I'm trying to develop a universal system for a number of traditional "loot-and-level" RPG's I'm working on.  I'm trying to determine how to make a mechanic that's friendly to swords, robots, magic...whatever.  It needn't be blindly interchangeable, but similarly structured. 

Ideally, I would like to lay a foundation that bases this universal mechanic off of similar features present in all of my "traditional" RPG's.  However, I've never built a mechanic from the ground up before.  I'm taking small steps with this one, but it's something I'd like to complete by the end of the year so that I canstart making large-scale games, once I've established some familiarity by publishing smaller, less resource-intensive games.

To make "loot and level" easy, you want the mechanics of each level to "look like" previous levels.  This reduces your game mechanics balancing problem.

Ie, you want your system to be affine in some space, and have character-level transformations be "cancelable" out.  A "level 2" character against a "level 3" monster should work out to be similar to a "level 20" character against one of a "level 21" or "level 30" monster (you have to make that design choice).

Variations from this core should be justified.

Your characters should "seem" to get more powerful.  Some numeric values should keep going up, and gaining a "level" should never make you worse at old challenges.

Creation of the "base" character should be simple and quick, and the complexity of advancement should be distributed over the character's lifespan.  Ideally starting characters shouldn't have to be optimized for later character options -- ie, a choice made at character progression should tend not to cause massive discontinuities in later character advancement options.  This makes "sit down and play" work better.

Loot and level games are all about character progression and the gameplay.  Possibly you will want to integrate the character progression mechanics with the gameplay mechanics.

Or you could try a completely different approach.  :)

Yakk's approach is solid from a certain perspective, but I would argue that games such as what he describes are very largely being dominated by MMORPGs where the loot and level paradigm has generated an entire arcane vocabulary of its own.  I would thus make the following 3 points:

1) is the Loot and Level itch being sufficiently scratched by MMORPGs already, so that a TTRPG would be better suited to NOT try to mimic that model?

2) Its becoming increasingly common for MMORPGs to seek some alternative to the Grind...viewing the Grind either as a necessary evil to get through, or something to be abandoned.

3) IMO I'd look is to cater to what the particular strengths of TTRPGs are that you can use in your design.

So, applying those ideas to the idea of a Loot and Level game, I'd come to the opposite conclusion.  Instead of each level looking like the previous levels but with higher numbers...investigate having each level radically change your character, its capabilities, and the way he plays in some transformative capacity.

You could even break it down into substeps, where the small levels are standard power ups while the BIG levels are transformative.  D20 kinda sorta tried this with Prestige Classes, but ultimately IMO Prestige Classes never fulfilled their potential and wound up just being a way to reintroduce 2ed kits.

Consider an example path:

First Big Level:  you're weak, you're skulking in the shadows, hoping to vulture some easy prey, swipe the low hanging fruit and bolt.  All the game mechanics focus on the fragility of the character vs. the environment and the choice between fight and flight with flight often being the option that keeps you alive.

Second Big Level:  you're strong, you're now striding through the dungeons scaring off the weaker denizens and engaging and defeating denizens who before would crush you.  All the game mechanics now focus on you as a hero defeating your opponents.

Third Big Level:  you're now a leader, you're success is legendary and people look to you to guide them.  You now have several henchmen and followers...more than just a couple, and are someone who is recognized in the broader setting.  All the game mechanics now focus on small unit tactics with you as the leader of the Warband.

Fourth Big Level:  you're someone of influence and importance in the world.  You almost never go delving into dungeons on your own any more, except when after the biggest prizes.  Now, you send others on the delve to do your work for you.  The mechanics have now moved away from the dungeon dive itself (unless players want to play it out with secondary characters) and into the machinations and politics of the broader world.  Much like games where the player is the owner of a chariot team that must be outfitted for successfull races, the player now outfits and sends out his teams of underlings with their success feeding into his source of income.

Fifth Big Level:  you're Mr. Big shot.  The overthrower of empires, the power behind the throne, the king yourself.  You no longer are outfitting adventuring bands, now you're outfitting armies.  The mechanics of the game become much more strategic in nature.

That's just one type of path...another path could be literally transformations by acquiring bestial or demonic powers that reinvent your characters capabilities rather than reinvent the playing field.

The point being, that one thing that TTRPGs can do well is completely cater to what's needed at the table.  A GM scenario created especially for their gaming group is the ultimate "instance".  MMORPGs are desperately trying to make "instances" more like actual role playing...with a TTRPG you're already there.  So consider that instead of "leveling up" being just another power grind...that it actually completely changes the way the player interfaces with the game.  I've seen some games kinda sorta do that (d20 feats kinds sorta due...and the introduction of ever more powerful spells kinda sorted did for older editions of D&D) but I've yet to see one take that as a core design goal and really run with it.

That would really challenge the way people think about "level up" games.

The Mentzer boxed sets for D&D (Basic/Expert/Companion/Master/Immortal) actually pushed in the direction Ralph is describing, with an assumption of different adventure types and character status in the different level ranges. But it was mostly reinforced at the content level rather than mechanically.

I think Greg Stolze's impending REIGN is also going to push in that direction.

Trollbabe has some VERY useful material for this traditional problem even though it's not a traditional game.

That said, if it's going to be universal, you want to think of the setting abstractly, and the role of the characters in it similarly abstractly. So the meta-levels would involve

(a) your locus of action - is it just your body and words and magic exerting local influence, or do they go farther out into the setting through agents?

(b) how much you can influence - local change, broad change, setting-wide change. Other PCs? The gods or laws of physics?

Traditional games tend to keep (a) constant at the individual PC while gradually creeping up (b) through powers and the GMs imaginative contribution. Working with (b) also seems like fertile ground.

It's ironic the systems that were mentioned in the insight I received.

I use the term "loot and level" for mechanics that advance a character through combat or some other contest.  Whether the player is rolling against a goblin shaman or a tightly locked door, the outcome contributes to the character's advancement.  The reward system for these games is usually stuff that makes future contests easier.

On a tabletop, this is manageable...on a server, it's a formula for a vicious circle called "the grind" that usually ends in tears...or boredom.

Since my background is in programming (I make games for the table too...I'm an author...I like books...don't hit me) I tend to work outwrd from the mechanic, looking at  what gameplay I want the system to support...like a framework.

Actually, the mention of the grind and the shortcomings of MMO's have given me a push in a different direction.  So far, I've created a combat system (a kind of "sliding round" system where each player keeps count of their own turn in the action que) and a "class" system based upon social interractions.

Sadly, it might be too complex for the tabletop (although I've been playtesting the combat system with dice...it works very well with two players commanding multiple units...it turns somewhat raucous with multiple players controlling individual characters as "challenges" start to stack) but, it would work very well with the aforementioned MMO. 

The class system especially, that makes no assumptions of motivation, and instead deals entirely with how others respond to you based upon your class rank in that society.  A fighter is nearly a king among the barbaric tribes while magic-users are treated like outcasts.  Conversely, a paladin or other noble class enjoys a position of status among civilized nations and works at a disadvantage when camping beneath the stars or in the slums of their own city.

Surprisingly enough, these non-motivational class systems haven't hit any snags.  If a player wants their character to be evil, they're evil.  If a player wants their character to be good, they're good.  However, the way they are perceived is the way they are perceived regardless of their actual deeds.

Try and play a dark-elf outcast ranger in this system and you're got a real uphill battle on your hands.

I'll start putting things up in playtesting (along with a summation of the system thus far) in a couple days...wish me luck.

Hi Scott (right name to use?),

Surprisingly enough, these non-motivational class systems haven't hit any snags.  If a player wants their character to be evil, they're evil.  If a player wants their character to be good, they're good.  However, the way they are perceived is the way they are perceived regardless of their actual deeds.

Well, that's a game that this system could support since the system, in it's infancy, does not push the character towards the next level, but rather towards expanding their experiences.  Instead of experience "points", added up and counted towards an incremental increase in power.  This current system takes experiences and weights them, each experience gaining the character a talent.

While it works well with traditional hack and slash (learn swordplay from master...get better with a sword), it also works well with lots of different roleplay.  Escaping a prison with a cellmate grants an affinity with that person and gains the character a reputation among his social circle.  When run through a computer program, a player can team up with a reputable rogue and go to that rogue's hometown to discover that their reputation has proceeded them.

It's even easier to track.  A certain amount of hours of wearing plate mail calculated with the difficulty of the tasks performed in them is madness compared to listing events and their impact on the gameworld, the group you're adventuring with, etc.

I'm liking it so far.  It's complex as all hell, but fun.  It also promises to offer grind-free gaming.