC through various strange worlds, partly dictated by random rolls, and a final confrontation.
Solos of most types can include a time measurement which can drive the adventure forward and stop PCs goofing off (not often seen in tabletop) - either ad hoc estimated by the adventure, or with a tracker than goes up. An adventure might either fail automatically on day X, or in some cases make the adventure harder e.g. in FFs 'Night Dragon' the final monsters Stamina increases based on time spent.

Other dungeon design notes: another note here randomly for now, something interesting but beyond the scope of the thread really (dungeon-design-related). Drgon 1980s the 'The Mansion of Mad Professor Ludlow' (reprinted in TSR's 'The Lost Adventures') is a regular (non-solo) adventure with 3 levels; to prevent confusion in which room is which, the first floor rooms use letters (e.g. room "J"), the second floor labelled in roman numerals (e.g. room "XV"), the third floor numbers (e.g. room "7").