For the purposes of this thread I'm going to seperate character creation from task resolution. Task resolution, combat, and so forth is the mechanics of the system while caracter creation is... roughly, the user interface. I hope the reasons for this comment will become more clear as I go along, but briefly: Any number of games get one right and the other horribly, horribly wrong.

I'm not setting out to refer to creation under a jargonny term, mind you. Just note that when I refer to Mechanics, I am typically not referring to caracter creation in any way.

It is unnecessary to note that this is heavily influenced by my own opinions.

In no particular order:

Traveller, any edition:  Mechanically Traveller is sound. Its simple enough to grasp, easy to use, and not actually boring.  While I have few experiences with the classic system in actual play (Mega/new era trav and gurps mostly), I believe it works very quickly at the table.   If it is limited it is in terms of adaptibility. Traveller is essentially realistic and gritty, from the rules up. Some games attempt to impose gritty reality on top of the rules, Traveller just started from realism as an assumption, I think. This is, mind you, a game with psychics and lasers.  
Mechanically: Fixed dice, standardized variable target number, small number addition to dice rolls.  I do find a minor communication error in that two opposite actions (making rolls easier and harder) seem to be expressed in virtually identical ways, but I could be misreading.

In character creation Traveller is a classic example of 'added complexity'. The heart of character creation is a set of simple randomized stats and a fairly small number of skill points that are distributed. The careers and other rolls that make Traveller character creation into an entire mini-game serve as a roundabout randomizer for skills more than anything else. It should be no surprise then that one can easily find an online 'generator'  that can produce complete classic traveller characters with a mere two drop down menues. If you're not interested in solo character creation, the system can actually bog down starting play.  That said: It is fully servicable and functional. Almost totally randomized creation will more rarely produce broken, unusable characters than allowing players to assign skills etc freely.  At least in Mongoose Traveller, however, the career paths are a little to... fat... for what they are doing. Giving six tables to chose from, but with almost each table having the same four (out of six) skills to chose from is largely an 'illusion of choice'.  Similarly, in MongTrav, the three sub-careers tend to be very simply a choice of risk/reward ( A 'dangerous' career moves the standard risk (death/mishap) one deviation, and the promotion one deviation the other way, a safe career exists that moves the deviations the other way. It IS a bit muddier than that, but largely holds true in many careers...) This sort of simplisity makes mathematical sense, but tends to make the 'game' boring.

Mechwarrior Third Edition/Millennial Edition:  I found these two games, seperated by nearly a decade, to be nearly identical systemically. Nearly.

Mechanically, Mechwarrior is simple and sound 2d10+skill vs Difficulty is easy to grasp, the two dice reduces the 'swing' of a single d20 while providing nearly as many possible outcumes (19 v. 20), so extreme outcomes are less extreme. Functionally, this is Traveller with larger dice and a different setting. To be honest, I'm actually surprised this wasn't a bigger deal than it ever turned out to be, as I believe the Battletech setting has a broader appeal than the Imperium, and as noted the games are clearly related, mechanically.  MW does, however, suffer a bit mechanically in combat. While the 'ability damage' of Traveller may not be to everyone's taste, the odd scalar damage of MW is a stumbling block, and less intuitive by far. Translating Traveller's damage to the character sheet is obvious once explained, while the similar damage dice in MW produce look up results, a stumbling block.

The similarity to Traveller is even more explicit in Character Creation, which uses a more evolved form of career path, and also the single biggest change between THird and Millenium editions. Where the simple random rolls of Traveller often created Illusion of Choice, the career path (which unlike Traveller actually doesn't gloss over childhood... this is a debatable addition, but I rather liked the idea of cradle to play lifepathing...) legitimately puts every choice out there. There is less randomizing, and all choices are legitimate. MW also has a deeper character system, not just stats and skills but background... merits and flaws if you like. Some prefer the simpler characters.  In general, Millenium edition is the better system here, by far. Third had some issues with attribute generation and allowing for some backgrounds as purely a product of choice (so to characters that followed the same lifepath (brothers, maybe) might actually be mechanically different. The Millenium Edition actually unified all of character creation with the addition of a simple points(xp, by the way...) system (buying blocks of background from a large pool, and using the leftovers to customize from there).  Third Edition had a diminishing returns mechanic to character creation. It was fully functional, so that after a while trying to scavenge more points for, say, rifles, was a fools errand. Its only downside, and this was a doozy, was that it was baroque. You could accumulate dozens of skills, each with dozens of 'points', and then looked up those points against a four column table, and had all sorts of leftover points floating around that could be redistributed... it was labor intensive to say the least, and not alway clearly explained. Millenium still carries the trappings of this system, but they improved it, largely by unifying the previous points system with the progression (xp) mechanic.  The numbers being used are larger, but also much simpler to grasp, and thats not a small thing.

Cyberpunk 2020: Mechanically simple: Stat+Skill+d10 vs Target Number. Damage is applied fairly simply to a fixed damage track.  The biggest complexity in the system is for the GM in determining accurate TNs, which have a lot more swing than previous systems. While I personally love this game, I do feel that CP202 provides a classic example of 'what not to do', in specific: It is trivially easy to overwhelm the variable with the fixed numbers. The lowest level of difficulty in the system is, essentially a TN of 10, and the highest is thirty (since Stats and skills both go to ten). While exceptions may push both numbers past their limits (a stat of 2 with no skill makes a TN of 5 a feasible difficulty, for example, and with cybernetics it is possible to have a stat hit 15, making a TN of 35 doable...).  Note that the STANDARD range is a twenty point swing, while the dice only covers ten points. Now, this may or may not satisfy some peoples conceptions of physics engines (that is: Some exceptional people don't actually sweat the small stuff, and average losers will never hit it big...), but fully half of all possible TNs will ALWAYS be extranious to any give character, as outside the variable swing.  Now, this is vaguely offset by crit/fumble mechanics on the dice, and the LUCK stat, but some argue that the crit/fumble actually compounds the problem, where the best shot in the world (20+d10, better with good gear) still misses ten percent of the time, and in potentially outrageous ways, even against the easiest of shots.  Remember two that a d10 provides roughly twice as many crits or fumbles as a d20.

Character Creation:  CP2020 provides some early examples of good design decisions combined with classic problems. You have something like ten stats, including plenty of duplication (you have three social stats, one of which is one of two potential 'dump stats'... Appearance, which can be improved cheaply with cash at any point. ) it offers random rolls, but the default is for the GM to pick a 'power level' for his game, though honestly the power levels aren't really all that noticable, as a difference of five character points between ten stats is fairly minor (and stats are the only thing altered).  A seperate pool of points (40, plus 'extra' points derived from two stats, so lets just say 50 and have done) is given to divvy up. The actual number of points given is very well balanced, as it is damned hard to argue a guy with four 10 point skills is actually viable, much less balanced. There are core classes, but this is (again) largely an illusion of choice. Each class gives ten skills, but only one defines the class, so you have nine generic skills that are largely thematic choices (and to drive the point home, the main book explains how classes can be adjusted to suit, or even created from whole cloth), and again there is some duplication.  Nomads and Rockers both have class skills that provide NPC goonsquads , Corporate and Fixer both provide a pool of cash and connections (being upgunned versions of normal skills). There is also niche protection skills. Only a netrunner can use the Interface Skill, making him much more capable online than anyone else. Only the Solo gets a massive initiative boosting skill. Only the Techie can Jerry Rig (jury in the book). As noted, however, its an illusion of choice. The Core skills can be removed, along with the classes, without harming the game notably, or they could be freely made available through the skill system without too much harm.  Another issue here is that CP2020 suffers from massive skill inflation. There are way too many skills, and too many artificially divided skills, even disregarding the special class skills.  There are, for example, four different mechanic skills, with no crossover between them. WHile I have no doubt that a helicopter mechanic learns some stuff that a vectored thrust mechanic doesn't, and neither cares much what the guy working on his car knows about, say, mufflers... that the master helicopter pilot is literally no better at working on the Vectored thrust vehicle than some chump off the street with the same natural talent (TECH stat) he has is laughable!  So we have near duplicate skills. We also have useless skills, like weapon maintenance. Now, sure: This is a good thing to know if you work with guns. But what do you make of a guy who makes this his best skill, at ten.  The zen master of cleaning his gun?  Is there as much to know about cleaning one's gun as there is to know about, say, computer programming? Olympic level gymnastics?  Has there ever been a CP2020 game where the adventure hinged on that one guy making that amazing TN 25 Weapon Maintenance check? I suspect not. Its just one example, I could name a good half dozen or more.  There is, for example, a 'feats of strength' skill that governs, among other things, such important events such as ripping phone books in half. Now, in theory knowing proper lifting and tearing techniques is a useful skill, but we are talking about a game where strapping on a exoskeletal suit is a viable creation option. Lifting CARS negates your ability to tear phone books in twain, I think.    
The last major sin of CP2020, to my chagrin, is that character power largely is a matter of wealth.  Its not as obvious as its closest competitor (shadowrun), but while gathering experience/time to learn new skills is useful, its better to get more cyberwear and better guns/armor.  The core book is actually a weak offender compared to the later supplements (power creep!), as cyberware isn't that hot (a high end reflex booster is a major investment, yet a newb solo without one is probably faster than you +2-3 vs, say +4 or +5 (professional competent but nothing special)).  An SP25 body armor makes you feel pretty bulletproof, but a 20mm cannon will still punch nice holes in you. Of course this has the attendant problem of having to lug that beast around (it is described, as I recall, as being six feet long).  

The Old World of Darkness Line:  Mechanically these were all pretty simple, though different editions did pretend to make them more complex. Essentially a dice pool, with a essential hard cap of ten, with a fixed TN, count successes.  Moving the TN around, changing minor rules about ones and tens... these were window dressings.  I do think that, as dice pools go, the OWoD hit a sort of sweet spot. Pools of ten were somewhat unusual, and were just functional.  Actually, all the minor alterations over editions was the worst aspect. Dice pools have many problems, but bar none they are dead simple to use, perfect for casual players who don't want to work hard. More dice good, less dice bad. Moving the TN from seven to eight, or the other way round, doesn't actually change how the game works, it just confuses people who know both versions.  Note: I'm not against nifty quirks (ones cancel successes, tens count double, or allow extra dice...whatever...), but I do think that in a single game line (or group of mechanically similar, thematically linked game lines), that constantly tinkering with the quirks is problematic without a corresponding improvements to the structural aspects of the rules.  Likewise, such little quirks are among the most heavily houseruled aspects of the games, simply by default.
Now, I am not a fan of dicepools in general, and despite hitting sweet spot of playability, the WoD implementation has always been the worst of the lot, but alot of that is simply taste. My big beef, mechanically, is that the designers never committed to making the games feel like they were played, a 'disconnect' between the author and the audience. Ironically they should have known that the audience wins that arguement every time (there is a story of a famous author... I'm going to erroniously declare it was Harlan Ellison... who snuck into a college English class where he knew his work was being analyzed by the instructor, then jumped up to declare her(?) wrong. When he revealed who he was, she told him his opinion, as the author, didn't matter.  I say this, because the WoD crowd has always struck me as the sort of collegiate elitist snobs (my philosophy degree means I know more about farming than you!), who should know this story and yet ironically not understand how it applies to them...   In short, the game is packed with artificial restrictions that keep your monsters from being much more than ordinary guys with a dietary restriction and a bad sun allergy.  However: That is not a structural flaw in the design, and could fairly simply be undone with a single permanent marker and a few hours of flipping pages.

Character creation in the OWoD has always struck me as ideal. Let me try again. It has struck me as Ideal (note the capitalization!). I'm going to bring NWoD up here, briefly: By sorting the nine attributes into a neat grid (physical/mental/social crossed by Act/react/resist) while actually intellectually interesting, actually sucked some character and... call it life... from the attributes without meaningfully adding to the game. What worked well for physical attributes didn't translate well to social or mental.

I won't claim its perfect, or even without flaws, but the concept of subdividing points into clear focus, and seperating out different areas to create more organically balanced characters is a good one for a point based system. The pieces have enough interactivity to make it organic. Likewise, the balance of skills is more or less in the sweet spot. It could probably be improved by removing x and adding y, but in the main most people wouldn't notice.  I do think some later versions overthought it by trying to recreate the divisions of the attributes, but not fatally, and not without some good reasoning along the way.  On the other hand, I'm not entirely sure that trying to build in forced growth assumptions into the design was wise (see again my comments about author v. audience debates...). I also think that, at least this far into creation, it is perfectly reasonable to build a WoD character without any need for character growth... eliminating the often problematic issue of character growth rate (too fast or too slow).  

Beyond that, however, things largely break down to questions of personal taste, power levels and supernatural whatsits. I could mention how the idea of Humanity wasn't a bad innovation of an older idea, and was nicely tied to what they were trying to say about the game's themes.

Exalted: Briefly, this is the sort of mess that happens when you stop thinking critically and you intend to insult your audience. As it is essentially the same WoD game with a different theme, I won't go too in depth. First, however, you have teh vastly inflated dice pools, which if nothing else proves my earlier point about sweet spots. Too big and the pools become more unwieldy than just asking players to do math.  Since so many actions involve the use of a vast array of clearly defined 'powers', this winds up being a mechanical issue and a character issue at the same time. The amount of viable action available grows too big for any one person to master, and the powers themselves are radically unbalanced (Unbeatable defensive powers make the game essentially a excruciatingly slow process of depletion, yet without unbeatable defenses everyone dies the first time they are hit. Rocket launcher tag, to steal a phrase.  The swing between these two states is horrific.  

Likewise, Exalted manages to have the worst version of the otherwise ideal character creation system, not least because of the massive proliferation of powers (and their requist groupings...an unnecessary cludge decision... the sheer number of bad decisions made in exalted's design process is worthy of an entire thread, and people have done just that elsewhere...).  its still not bad, but it is the worst implementation of a good idea.  Some of this comes from forcing the skills to fit their 'class' breakdowns, and the odd ways they occasionally did so. This was marginally improved in the second edition but...

D&D: Ah... which edition? I'm going to catagorize 4E as its own game, which could be a judgement call. For the purposes of this discussion I'll keep myself to some universal traits of the prior editions from AD&D through 3E. Sorry basic guys, but I gotta go with what I know.
Mechanically, D&D is sound enough. Its not elegant, but it is functional, with differences between editions. AD&D required more table look ups, 3E got into exception based design with feats and the sort of exhaustive attitude that everything needed to be covered. The phrase 'That which is not permitted is forbidden' springs to mind, and since they wanted to avoid forbidding too much they tried to permit too much... if that makes sense.  One oddity that strikes me is that even in previous editions all those funny extra dice are just that, extra.  They really only ever got used in damage, either giving or taking it. Oh, there might be a few edge cases I've forgotten about in older editions, or some randomized tables the GM could use for ideas, but in the main... d20 was king, which is oddly enough why they named the 3E edition the D20 edition, I suppose.  As dice go, its fine. It works a bit like a faster percentile system with a little less granularity. It is swingy as hell (the swingyest!) but it works.  The problem, which has long been an issue in D&D, was that as the game progresses and various fixed bonuses accumulate that it gets a bit overwhelmed by the stuff you are adding to it.   The high swing factor means that preditive results are actually somewhat hard to come by, and the system does stabilize the further you 'break' it by piling on numbers, after a fashion.  
Balance between magic and mundane has been an issue, and one that generates LOTS of debate, and I won't go into it. I just can't ignore the existance of D&D.   More interestingly is how D&D, particularly 3E failed in handling scaling.  However, neither here nor there.

What I want to get into is character creation. This is a shockingly big part of the game. As I pointed out in Traveller and CP2020... pretty much every game so far, the character creation system could largely be altered in any number of ways without affecting the game noticably.  Remove or significantly alter D&D's character creation and you wind up with something radically different, though there is some room for flex (see D20 Modern and its kin, for example...). The Class system is ruthless, your class literally gives you access to MAJOR portions of the game that other classes simply don't get. This is less true of 3E than previous editions, both with the removal of certain sub-systems (thief skills being classic here), and the easy cross-classing.  Classes serve as your primary number matrix and your access to the game.  Levels, of course, serve as your power determinant.  I do think that 3e made a good decision to cap the levels at 20, as nothing exhausted my teenage self nearly so much as having some asshole tell me all about their awesome level 420 wizard who had survived ragnarok... a not infrequent enough occurance, but they fumbled a scalar jump into epic, which was clearly a core idea (as indicated by its inclusion in the DMG in a much reduced format...), which is too bad.    
Anyway [break to make dinner... must divide post!] one of the things that occurs to me, when looking at it is very much that D&D winds up being a fairly losely defined mechanical system for resolving actions from a very strong, if not necessarily good or bad (taste matters here), character system.  People worry less, between editions, about the changes to mechanics like Thac0 or AC than they do to the fundaments of the characters system, and fundamental changes to THAT system (4e) make a stronger change to the mechanics than almost any other approach.

Seeing at how massive this is growing, I will include one more in this main post before taking a break and moving to a new post.

GURPS:
Gurps gets a lot of flack for its cludgy, turgid system. It is overwrought, overblown and just too god damn big.  Well... yes.
However, that is almost entirely the character creation part of the system. Mechanically GURPS is fairly simple.  3d6 and roll under your skill, with modifiers applied directly to the skill total (thus eliminating the confusion I found in the visibly simpler Traveller... which may be cause by different editions approaching the same mechanic from different angles... I will look into this.) What GURPS does have, mechanically, is a lot of different options. The term 'toolbox' does get thrown around, and I find a metaphorical use for that metaphorical term does adaquetly supply the illustration of the problem.  Let us say that you have never fixed a leaky faucet (if you have, pretend, m'kay?). Its a pretty simple job, as jobs go, and mostly requires a single wrench to get it done, but if you've never done that...well.  So you have this leaky faucet, and you want to fix it so you ask your buddy to bring over what you need. And he comes in with a wheely toolbox with eight drawers and an expanding top, holding some twelve cubic feet of space, and a couple of hundred tools. Including a wrench.  Heck, this bad boy even has speakers and an Ipod hookup (now you know what I want for xmas...). Swell.

Now, if you are comfortable with what you're doing, you reach in that sucker and grab a wrench, maybe two if you're not sure what size will work better, and you get to work.

But if you've never fixed a leaky faucet?  Confronted with the sheer mass of options your brain shuts down and you check out. System overload, fuck it, grab a beer and call a plumber.

Of course, most, but by no means all, of the options are front loaded into character creation. And of the options that are left its not entirely clear which are actually optional. I mean, I can look, right now, at a GURPS character sheet and point out one area where a lot of people start getting lost: the weapon spot. Seriously: For ranged weapons you have ten columns for information AFTER the name of the weapon itself.  At least half of them have, at best, marginal utility on the character sheet or even in the game. Hell, I've been a fan of GURPS for almost twenty five years and I couldn't even tell you, off the top of my head, how some of those columns work. I can guess, of course.  However, really, the ONLY things you might NEED to make a functional GURPS weapon is the damage, maybe the range, and the number of shots.  You MIGHT find a accuracy useful, but seeing how its actually a fairly complex subsystem, rather than a simple addative, I'd actually suggest not. (I do note that 4E Gurps is  simpler,  I do note it is simple enough to 'get' at a glance, and thus is functional for more than 10% of gamers...).  I, for one, have found that GURPS actually breaks (IMNSHO) when you attempt to play it with all the bells on, requiring extreme degrees of specialization to acheive modest levels of success.
Mechanically, the 'hardest' thing about GURPS is, possibly, that it follows radically different assumptions than most games. Turn length, for example, is a rather brutal one second, and thus actions appear to take 'ages'. A single 'round' of D&D combat is, theoretically, TEN rounds of GURPS combat. Now, I have seen even more tight games (Phoenix command has quarter second phases!), but most of them are a bit looser about 'grouping' actions. Another is that GURPS doesn't have an Initiative 'system', or more accurately: Initiative is fixed. Randomness is reduced. It is one of the few systems that can honestly suggest that character creation determines seating arrangements without being the least facile or glib. I do note that this 'system' is not entirely unique, but it IS unique, so far as I know, in that it is not presented as a lazy option to the main system but is, in fact THE ONLY system given.  Another, not so unique but still 'difficult' aspect is the sort of gritty realism that is aimed for.

But Character Creation is where GURPS loses, well, everyone. IN 4E GURPS they don't pretend otherwise and actually make it an entire book, seperate from all other rules.  It is, of course points buy, with attributes providing a powerful foundation for skills, largely being a dominant factor in character creation (and likewise often consuming the majority of the points...).  This has never changed so during my time with GURPS. I note that divided point systems have received my general praise, and it is in comparing them to GURPS that I am able to say that without feeling like I'm stating an unsupported opinion.  However, that is a minor issue (one I note that my first ever GURPS GM actually corrected, in that his character creation guidelines explicitely capped point expenses for various catagories, including a brutal 85 point maximum for Attributes (Brutal I say, as a: This was a 300 point game, and B: I'll spend 115 points on a 100 point character (legally!) if I have the chance. And it will be a well rounded character when I'm done. Mad skillz.)

Where GURPS fails, to me, is ultimately Presentation. I do think it 'tries to hard', but presentation is the primary issue.  This shows in a few different areas. Namely, and especially in 4E in the Advantages and Flaws portion of the game.  Not only is it distracting to find 360 degree vision next to more universal advantages (and in fact is the first advantage!), but really more exotic advantages are more usefully seperated out into their own chapters.  Most games will include more or less human (or human like) characters... characters who are highly unlikely to have eyes on the back of their heads. Most games will not have timeline altering dimension hopping characters who can reach across realities to blindly find exactly what they need right then. Yet all of these abilities are presented right alongside more 'universal' abilities like being, oh, ambidexterous.   Skills suffer much of the same problem, being organized into like catagories but not into usable frameworks, so you look for broadswording a motherfucker right next to lazering a motherfucker, instead of next to horseriding (which is itself next to flying starships...). Now, if you are just grouping things out of an OCD sense of order, I suppose either method is equally valid, but if you're organizing according to principle of use for gaming? The method they chose is somewhat less useful. ( I note that 4E organizes by strict alphabetical lines, which is no worse than the previous editions (by catagory), so long as you know what the proper GURPSian name is for the skill you want. Again, for a GAME product I still hold that grouping by 'typical setting' is probably a better starting point.)

In the trying too hard catagory, GURPS does suffer from Bloat. Lets start with skills this time: I will note that the Skills chapter is 66 pages long, and the index for skills is five full pages of double column entry. I was going to count them but I'd like to get some sleep tonight, and that doesn't even reference the various sub-skills.  I will note that Mathematics has six specialities (and you have to take one, thus there are technically six Mathematics skills), and that making a gesture like 'drinking' to indicate you are thirsty is, in fact, covered by a skill (it is, I believe, considered a language skill, and is, in fact, called Gestures. D'ya need a page reference?).

I'd make a joke about having a skill for walking, but it's a bit sour when you notice the Hiking skill.  Now. It is not innacurate to suggest that there is some measure of skill involved in long distance cross country walking, or that a mathemetician working on spatial theory has a different skill set than a mathematician working on cryptography, which is a different skill set from what a stastetician does. On the other hand, even leaving aside the fact that this is a game (where they really are functionally identical uses of mathematics!), but that it can't even claim to be that good at emulating reality, as I'm guessing that the difference between them is significantly less than the -5 penalty the game assigns (or one point less than the penalty for not having studied math at all, technically (plus, you know, the higher starting point, but still, try to keep up!).  In short GURPS is fat, so fat that it defies itself (since Bicycling, for example, is an actual pointed skill in GURPS, that means that your average person (100 points, theoretically) has spent at least 1% of his entire life mastering the art of bicycle riding if he knows how to ride at all.  Since your average person can be taught in a single day, and one day is less than 1% of the average year, and the average person of 'adventuring age' is in fact far more than one year old... well...)

The problem in 'trying to hard' isn't just bloat, its actually weighting.   Again, according to GURPS (real!) it takes 200 hours of study to be minimally competent at riding a bicycle. That is five weeks of moderately intense training.  In reality, its over weighted.  One point of skill should either represent the product of five weeks of training OR one point of skill should represent a far smaller investment of your time.

In the advantages the weighting is fairly scattershot as well. Being ambidexterous is worth five points, despite providing no real advantage (in GURPS rules carrying a gun in your 'off hand' doesn't give you extra shooting, as it would in many games that have ambidexterity... it just reduces the penalty for chosing to shoot with your off hand when you do so, for whatever reason. Combat Reflexes is worth 15 points, and is useful to anyone in a fight. Sure, I get that its useful to the point of being nearly mandatory, and therefor should cost, but in comparison that same number of points nets me a +4 bonus to almost any damn skill I want! (To clarify: An average skill gives me Attribute +4, an Easy Skill would be +5, hard (only a tiny handful of combat skills are 'hard') is +3, unless I'm adding those points to the skill I already bought, in which case its back up to a massive four point improvement.

So, is a one point shift of initiative order, a +1 bonus to a few extra actions (quickdrawing and defensive rolls) really 'worth' the same as being effectively a grand master of stabbing a motherfucker in teh face?  Debatable.  The supernatural version of this same advantage is 45 points, or damn near half of an 'average' person (now, a supernatural ability and an average person aren't exactly scaling the same...).

Its not that the weights are horrifically broken, its just a sort of off-beat icing to the bad presentation cake.

Anyway: That's all for tonight. I'll be back later to address more game systems. Sorry for the short shrift to D&D along the way, but seriously: I don't have the time or energy to wade into the most talked about game on the planet, and I don't actually see much need.

Interesting, look forward to reading more.

I'm glad you like it.  Its not quite as deep as I'd like, but then its also longer than I'd like per entry... lose lose, really...

In honor of Gliechman's personal peeve, today's entry will be Dark Heresy and its bastard children.

Dark Heresy: Mechanically it appears at first glance to be a robust and simple system, a percentage mechanic with relatively fixed target numbers (attribute based). You have low hit points, making the system reasonably lethal, which matches certain design expectations.  So far, so good.

However, it suffers from two problems. I'll deal with the simpler one first. I will call it, here, the Assumption of Zero drift propblem.  When designing a task resolution system you have to have certain built in assumptions, which will include a sort of baseline difficulty.  What is a completely natural, unmodified, roll actually representing?  This is Zero. Zero modifiers, baseline difficulty.  Most games keep it intuitive: natural rolls are 'average', not just in terminology but in application. Shooting a gun with no modifiers? You must be plinking at relatively close range at a static target in daylight.  Penalities (or shifted TNs) are then piled on as things change from this default.  Dark Heresy, however, assumes you're shooting in adverse conditions at moving targets.  The default Zero, for Dark Heresy, is actually Hard. They call it average.... no wait, they actually don't. They call it challenging. Average would be 'routine' checks, at +20.  
Systemically, I have to applaud this. I DO applaud it. However, as a game design decision it sucks balls.  People complain all the time about the wiff factor in Dark Heresy. Are they wrong? Well, yes they are, but it really is the game's fault for forgetting that most people are lazy and stupid, and more than that GM's absolutely hate giving out positive modifiers. They can be sweet talked into it, they can be 'forced' by look-up tables. They can even be relatively generous in benign circumstances to avoid bogging down the game play (driving to the mission? Easy check, since you apparently really want to roll that driving skill...no problem!).  I'd go so far as to suggest that most of the time, if a check 'deserves' positive modifiers, that most GM's would rather just assume it worked.   Thus there is a serious drift between what the game assumes Zero to be, and what Players and GMs assume Zero to be, hence the name of the problem.  
Mechanically it is perfectly sound, of course. We could call this a PR or marketing failure. Design wise, one supposes that simply reinforcing the notion of 'less difficult rolls' wouldn't be too hard to do, even if it would only be a bandaide.

The other main mechanical problem, however, is thornier. I could simply call it 'talent bloat', but that makes it sound like a purely character design issue, and it isn't.  In Dark Heresy the talent bloat is so bad that it has consumed most of the game. Half or more of the actual rules of the game are found in Talents. This isn't quite the same as D&D's exception based design(but similar). Usually the talents don't actually allow you to ignore various rules (though some do), so much as they give you all sorts of new things to do. It isn't just combat, but everywhere. Want to motivate men to charge into combat? There's a whole raft of talents for that!  Want to fly? Talent!

A lot of games do this sort of thing, I suppose. I can think of a few games that were built from teh ground up to support a collection of powers (Champions), that happen to work really well.  Ironically, while ultimately DH has fewer 'talents' than D&D has feats, the later actually works better, and for one simple reason: A high level D&D chacters is only going to access a tiny percentage of the available feats throughout his career. A DH character will accumulate a majority of the Talents, and without them is essentially crippled.  Both games 'chain' their feats/talents, but in D&D there is a sort of logical mathematical progression that is absent in DH.

Its worse. In D&D you can largely ignore Feats when designing opposition. Monsters don't require feats to function, though they can be more challenging with certain ones. In DH, however, a high level monster (a Tyranid Hive Tyrant, for example...) Requires dozens of talents to remain remotely competetive with Player Characters. His raw numbers may be terrific (as appropriate for a twenty foot tall killing machine), but the talents are required for him to make an even remotely decent use of them.  Think I'm kidding? There are nine seperate entries under traits in the Hive Tyrant, two of which are stacking (Heightened Senses (all) is four traits. Psy talent ( is, well, eight.), plus the monster/NPC traits, which are really just reskinned talents, of which there are 13 (again, some are stacking). Now, some of this is idiot nonsense which could be removed. We don't really need a trait entry for armor or weapons, we need an armor line (we have) and a weapon line (we have). We ALSO don't really need entries for unnatural attributes (They are listed on the attribute profile), and I'm not clear how useful listing 'Tyranid' as a Trait actually is.

In some ways it is presentation, at least for monsters. A lot of unnecessary talents and traits when what you really want is the raw data to use at the table (which, btw, is often hidden by the trait listings. No where, for example, in the Tyranid Hive Tyrant entry does it tell you what Heightened Senses actually does. (+10 bonus for tests using that sense)).  For characters it is a bit thornier, as they must access those talents (or traits...) in order to get those benefits.  

This makes things more difficult all around. On cannot simply glance at a character sheet (or monster entry) and make a reasonable assumption about its capabilities, one has to parse through a double dozen talent/trait entries. Each of the 300+ talents must be known perfectly in order to intuitively run the game, while players have a lower hurdle (but still a hurdle) in that they only have to remember the dozen or two they personally use in order to be fully effective.

Character creation is naturally dominated by traits.  Most of what's wrong can be summed up in the fact that any given 'class' is almost purely defined by the order in which they access these traits... as given out over a four or five page entry in a book.

Think about that for a moment. D&D character classes are DEFINITIONAL to game play, and yet can largely be summed up in a single page chart, rarely two.  Dark Heresy classes are mostly notional, yet take up five pages. Mechanically: What is the difference between an Arbiter in carapace armor with a shotgun and a trooper in carapace armor with a shotgun?  At level 1? At level 9?  I submit to you that there is a greater mechanical difference in chosing to be human or elf in almost every fantasy game ever made.

Its more than that, however. Most of the classes are somewhat broken, at least in a half dozen small ways.  As they are merely long lists of talents and skills, and the order they are allowed to be gained there are a lot of simple errors. This skill is a prerequist for that skill (or this talent rolls that skill), but you can't get that skill until after, or at all, thats common. Talents that make logical sense in a career aren't present because... well... someone should look into that, I guess.  Some of it is arbitary and capricious, and why wouldn't it be? You've got to put some 100-150 talents and skills down for each class, divide classes into meaningful branches and try to make very similar seeming classes at least a little different from one another. Thus a gunslinging criminal might lack a crucial pistol talent simply because they gave it to the trooper-sniper already and don't want too much cross over. But you aren't likely to discover that until you try to make a given character type and play through, oh, level five or six.

In short, its a mess. So much so that it handily papers over any number of lesser problems that I might be inclined to talk about (money and equipment availabilty, the way races are handled, or not handled... that one winds up being at least half a bitch about talents and 'classes'...) and progression.

Eh. One more point on theme, then I'm done with this post.

The advanced class entries are all rather nifty seeming, but given the insane method of character design due to the nasty talent bloat, it winds up being a massive, flavorful, list heavy way of saying something like: Trade Talent/skill X for talent/skill Y.  Let me illustrate with an example from the books. Sister Oblatia (chosen at random because I remember the Sisters of Battle easily), replaces your normal level 5 level, or higher. I'm not going to list the 22 entries in teh Elohiem (level 5 militant path for a SOB in the Inquisitor's Handbook), nor the similar number of entries for the Oblatia. I will point out that what you lose are things like Acrobatics and Climb, and what you gain are things like Interrogation and Inquiry (obviously, if you were a Sister Dialogous you'd trade some other skills for more combat oriented abilities), and the Oblatia Unique Talent of Duty Unto Death.  Now, let me ask: If you took Oblatia instead of Elohiem, but then return to your normal progression at level 6... does this mean you will never be able to take, say, Acrobatics +10... I mean, you didn't get Acrobatics at 5 (since you were Oblatia and all...)...

Its just... sloppy. A mess, bad design work. Problems that shouldn't exist are created simply because the framework was badly chosen, and the framework was chosen because the Talents wind up being the very heart of the system when they shouldn't be. (attribute checks should be, since thats how tasks are actually resolved.).



I was going to do this entry anyway, but I did move it up the queue as its been a topic of discussion around here.  Sorry for the slow posting speed, I've had a lot of different things going on.

I'm going to stick with my 'one system per post' format due to general exhaustion and 'wall of text-itis', but I'm going to break faith with my 'popular' heading.  

Today's Post is about Phoenix Command!

Designed by part time NASA engineer (or something. Seriously, this guy has street cred) Barry Nakazono in the early nineties (late eighties? I played it sometime around '91, maybe earlier...), Phoenix Command as a stand alone system is something less than an RPG and something more than a purely tactical game.  So why is it here? Because I can, and because Leading Edge Games also put out a shedload of product that WAS RPG stuff, to include Living Steel, Lawnmower Man, Aliens and so forth. For a while there they looked to be aiming for 'king of the movie franchise games', until WEG beat them down with Star Wars and Ghost Busters and Men In Black and more... and yes, at least one WEG system will be handled here, eventually (Masterbook related, certainly.), showing that with a bit of work PC really did have RPG chops, and somewhat respectable chops at that.

Mechanically, however, PC is a fucking nightmare. Not because it's actually hard (I've owned the books for twenty years and I bought that hype), but because the presentation is really, really bad. Its not an EASY game, by any means, but it is actually simpler than, say, rolemaster once you start digesting the... quirks.   First of all, it is very information dense.  Let me give you some hard data on a relatively 'simple' weapon, a standard pistol.

You have a column on physical characteristics that includes 8 or 9 entries (depending on how you interpret 'mag').  Those entries are helpfully labeled: L, W, RT, ROF, Cap, AW, KD, SAB... with Mag shoved between AW and KD like a bastard child.  Most of these are numbers, but looking over the pistol entries, I see ROF is often an asterix, and in one case a double asterix, and in another case a number with an asterix. Hmm...  The next block is somewhat easier to parse. Its a double column labeled Aim Time, with two lists of matched numbers labled AC and Md. Pistols generally have six entries, Rifles twelve.  The next block is "Ballistic Data" and has nine or so columns helpfully listed as Range in 2 Yard Hexes, and nine rows (in most cases), the top six broken down into two row sets of bullet types (JHP, FMJ and AP), with one row being Pen the other being DC.  The last two (or three) rows are labled various things. BA and TOF being the most common, MA and 3RB seeming to switch out.

In short there is a lot of data to parse for a single weapon, and a simple weapon at that. Grenades, by the way, have virtually ALL the same information, and a Grenade Launcher actually has two 'ballistic data' blocks... or rather the ballistic data block and the equally information dense, and range dependent explosives block.

Now, in isolation the various numbers can be parsed and fit into a standard resolution system, but the sheer amount of things that have to be parsed and fit is somewhat daunting.  Even though the book itself is only 54 pages of rules and thirty or so unnumbered pages of weapons and resolution charts.

To actually do something, namely shoot someone (since this version doesn't really include non-shooting tasks...), you find out your accuracy based on character skill, aim time, movement penalties and so forth, then you roll a d100 against a chart based on your range to the target.

No, scratch that. You MIGHT roll a d100 against the chart. You MIGHT find that you automatically fail, or automatically hit. While somewhat stasticially invalid (I suppose...), I actually rather like that part. You only roll when the outcome is in doubt, even if that means blowing a motherfucker's head off at a hundred yards.

Damage... now that is a horse of another color.  First you compare PEN (based on ammo, weapon and range) against armor to determing penetration line (one through four), then you (I believe) roll a D10 to determin if glancing damage is inflicted or, in the case of line 4, you overpenetrate.  Then you roll location based on wether the target is in cover or not (you don't hit the cover if they are, you hit 'unprotected' areas instead...), which includes 24 possible locations but does not appear to seperate left from right for arms and legs. I double checked and found no actual rules to seperate left and right arms. Anyway: The damage table covers numbers from 1 to 100,000, and includes (only in the overpenetration portion) several entries for 'Dead', which is convienent. Wounds can do PD or can be critical and/or incapacitating. For example a low pen shot to the forehead does 2000 critical.

HOWEVER: Two pages later there is a bigger chart that cover Effective Penetration up to EPEN 10, which I believe covers the 'advanced rules' which allows for damage up to millions of points.

I will point out now that there are absolutely NO hit points in the system, and I am glossing things greatly.  Essentially, you can rack up millions of points of damage, which accumulate from multiple injuries, and you roll, essentially, a damage save against them to stay alive and fighting. You can be knocked out by trivial wounds or great ones, and you can just outright die.

That is Phoenix Command's system. Living Steel went quite a bit simpler... optionally (their weapon entries, for example, included shaded portions for people who wanted the whole PC monty...), and it included a somewhat bare bones skill system that was largely percentile based for non-combat tasks.  I believe the actual emphasis was less on rolling to succeed than checking to see if you knew what you were doing (I am an engineer, I will build this bridge eventually... when the GM says I do. I am not a radio technician, so comms are down indefinitely... not quiet that simple, but you get my drift, eh?)

Character creation is... interesting.  Somehow Phoenix command gets all the peices into place and then manages to bolo then entire deal.  Phoenix Command gives you five fairly ordinary attributes (strength, intelligence, etc...) then a bunch of derived attributes, to include your initiative chart (four half second impulses in a round, potentially 21 action available per round for the best of the best of the best... more on that soon!).

However there are only two things any decent player will care about once they've learned the basic system. The first is the little entry at the top of the character sheet that says 'Skill Level'. The second is Intelligence. Lemme Explain.

Strength and your encumbance are cross checked to give you a basic speed. Since the only things that give you encumbrance are things you use in a firefight (not stuff in your backpack, its assumed dropped) you pretty much have an assault rifle, some armor and some ammo, maybe a few grenades. Your encumbrance won't change much, and you'll only be able to twitch it a column or two anyway. Basic speed is in .5 increments and seems to go up to  4.5. THAT is cross checked against your agility to give your maximum speed.  Got it?  So unless you are incredibly strong, and/or stripped naked your basic speed will generally fall between 1.5 and 3.5. and is far more likely to be within 2-3, and your maximum speed thus will fall between 3 and 8, and those are the outlier numbers. This is purely ground movement speed and you will never use either strength or agility again.  Health and Will do affect you by determining your chances of staying conscious after you've been shot, but then we are talking a system that slams you with potentially millions of points of damage all at once.  No, let me correct that. Will is used. Health only shows up after you've been wounded and are healing... and at that only Half of Will is used.

The single biggest determinant is the character's skill level, and that is determined entirely by the GM. It is assigned. It covers your accuracy in combat, it covers your KV value (Knockout save. Not half, like Will, but full value), and it determines your actions per round.

The most amusing thing in the game is determining your SAL value. There is a chart for this. Chart 1C in fact.  Do you know what is in Chart 1C? Your skill level. You could just as simply say your SAL is equal to your Skill Level+5, except for skill0 and 1, and been done with it. Literally, that is all there is to it, but god damn it, there is a fucking chart. A one column chart.  

Anyway: Your SAL is your accuracy determinant. Added to your intelligence it is also your actions per round. Eh. It IS cross referenced against your Maximum Speed, but at this point you need to recall that you have very little direct influence on your maximum speed, as a range of 3-18 strength actually only produces a likely range of three or four steps of movement (On the extraordinarily rare chance you get a 3 strength on a 3d6 roll, you could, in theory have 1 bs, and on the off chance you both get an 18 strength AND you go into combat with nothing but some cutoff bootyshorts and a pen knife (literally ten poinds of encumbrance) you CAN get a 4 Base speed...), and THAT is then further reduced by the agility comparison, the straight addition of INT to SAL makes INT by FAR the strongest player determined factor in actions per round... and that, in turn, entirely governs almost everything, like how fast you are on the map and how many half second impulses it takes to complete your aim actions before you shoot and so forth. Yes, MS is important, but it requires two attributes and a keen appreciation for the value of rambo combat to even think about maximizing.  No, its not as broken as, say Dream Pod nine and Agility (used for accuracy and defense, and accuracy is a direct damage multiplier. Strength on the other hand is only used for melee damage, is a direct point for point upgrade, and requires raising two attributes, since its averaged... Hmm...), but it is an obvious flaw, showing a certain attribute bias...

Living Steel ups the number of attributes and takes the GM assigned skill value off the table, both of which are nice.  The increased attributes are largely social, but there are some nice additions to derivative characteristics.

However there is an interesting flaw in the method of 'rolling up a character'. LIving Steel posits a fixed time line of events leading up to teh campaign, the characters are 'dead' war heroes who have been stored 'on ice' for the right moment to restore the glory that was the Sword Worlds, the 'Dream' of the Sword worlds (which isn't terribly well explained, but we can presume an acceptable to modern democratic citizens meritocracy with strong concepts of individual rights and honor, as compared to the better detailed 'bad guys' teh Star League, which is a ruthless and cynical caste based noble/corporate syndicate...).  You essentially roll your pre-heroic days on one simple chart (including options for being a former star league soldier, which is a nice touch), then several steps down the 'lifepath' chart where you work out your various military background (or, if you are 'unlucky' you were a civilian, with an assload of those sketchy civilian skills I mentioned earlier and shit for combat.), along the way earning decimal point skills in five catagories, and a chance at the four 'binary' skills.  I love the binary skills concept by the way: You either can use lasers, or you can't. You either can use grav vehicles, or you can't. I don't necessarily buy their choices of binary skills (lasers, grav vehicles, power armor and... um... something else I think....damnit! Where's my book???!!!), but I like the idea and it does work.  There are two major spots on the path that are notable for being your 'heroic choice' moments, where you becoming 'living steel' or become a civilian psychic (or later, even more heroic living steel or... psychic living steel...). I will note that the psychic stuff is almost entirely notional. Off the top of my head you get some 'vir points' which... at best, work a tiny bit like weak action points from any number of games, buying rerolls or something. I suspect they planned more but never got around to it (in teh power armor book, for example, you can have  'vir battery' module installed, giving you extra vir points. Woohoo! What do they do again?).

Anyway, the last entry in the chart is how you 'died' , or rather which 'final battle' you died in. Its actually a pretty decent way to introduce and cement the world history into the player's minds. I still could tell you about the Fall of Hyrken and the Final Stand on Alpha just from making a few characters (for actual play and for fun...) years ago.

There is, however, an interesting quirk to the system.   The entries in any given part of the path are not remotely equal. THere are alway clearly superior and clearly inferior choices, which forces you to do it by random rolls. Becoming a psychic soldier is clearly superior to just being an ordinary damn hero... as in: In addition to gaining the psychic power shit you ALSO get much better skill ups across the board.

Anyway: The five combat skills translate more or less directly into various sub-systems of the greater phoenix command line of books.  The A column, for example, gives you your shooting skill (as detailed in the review), while the B column gives you your melee skill (as detailed in another book which I have but did not bother to review here...), while your C column is basic combat driving (I believe...), and another column covers your generic athletic skills (climbing shit and so on...).  Its actually a decent implementation, and the setting is good.  I absolutely adore their power armor in a way that more fantastic versions (SPace Marines, fer ex), fails to capture (my inner engineer coming out, or something...).

I'm going to go out on a limb and try a bit of post description editorializing on you, a first for this thread. If I (or you) don't like it, I won't do it again.

Living Steel I would recommend to anyone interested in a sci-fi post-apoc senario with reasonable high tech (and balls out awesome power armor), who don't mind some serious crunchy rules and a bit of wiggly/restrictive character creation.  

Phoenix Command I highly recommend for RPG flavored tactical military play, if you can find it, or as a crunch enhancer to Living Steel. Played straight, you roll 3d6 in order for stats, so the extra weight for Int is pretty trivial (and the chart is brutal...). I would recommend that, for most cases, the damage rules be more handwaved.  Since hit locations are pretty regimented you can just rule a heart shot is fatal rather than parsing KV rolls over 3k Critical damage... which will speed up game play a lot without actually increasing or decreasing lethality much.  If crunch or lethality aren't your thing, avoid like the plague. (note: The main PC book is focused on military style play, but there is a wild west supplement (like: Three pages of rules and 5-10 of weapons..), a tank supplement, a melee supplement that is good for historical play with a little effort, a faaking artillery supplement (as I recall. I regret not picking it up now...) and so on.  Obviously this is an older game, where non-combat gets short shrift, but most of us survived the eighties gaming scene just fine, I'm sure the 'damn kids on my lawn' will figure it out too.

:cheerleader:

Interesting stuff...DH and Phoenix Command are both games I'm unfamiliar with. (Though I have another thing by Nakazono, Swords's Path Glory, and its pretty freaky.)

Minor question if you don't mind: how do the 'Unnatural Attributes' work in Dark Heresy ?

Swords Path Glory sounds like a 'Red Dragons' supplement for Living Steel. Each of the seven swords gets a color, a thematic goal (glory for Red), and a nice symbol (Coincidentally, the one for teh White Swords (the pyschics) looks an awful lot like the diamond for Skyrim... hmm...)

Black is the path of battle, Blue was duty, Green was... loyalty?  Red was glory. White was the Kami, and I'm not entirely clear if they had the same sort of path. Gold was... um... damnit...


Anyway: Unnatural attributes in Dark Heresy:

So many of the attributes in Dark Heresy provide a fixed numeric bonus based on the tens digit of the attribute. So a strength of 45 provides a 'strength bonus' of 4 for the purposes of melee damage. Agility bonus covers initiative mostly, and toughness is damage resistance, Will Power governs various psychic stuff (how many psychic powers you unlock at each tier of power, for example).  I presume that this fixed bonus can be extrapolated to any number of situations by a canny GM, but I won't swear that that's actually in the book.

Anyway: Unnatural Attributes essentially act as multipliers for that fixed bonus, starting at x2, then x3 and so forth. So a Toughness of forty three remains a forty three for any percentile rolls based on toughness, but when using the toughness bonus (for example: Resisting damage) you would resist 8 or 12 points (or so on) instead.  There are times when that advantage can be negated... for example warp based bonuses (demonic or psychic) can be negated by any weapon with the Holy trait, there are a tiny handful of weapons that specifically ignore a set number of unnatural toughness bonuses, depending on which game line you are referencing... as they've evolved a few edge case rules to work better and included new toys.

Anyway: While I don't believe there is a canon use for, say, Unnatural Fellowship (charisma), we could postulate both a 'social damage' effect vs Will Power, a limiter on how many people can be persuaded personally with a single check and so forth.

Weird that it boosts some stat effects (those using the modifier) and not others (the %)- the idea reminds me of Rifts' "Supernatural Strength" except that Rifts' ability check system generally isn't complete enough to actually have problems from that...yeah...well, its unique.

Sword's Path Glory probably sounds more exciting than it is. Lots and lots of tables, with charts to calculate assorted values like Base Acceleration Increment and parry %, then loads of hit location charts by weapon type and body area, where you cross-reference Effective Impact Damage with armour type worn to get flesh depth penetrated and so shock/damage points. No fluff text in there, unless you count the combat example.

Well, the only thing it doesn't boost is the percentage checks, which is a good thing if a bit counter-intuitive.  Now true: Strength and Agility and all that don't actually influence combat accuracy (Those are Weapon Skill and Ballistic Skill attributes, which don't get 'unnaturallized'), but the idea is to keep all 'skill checks' within a functional range.

So eldar can be inhumanly fast and graceful (unnatural Agility x2) without negating the fact that they still need to learn to dance like everyone else, or what have you.... since their 100+ potential Agilty doesn't negate the unskilled penality or standard probabilities, because its really a 'human normal' 48 or something with an 8 bonus instead.

One of the issues I have in doing this sort of comprhensive analysis is simply choice. I am spoiled for choice. What system should I do next?  Since I went 'exotic' with Phoenix Command, should I do something more 'mainstream' next? Or can I follow up with, say, Anima: Beyond Fantasy?  Even by a quick scan of my bookshelves I have enough material for at least 15 or so posts, even with just the very lose groupings I've already done.  But perhaps I'm boasting. Well, if I stop posting, you'll know its geek hubris which done me in, pa...


So: Shadowrun:

Mechanically the first three editions of Shadowrun are essentially identical with minor tweaks. The fourth edition introduced a major systemic overhaul that virtually eliminated cross compatability, but didn't fundamentally change the nature of the game. THis is comparable to the 3E overhaul, with the attendant fan bitching from old grogs that was mostly swallowed up by the fact that the new system simply worked better. Mostly.  

Shadowrun is and was the first significant Dice Pool system, using d6s. I understand one of the major designers then went on to help design White Wolf, and the similarities in presentation are worthwhile as well. Both game lines present character choices (Archetypes for Shadowrun (Pregens), and Factional choices for WoD) with large full page+ splash pages with large character drawings. This is, to be blunt, a brilliant marketing move that really helps draw in players, and is something that should be emulated more than it is.  If you're gonna spend on art, spend it on something that will spur player imagination and allow them to really 'get' their character's place in the setting, rather than on brooding NPCs doing NPC stuff.

Anyway: Earlier editions of Shadowrun rolled smallish handfuls of dice and simply took the largest single die out of the pile. THere was a rerolling mechanism so that d6's could hit higher numbers than 6, which largely negated the speed value of this system and made calculating odds of success truly wonky (TN 6 and 7, for example, are identical mechanically...). Fourth edition counts every die of 4 and up as a success, and counts successes. All modifications are to dice pools, which vastly sped up game play, though the soft cap on dice pools is significantly higher than optimum (22-24 dice is easy to reach, very hard to surpass for most actions). The Math is very simple, and for some tasks you are encouraged to exchange dice at a slight disadvantage for successes rather than waste time rolling.  Hit points are fixed with a success based damage/resistance mechanism, though in fourth edition the fixed hit points are determined during character creation, an additional bit of fiddly-ness that doesn't seem to improve the game significantly.  Still, it is front loaded, so the imposition is minor.  It does lead to the loss of a signature Shadowrun feature: Scalar damage codes.  In prior editions, a weapon's lethality was obvious at a glance. This weapon did serious damage, that one medium and for a few rare weapons... Deadly.  Now the lethality is strictly found in the Power rating (which weapons always had). This has led to a problem.  Before you could have a nastily lethal weapon that was not 'broken' in terms of mechanics, but under the new mechanics, the only way to get more lethal is to raise power higher, which is obviously a problem, as Power also governs resistance and other things. I'm doing a poor job explaining, so instead I will illustrate via the 'Troll Bow'. Now there are two problems with the Troll bow, the fact that power is raised linearly by strength in a bow, unlike every other weapon in the game, and two, a troll's strength rating can easily beat the power levels of even heavy weapons. While usually used to illustrate what is broken about the bow's damage mechanic, it also serves to illustrate the problems with simply raising the power level: Troll Bows are known to shoot down helicopters! The simple addition of a few points of Power over the carefully monitored (and slightly flat) power raises of more conventional weapons makes a simple bow better than man-portable artillery. Previously this wasn't an issue, where a 20mm explosive shell was a deadly weapon, while an arrow was a moderate weapon, regardless of power ratings.  A case of simplification gone too far, and unnecessary except for the fact that they changed the health mechanic.

Character creation has always been a mess in shadowrun. THey have always had pregenerated characters, but in almost every edition those, or at least a significant number of them, did not add up in the character creation system.. being either too weak or too powerful.  Earlier editions used a simple priorty mechanic to give a fixed number of catagory points (similar yet distinct from teh WoD mechanic already addressed. The WoD mechanic is the more refined one, I believe).  In 3E shadowrun they introduced an optional point buy system, and in 4E shadowrun that point buy became THE system, only its a bit of a mess.  You spend four hundred points (which is a lot), that are unrelated to character advancement points incidentally. Curiously: Almost every single point value (excepting starting wealth) seems divisible by four. With a ridiculously small number of tweaks you could have divided the entire system by four, spent 100 points and greatly streamlined the process for most players. Attributes, for the first time, became universally important parts of the character, rather than near useless appendixes used only to generate more important stats barring a handful of resistance rolls (health loss), which, while appreciated, does lead to the explosion in dice pool sizes.

Curiously: A pre 4e character should be directly importable to 4e, since the characters look the same (a minor adjustment of skills may be necessary) except for the addition of two stats, which could be done by cloning the newly divided stat, since the majority of the changes are to resolution. A six strength is a six strength on either side of the change.  

However we do come to a major beef I've always had with Shadowrun (at least after my first major campaign): while characters grow via karma Points, true growth is measured largely in Money.  That's a bit glib.

Power on the character sheet comes from Cybernetics and 'rare' equipment (Panther assault cannons, heavy armor if you want it...) OR it comes from your collection of magic (which costs money AND Karma), more than any other factor.

Much like a high end D&D character you cease being a character and become a collection of powerful items and spells.

This doesn't strike me as morally or factually wrong: After all I'm reasonably sure that real world militaries care more about the numbers of riflemen on the ground than they do about wether or not Joe skipped his last two ranges, or whatever. Gear matters, and beyond a relatively minor variance in ability, it winds up being most of what matters (discussions about morale are welcome...).

For an RPG, however, I think that focus is misplaced.  The skills of Tom Berenger and Billy Zane, and their personalities, are what matters in Sniper far more than wether or not they've been issued APDS ammo (a Shadowrunner's favorite) or not. They don't spend any time at all talking about what model sniper rifle they have... its a tool they need to do their job, one anyone can get. Its how they use it... their, their PCs if you will... that matter. That is how RPGs should focus down: On the character rather than their equipment.  Shadowrun has all the elements of important charaters, but the emphasis winds up being on the tools. A mundane character just can't compete with a cybernetic beast or his enchanted spell flinging pal.  This problem surfaced early, when spell casters started getting magical equivilents to certain cybernetic implants (notably the initiative boosts) that they could sustain for long periods of time.

Shadowrun is still a good game, though not without flaws.  I personally leap at chances to play it, but I won't buy any more books until I'm sure I'm not feeding theives in return for crap products... but then, the good stuff came out early, so I'm pretty set.

Just a short "thank you". This is really interesting. I am looking forward to the next instalment(s).

Some thoughts on character generation:

I tend to view any sort of system as a makeshift compared with simply modelling whatever conception one has in mind, in consultation with the GM. The latter is how I have always run Marvel Super Heroes, despite having previously both "rolled up" characters in V&V and used "point buy" in Champions (and a mix in Superworld).

That said, sometimes it's fun to go with what chance happens to provide. It can be a stimulus to the imagination, as well as more realistic to the extent that one steps into the shoes of a persona with no say over things that he or she really could not control.

Classic Traveller perhaps goes too far in randomizing careers, but one might rationalize that (as one rationalizes the results) by making inferences about the imagined universe.

One difficulty in CT is that the 'advanced' systems (in Book 4 and on) tend to provide more and higher skill levels, as well as making available skills not on the tables for 'basic' generation.

I got around that with a rule allowing skill substitutions. That was originally conceived because it seemed odd that the sciences were not represented in an SF game, as well as that there should be so many computer programmers. Simple solution: allow the player to trade out Computer skill for a scientific (or other academic) field. This got extended to give access to skills from later Books.

Someone else came up with a quick method to give characters skill levels roughly similar to those produced in the 'advanced' systems.

I have also found that it can work well and quickly to distribute a given allotment of points among ability scores and skills, choosing whatever seems appropriate.

I thought Living Steel had fun potential, but couldn't find interested players. I ended up donating (literally) both that and Phoenix Command to a local shop that had game tables, hoping that somebody might get some use out of them.

The Tri Tac system (Fringeworthy, FTL:2448, Stalking the Night Fantastic) was overload enough of all that "gun nut" stuff for Yours Truly, who is contentedly ignorant of most of it!

Quote from: Spike;643367One of the issues I have in doing this sort of comprhensive analysis is simply choice. I am spoiled for choice. What system should I do next?

How about RuneQuest/BRP/Legend? RQ is one of the oldest generic rules systems around and BRP/Legend are offshoots of RQ.

Hope this is still going?!

I'd up-vote Savage Worlds or Palladium for review I guess, but whatever really.