Topic: Japanese-style computer-rpg mechanics? [ot?]
Started by: afray
Started on: 5/12/2005
Board: RPG Theory
On 5/12/2005 at 9:18am, afray wrote:
Japanese-style computer-rpg mechanics? [ot?]
First off, I realise this may be out of your area of expertise, but thought I'd ask anyway. :-)
I'm a video games developer who's trying to develop a new computer-rpg as a side project. I haven't played a lot of table-top gaming, but there's a lot of info on western table-top rpgs and their mechanics. However, this style seems more suited to the attribute/skill style.
I've played through a couple of japanese computer-rpgs and they have an interesting attribute/natural atribute approach. Disgaea, for instance, gives you the standard western rpg attributes (hp, sp[mana, special], atk, spo, res, etc.) but also a percentage "natural" rating for each attribute. When you level up, each stat increases in proportion to their natural ability. So everyone starts off pretty equal, but their natural ratings make them diverge quickly.
This is fundamentally different to most table-top games because each stat increases as you level up, eventually ending with a lvl30 warrior character typcially having around 1000hp, 800atk, etc.
It also means that your initial class choice is pretty much what you're stuck with, and a high-level theif will look like pretty much any other high-level theif. In Disgaea there's a very powerful system for creating multi-class characters over time, but I'm not really interested in that at the moment.
I'm wondering how this would be applied in combat. A straight comparison wouldn't seem to work, as in-game a few levels difference can mean 100 points of atk, but you *can* take someone higher than you.
Does anyone have any experience of combat in this type of system, or know any good web resources?
Thanks for putting up with my OT stylings!
Cheers,
Andy.
On 5/12/2005 at 3:26pm, Bankuei wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Hi Andy,
If you've decided to use a system where characters can infinitely level up, there are definite issues that can crop up. Many games choose to either cap the maximum power levels, or have areas where the game rises to match your effectiveness(such as Disgaea's Item Worlds).
Probably the -MOST- useful information you're going to find online is going to come from other videogames. I'd try taking a look at gamefaqs.com and check out some of the more indepth FAQs that often will break out the exact mathmatical formulas. A good example is this:
http://db.gamefaqs.com/console/ps2/file/final_fantasy_x_stats.txt
Chris
On 5/12/2005 at 3:43pm, afray wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
No worries about the infinite levels -- it's essentially a huge random dungeon anyway, and *very* combat-centric. It's more of an experiment than a practical game, although we can dream. :-)
That link was great, it's exactly what I needed, but you can't link directly to gamefaq faqs. You have to link to the game page, and say which faq you refer to. Found it anyway, though.
Cheers,
On 5/12/2005 at 3:50pm, Bankuei wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Oh yeah,
And, for random dungeon love, do check out Nethack:
http://www.nethack.org/
Chris
On 5/12/2005 at 4:40pm, Mike Holmes wrote:
Re: Japanese-style computer-rpg mechanics? [ot?]
afray wrote: I'm wondering how this would be applied in combat. A straight comparison wouldn't seem to work, as in-game a few levels difference can mean 100 points of atk, but you *can* take someone higher than you.
Does anyone have any experience of combat in this type of system, or know any good web resources?
I think if you restated the question it would help. For example, with regards to the question about combat (it is a question, isn't it?), do you mean that being able to take somebody with 100 points of atk difference is something that a system like Disgaea does, or is it a design goal that you have?
And what sort of comments are you looking for from people who may have experience with this sort of system? How to implement it? How to limit it? How to make it work in combat?
What are your design goals that you're trying to get with such a system?
Then, I'd add, have you considered something more interesting from tabletop? Or, rather, from a TT player's perspective like mine, I've always wondered why CRPGs don't have more complicated systems than they do. It's a strange irony that CRPG systems are often far, far simpler than TT systems (except perhaps in the execution of the math), when computers could handle much, much more interesting modeling.
Mike
On 5/12/2005 at 4:54pm, afray wrote:
RE: Re: Japanese-style computer-rpg mechanics? [ot?]
Mike Holmes wrote:
I think if you restated the question it would help. For example, with regards to the question about combat (it is a question, isn't it?), do you mean that being able to take somebody with 100 points of atk difference is something that a system like Disgaea does, or is it a design goal that you have?
Not so much 100 points of attk diff, but one or 2 levels -- which, on high levels, in disgaea, can around 100 attk points (or more!).
I want my system to have infinite levels, and allow players to compete to within +3ish levels of themselves.
Mike Holmes wrote:
And what sort of comments are you looking for from people who may have experience with this sort of system? How to implement it? How to limit it? How to make it work in combat?
What are your design goals that you're trying to get with such a system? [\quote] (<< hmm, I can't get this to parse, but the others will. why?)
World-on-a-plate-style, a few lines of maths showing how a real-world system like this works for simple physical combat. Otherwise, pointers on balancing and resolving simple physical combat.
This is more of an experiment for me, tyring to learn what I don't already know. There's lots of websites on tt mechanics and design, but not so many on this type, which seems to be exclusive to crpgs.
Mike Holmes wrote:
Then, I'd add, have you considered something more interesting from tabletop? Or, rather, from a TT player's perspective like mine, I've always wondered why CRPGs don't have more complicated systems than they do. It's a strange irony that CRPG systems are often far, far simpler than TT systems (except perhaps in the execution of the math), when computers could handle much, much more interesting modeling.
Mike
Maybe they're not more complex because a few people still have to design and balance these systems, which can get exponentially harder the more complex a system is. My limited experience of comerical game design is a large management game, where we had to balance the economy with a very big consumer AI. It was not fun, and it didn't *actually* work, tbh. :-)
Cheers,
Andy.
On 5/12/2005 at 6:25pm, Miskatonic wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Mike Holmes wrote:
Or, rather, from a TT player's perspective like mine, I've always wondered why CRPGs don't have more complicated systems than they do. It's a strange irony that CRPG systems are often far, far simpler than TT systems (except perhaps in the execution of the math), when computers could handle much, much more interesting modeling.
I'm always perplexed by this untapped area in CRPGs, too. I'm a firm believer that computers are where the future of the Simulationist agenda is at.
Here's my understanding of the "Creative Agenda" of Japanese console RPGs: Originally, games like Ultima & Wizardry, themselves trying to emulate face-to-face D&D, took the Japanese market by storm. Future game designers, exposed to roleplaying only through these games, did not make any attempt to be "like" a table-top game. Instead, they latched onto the process of killing monsters, getting XP and loot, and gaining levels. Thus console RPGs like the Dragon Quest and Final Fantasy series are drifted beyond gamism into what is effectively a cool-looking spreadsheet interface with fate resolution and elaborate railroad plots. (I don't know that CRPGs actually have points of contact, or I'd be tempted to call them "ultra-pervy.") Therefore, aside from historical origin, such games have become a distinct creative discipline separate from what we call RPGs. There's no social contract, no shared imagination space, none of that.
There have, however, been no shortage of a distict subgenre of fantasy heartbreaker inspired by CRPG mechanics and level-progression. These are usually homebrew attempts by novice gamers who were turned on to RPGs by Final Fantasy. I have not, in my experience, ever encountered one of these which might be considered a good design, or for that matter able to find regular players. The most famous published attempt, Zak McCraken's Synnibar, is in fact something of a target of derision in certain gamer circles. (But puh-lease let's not start a Synnibar bashfest.)
All of which is my long way of saying that I don't see how this thread is relevant at the Forge.
On 5/12/2005 at 9:01pm, Mike Holmes wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Oh, I'd say that the relevance is that I think that one could, in theory, create a fun and functional game of this sort. For example, an actual case would be not to look at the failed Synnibar, but at Sen-Zar (http://www.senzar.com/). That game goes in much the same direction, but seems to be much more functional. A "munchkin's" paradise, to be sure, but that seems to actually have been a goal of the design.
I'm not saying that it would be for everyone, but I could see a system like this working for a tabletop game. Especially if you used what I fancifully term CARP, Computer Aided Role-Playing. That is, where each player has a computer in front of them, but it's only used for chargen, computation, and record-keeping.
Anyhow, I think that as long as it's speculatively possible, it's worth at least discussing. Just my opinion.
Back to Adam's needs - if I read you correctly, your goals are to have a system with the "natural" ratings which produces characters that, when within three levels of each other, can compete. I assume you want a curve? Something like the following odds for the weaker character to win:
Same level = 40% to 50% chance to win
1 Level lower = 30% to 40% chance to win
2 Levels lower = 15% to 30% chance to win
3 Levels lower = 2% to 15% chance to win
4 or more Levels lower = Less than 2% chance to win
Is that it? Then you're looking for the math that would make this come out right? I'm assuming a standard IGO/UGO round system, or the typical CRPG model where it's based on attack times and such?
Before diving in, I assume that you're familiar with the algorithms by which this is done in most CRPGS? Or are you completely new to this sort of thing?
Also
On 5/13/2005 at 8:30am, contracycle wrote:
RE: Re: Japanese-style computer-rpg mechanics? [ot?]
afray wrote:
Maybe they're not more complex because a few people still have to design and balance these systems, which can get exponentially harder the more complex a system is. My limited experience of comerical game design is a large management game, where we had to balance the economy with a very big consumer AI. It was not fun, and it didn't *actually* work, tbh. :-)
I strongly disagree, not least becuase I used to write die-rolling programmes to balance my die mechanics and detect any results I had not expected. In this manner I could produce the results of many millions of rolls that I would simply have been unable to do in manually.
I don't think that what Mike was driving at was complexity in terms of the quantity of detail or decision, but rather on the thinking behind the stats and the game reality they seek to model. It is this aspect, in CRPG, which is still crude and simplistic, IMO. What we do see in CRPG is an increase in complexity, but not an increase in sophistication.
thus I am still a little confused as to what your question is exactly. If you wanted proposals for algorithms that use a device like this "natural" score, that should be easy enough to suck out of our thumbs. I'm pretty cojnfident there are already some existing systems that distinguish between inherent Talent and learned Skill.
One system that could certainly be used in this way is L5R's "roll and keep system", which works by establishing a number of dice to be rolled, and then a subset of those (the highest) to be counted towards the roll total. So the notaition "5k3" means "roll five dice keep the highest three". Obviously, you could manipulate acquisition of dice rolled and dice kept to reflect the influence of natural talent.
One such system I could propose off the top of my head is as follows: roll Talent plus Skill where Talent is a fixed number and Skill is a number of dice, so something like X + 3D6. Depending on your scaling of the range of X in relation to the rolled dice, you could again produce the Talent effect, I think.
To take the levels of competition issue, Conspiracy X has a semi-random system to achieve a similar goal. This establishes that the raw ability ratings of each character are compared to determine the difficulty of the task to be rolled, with automatic success or failure applying for great degrees of difference. This means only a subset of conflicts are rolled, those that remain indeterminate. We descroibe this system as one that mixes Karma (comparison of absolute values) and Fortune (randomly generated output).
There is a huge variety of mechanical designs in RPG, even if the bulk of printed products uses a rather staid convention. It might be worth your while picking up a few of the small press games to see how weird these can be.
On 5/13/2005 at 9:12am, afray wrote:
RE: Re: Japanese-style computer-rpg mechanics? [ot?]
contracycle wrote:afray wrote:
Maybe they're not more complex because a few people still have to design and balance these systems, which can get exponentially harder the more complex a system is. My limited experience of comerical game design is a large management game, where we had to balance the economy with a very big consumer AI. It was not fun, and it didn't *actually* work, tbh. :-)
I strongly disagree, not least becuase I used to write die-rolling programmes to balance my die mechanics and detect any results I had not expected. In this manner I could produce the results of many millions of rolls that I would simply have been unable to do in manually.
I don't think that what Mike was driving at was complexity in terms of the quantity of detail or decision, but rather on the thinking behind the stats and the game reality they seek to model. It is this aspect, in CRPG, which is still crude and simplistic, IMO. What we do see in CRPG is an increase in complexity, but not an increase in sophistication.
Your dice-rollers are exactly how far I've got at the moment, and it's done its job in detecting the unbalance I'm trying to fix. Video game designers use a lot of prototyping to make sure their system is balanced, but there are other limiting factors, too.
The video game industry is publisher-lead, meaning that publishers supply the money to us developers, and they have the final say when it comes to design. They're lead by margins and profits, which can be *huge* (the latest GTA game mad £60m in one month, in the UK when it came out >_<), so they naturally want easy-to-play large-demographic games. One side effect of this is that we in the west only get a subset of the best japanese CRPGs. Another (and maybe this is good design practice) is that overly-complex systems aren't allowed by publishers because it alienates non-hardcore players, hence lowering profits. I say it's good design practice, because for all the realism you could model in a crpg, it must still be intuitive to the player. He must be able to know that he's got a chance against this monster even if he doesn't know the wind velocity or the month when these monsters are on heat. Simple (-ish, it's still more than a few dice rolls and modifers, check out the FFX stats page linked above) stat comparison gives us this. And at the end of the day, many of these games are brilliantly fun as-is: See Disgaea, Phantom Brave, Knights of the Old Republic, or Golden Sun.
So there. :-)
There are more complex games, but they leave behind other elements of RPGs and have morphed into other genres. First-person shooters, for instance, are incredibly complex when it comes to enemy AI and weapon damage, but it's all intuitive.
And thanks for everyone's help. You've given me lots of hints and lots to chew on. :-)
Cheers,
Andy.
On 5/13/2005 at 11:31am, contracycle wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
I am aware of the problems related to the industry being publisher led (and indeed, these are problems inherent to Capitalism). But what you have mentioned here is one of the reason many games are actually quite poor IMO: intuition counters mastery.
I do understand that you want the player to be able to play as quickly as possible once they get the disk out of the wrapper. But that also means the system isnquite opaque, unless your manual is very detailed - and the trend appears to be away from that. But IME, if you want to grab *and keep* players, you need to give them some way to learn, understand your algorithm and therefore make informed, rather than inuitive, choices.
You see, I specifically did not mention realism anywhere. Part of what I was trying to get at is that yes, some games do have matehmatically complex, realistic models of the game environment. But thats actually easy - just the borrowing of formulae from physics.
What I was trying to0 get at, in terms of designing your game systems to govern character interactions, is that increased realism is not the grail, and inution is not the counterpoint to a representative system.
On 5/13/2005 at 11:59am, afray wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
I'm not sure if I understand what you're getting at.
Why don't crpgs have complex stat calculations? The FFX stat page linked to above has damage calculations that involve powers and division. You couldn't do that by hand.
Why don't crpgs have more stats to simulate more factors? Disgaea has many hidden stats, including master/pupil skil adoption, spell and item leveling, promotions, buyer/seller ranks, team attack percentages, attack height difference, geosquares, etc, that all go far beyond what you can "see" in the atk/spd/res stats, but those stats still give you a good idea of your chances in combat. However complex your background stuff is, it still has to be abled to be grasped by an advanced player, or not surprise a novice player with unexpected results. Disgaea does this *brilliantly*, and it's not easy.
As far as innovation and difficulty in video games go, you can blame the suits. Many video game designers (and I suspect tt rpg designers) fall into the pit of "designing games for themselves." This may not sound like a bad thing, but you can quickly loose sight of the novice or newb's point of view. While the whole game shouldn't be overly simplistic, There should be a learning curve, and hence your system has to be introduced slowly.
In shooters you get one weapon at a time, so you don't get overwhelmed. In Disgaea whole aspects of the combat are introduced one-at-a-time, like the geopanels not featuring for a while. And when it comes to them, you can use them fully and gain a lot, or leave it and play normally.
All RPGs have a natural learning curve with the leveling, but this has to be augmented sometimes.
On 5/13/2005 at 12:38pm, Selene Tan wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
I think what contracycle is trying to say is that many attempts to make games more intuitive also prevent the players from making informed decisions. i.e., games will "hide the numbers" from the players to make things simpler, but then the player's decisions will basically be random guesses because there's not enough feedback.
Meanwhile, I'm still confused as to what you want out of this thread. Are you just fishing for ideas?
On 5/13/2005 at 12:47pm, contracycle wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
We are clearly talking past each other yes.
Let me try this angle: why do you have levelling at all?
I know its a stock staple of the genre, but thats part of the problem, it IS a stock staple and whether or not is being used appropriately or intelligently is open to doubt.
It is that kind of thinking which I suggest needs to be re-examined. What does levelling do for your players experience of play, and why have you chosen that aspect to be part of your game?
On 5/13/2005 at 3:03pm, Mike Holmes wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
In any case, you mention stuff that is both complex and intuitive. Why would you assume that more complexity would have to be less intuitive if that's what you want? You seem to assume that any complexity we can add would automatically become unplayable.
We're actually quite a lot better at design than you seem to think.
For example, to add to Contracycle's comment about leveling, what about detailed damage? One thing that's always boggled me is why damage is always doled out as Hit Points making it a very simple resource management part of gameplay. Why not include hit locations and various effects of damage to different locations? My god, how well would a game sell if your character might suddenly have a limp?
Is this hard to understand or counterintuitive? Hardly. Instead of saying "1237 Points of Damage!" wouldn't it be more intuitive to have the display show that the character's arm is broken? Imagine how much more interesting the game-play of healing becomes if you have to decide which wound to heal up. Instead of just taking a potion and getting back a lump sum of HP.
And I can go on and on and on with ideas like this for how to make a CRPG more complicated and interesting without making it one iota harder to grasp. In fact I can make them easier to grasp at the same time. The fact that CRPG designers haven't figured any of this out is beyond me.
We don't do a lot of this stuff in TTRPGs because, as you note, the math is far too complicated to do by hand. Oh, we want to do it, we just can't. With a CRPG you could. But you don't. I don't get it.
Mike
On 5/13/2005 at 3:54pm, afray wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Mike Holmes wrote: In any case, you mention stuff that is both complex and intuitive. Why would you assume that more complexity would have to be less intuitive if that's what you want? You seem to assume that any complexity we can add would automatically become unplayable.
You got me, I am being a bit cynical. :-)
My point isn't that complexity is bad, it's that needless complexity is bad. Each layer has to complement each other, but still function independantly, and *each layer* has to be intuitive to grasp. There has to be a learning curve for the user to grasp these mechanics, and above all, they have to be fun. ;-) A big pot of great ideas doesn't always make a great game.
Mike Holmes wrote:
We're actually quite a lot better at design than you seem to think.
I didn't mean to slight you or the others here. In fact, one of the reasons I posted here was to tap the wealth of experience that's obvious even to an outsider like me.
Mike Holmes wrote:
For example, to add to Contracycle's comment about leveling, what about detailed damage? One thing that's always boggled me is why damage is always doled out as Hit Points making it a very simple resource management part of gameplay. Why not include hit locations and various effects of damage to different locations? My god, how well would a game sell if your character might suddenly have a limp?
<snip other valid points>
Mike
Location damage, limps, etc., are all good ideas, and would add to a game experience if done well. As far as more intuitive than plain ol' hp goes, I'm not so sure. Hp may not be realistic, but it sets out an obvious constraint. The user can work out immediately how much more damage needs to be done to kill a monster, etc. If the monster has two damaged arms but a perfect torso, how close are they to death? Even if you add hp to each limb, how do they total? And you have to take account of all of this every turn and the monster's natural resistence to elemental spells, armour type, his buddies, middle name...
Plus there's probably other cultural reasons why crpgs are still quite similar to ttrpgs. Historically, they were just computer simulations of ttrpgs. It's only recently that they've had the computer power to branch out in scale compared to TTs, and yet there's already a "tradition" now of doing crpgs in a certain way. Also, crpgs don't make a lot of money (except in Japan) which can stifle innovation.
In addition, you guys have been designing RPGs for years and years. CRPGs to many people mean great storys and characters, with added character development. Producing an original and interesting crpg isn't about creating new mechanics, but about producing a fine story. Your rulesets are so well balanced that we can just grab them off the shelf and spend resources from design on story creation. So we've seen a lot of critically acclaimed and satisfying games that just use D&D rules -- but they're fun because they're so well told. Never as much as a good dm, granted, but still excellent.
And, like I've said and given examples of, there is *plenty* of innovation in crpgs, from full real-time systems and strategy-rpgs to geo-panels and item worlds.
Answering contra's point, I'm using leveling because the point of my game is to take a lowly team of characters and work them up the ranks to the top. Levels provide an obvious way to show the user this progress, and compare players. I know what you mean about challenging tradition, but it's a tradition for a reason -- it's a great system. :-)
Cheers guys, and thanks again for your help. I don't want any of you to think I was trying to "use" you, I just wanted a few pointers. With your help in this thread I should be able to get some prototypes running, and I hope you'll have a gander at them then and let me know your opinions.
Andy.
On 5/13/2005 at 7:16pm, Mike Holmes wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
I know what you mean about challenging tradition, but it's a tradition for a reason -- it's a great system. :-)I can't help myself. Levels a "great system"? Oh man.
Levels were invented in 1973 (earlier, really), and became obsolete in 1976. Yes, loads of people play D&D still. But they play D&D despite many, many, many, many ways that have been introduced that are, at least, intended to be improvements. Sure there will be some people who prefer levels to this day - just as there are people who still think that listening to albums on vinyl is superior.
When it comes down to it, it mostly has to do with people not wanting to get out of their comfort zone. So, if you want to make just another game, go ahead, assume things like leveling is the only way to make a player feel a sense of progress. Despite the fact that even Final Fantasy doesn't really use leveling anymore (when you get a level every fight, and possibly several it's something else entirely). There are so many other ways to do this that I wouldn't even know how to get started telling you about them. It's like you're speaking a whole different language.
The problem with getting advice here on what you want to do is not so much that we're TT gamers, but that we're indie gamers, and can't stand the mediocrity of the standard game. So we're not going to be agreeing with you any time soon that it's a good idea to make a game with levels just because everyone else does.
Just for some context, check out these: http://www.indie-rpgs.com/articles/9/
http://www.indie-rpgs.com/articles/10/
Have you ever played the CRPG Darklands? Ultima Underworld?
Mike
Forge Reference Links:
On 5/14/2005 at 11:00am, Noon wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Mike Holmes wrote: Back to Adam's needs - if I read you correctly, your goals are to have a system with the "natural" ratings which produces characters that, when within three levels of each other, can compete. I assume you want a curve? Something like the following odds for the weaker character to win:
Same level = 40% to 50% chance to win
1 Level lower = 30% to 40% chance to win
2 Levels lower = 15% to 30% chance to win
3 Levels lower = 2% to 15% chance to win
4 or more Levels lower = Less than 2% chance to win
Mmm, probably off topic, but although I've begun to think that when you work out things this way its supporting simulationism rather than gamism. To look at it in a gamist light:
Same level = Small amount of resource management to win
1 Level lower = Medium amount of resource management to win
2 Levels lower = Quite a lot of management to win
3 Levels lower = Tons of management to win
4 or more Levels lower = Shit loads of management to win
When you work in percentages of who's going to win, you start making a game that tells you how the world works. And the gamist says 'so what?'.
Is this hard to understand or counterintuitive? Hardly. Instead of saying "1237 Points of Damage!" wouldn't it be more intuitive to have the display show that the character's arm is broken? Imagine how much more interesting the game-play of healing becomes if you have to decide which wound to heal up. Instead of just taking a potion and getting back a lump sum of HP.
Because most programmers don't know when to stop, and code in a death spiral. Ohh, so fun.
Personally I'd toyed with the idea of a one step death spiral. So you can have a broken arm, or broken leg or a spraned ankle...but only EVER one at a time. No stacking, no matter how the lure of realism gets to you. Once you have one of these wounds, you just can't pick up another until this one is healed/gone. This way you have some resource issue to deal with, which can change (heal your ankle (yay, I can run now!) and damn, get a big black eye from another fight...can't rely on sniping so much for awhile). Descrete wounding adding challenge to gamism.
On 5/16/2005 at 1:07pm, Mike Holmes wrote:
RE: Japanese-style computer-rpg mechanics? [ot?]
Not sure what you're getting at. I'm not proposing that the players would know these odds. Just that this is, given the same tactics by both sides, and all else being equal, what the odds shifts might be vis a vis two characters of differing levels. The system wouldn't work out a percentage like this, and then roll to see who wins. You'd do round after round, etc, etc, playing out combats until, on average about 35% of characters one level lower won against those one level higher. That is, the percentages I gave would be design goals, not part of some mechanic.
And I can't tell from your post if you're pro or con death spiral. In either case, detailed wounding only seems to imply a death spiral - it doesn't have to end up being one. In any case, I think discrete wounding and being able to work around it in terms of fighting well or knowing how to deal with it in other ways would add all sorts of gamism.
In addition to all of this, I think that the way that many people play such games is purely explorative. That is, they're not playing to do well, they're playing just to get to the next cut scene. Watching my wife play FFX for example, convinced me that this is the case. Takes her about twice as long to get from scene to scene as me. The only incentive she has to do better is not based on a personal desire to display how well she can do vs the engine, but instead just to get to the next cut scene more quickly.
The real player rewards in this case are purely participationism exploration. So the assumption of gamism is, I think, at least not what all CRPG designers are really looking for.
Mike