Critique of Eschalon - Learning Cliffs, and real strategy
Posted: April 2nd, 2013, 1:01 pm
My apologies to anyone on the "TL;DR crowd", but this will be a long one. What I am about to write, however, is not a simple statement, but rather a persuasive argument for a different perspective on this game, and arguing for a different perspective unfortunately takes far more time and verbosity. (Plus, due to the fact that it took me a week to get registered here, I kind of started writing what I wanted to say in notepad, and never particularly stopped...)
This thread is a review/critique of Eschalon: Book I primarily, but I'm putting it here, since I am critiquing the mechanics of the game that a trip to the wiki tells me are largely unchanging throughout the series, and hence, are applicable to Book II, and (probably) Book III, as well. Since Basilisk Wrangler seems to actively read and participate in these forums, I'll largely address this thread to him. (But that is not to say I am excluding anyone from the conversation.)
I am someone who might be called a connoisseur of strategy and RPG games. I bought Eschalon through Steam because I love game mechanics, and love sampling many small indie games for their mechanics before either "digesting" a game (in a superfluously loquacious manner such as this), or just moving on to the next game.
I have read some of this forum's threads, and see that people have complaints about the game, but that they aren't quite arriving at the conclusion to which I have arrived. Some people say this game is "too easy", and others "too hard", but there is a lack of the proper framing of the debate (a lack of the proper terms to define what the problem actually is so that everyone can recognize the problem for what it really is) to make sense of why this is the case.
In this game, I see a main problem, and two smaller problems that serve to obfuscate the main problem:
Problem 1: The Learning Cliff
Simply put, the "Learning Cliff" is the key reason why many newer players think this game is "too hard" when the game is, in fact, too easy. (This is why it is a minor problem that obfuscates the larger problems.)
The simplest way to explain what the Learning Cliff is, I'll say this:
I play Dwarf Fortress.
Dwarf Fortress is frequently described as an "extremely hard" game, and extremely frequently as a "fiendishly complex" game. The second of which is almost true. In truth, the main reason why so many people have trouble with Dwarf Fortress (aside from general interface shittiness) is that the game basically just dumps you out in a field with an interface filled with hundreds of buttons, no manual, no tutorial, and tells you to just do your best to survive, and that, "Losing is Fun". (In other words, keep trying until you get it right.)
In truth, DF has a very simple formula to solving basically every challenge the game will throw at you: excavate an easily-defended area, seal all dangers off with walls, and get a farm running to supply you with infinite resources. After you have established a basic defensible position, (which takes an experienced player only minutes,) you have overcome basically every challenge in the game.
Most players, however, fail in their games not because they are stupid, unskilled, or otherwise incompetent, but because they are simply flooded with tons of junk data and no capacity to make sense of how to figure out what was the actually important information or priorities in the midst of a sea of bloat data. They start worrying about forges or crafting trade goods when only farming is really important.
Comparatively, a regular game has a "learning curve" - Each element of the game is introduced one at a time, explained to the player, and then tested. You start with a little skill, and only need a little skill to advance. Then, you need gradually more and more skill.
This game has a learning cliff - the game's rules are actually very simple, and winning is very easy, but the problem is that players are forced to start out in a character screen that demands they allocate character points at a time when they have no idea how any of the skills they are deciding on buying or not will actually work.
Rather than being a slope, you need to know a ton of information, have many skills, to make any headway on the game at all, but as soon as you do, you have all the tools to conquer every single challenge in the game - it's a flat plateau of difficulty from there on out, and the game is "easy".
This is why IJBall "fails" twice at the game, only to find it "too easy" on his third and every successive playthrough.
Put simply, the game isn't hard, the players just aren't succeeding because you aren't telling them the rules fairly and simply before you ask them to start playing. The only "difficulty" in the game arises from players simply not knowing the rules, and "skill" is defined solely by whether you understand what you are doing or not.
You already tell players who are looking at spells, for example, how much damage fire bolt does, and deep freeze does, (1d3+2 and 1d4+4, respectively,) when they are looking at them in the game so that they can actually compare these spells situationally. (Twice the damage versus three times the cost.) However, you seem to consider it a "spoiler" to tell players that they gain massively more MP regen benefits for the same flat cost the more points they put into Perception, which is something they really need to know in order to make proper decisions.
There is a relatively simple fix for this problem - simply tell players the rules. Give players more information on the character creation screen about what each skill actually does, mechanically. D&D, for example, doesn't consider it a "spoiler" to tell you that a given feat or skill will give you a +1 bonus to a specific list of situations, or how to-hit calculations are made. Tell players fairly and simply what a single point (or rather, every two points) of weapon skill will actually buy a player for their skill points, and then let them compare that to what the exact mechanics of spending their points in other skills will be.
You also need to tell players what challenges they will face (I.E. monsters, and how they will attack you, or in other words, the combat formula, as well as traps and locks,) and then what skills there are that you will need, and what general focus you will need in them to overcome those problems. Critically important is that you list the alternatives to solving problems, because choices are all about opportunity costs. The alternatives to picking a chest's lock (rogue skills), for example, are bashing it (fighter skills) or casting lock melt (wizard skill). Knowing this, and how much focus they have to put into any of the given "paths" in this game they will need to overcome an obstacle, they can actually start making strategic choices about how to build their character even on their first playthrough.
You claim outright on your advertisements for the game that this is a game that demands "strategy" on the part of the players, and that stats are everything... But the basis of strategy is the capacity to predict the ramifications of your choices several moves in advance, and compare opportunity costs rationally. You're asking players to make the most important choice they will ever make in this game completely blind and guessing the first time they play (unless they "cheat" by reading the actual rules on the wiki).
By setting up a learning cliff, you cripple your ability to achieve your game's own stated goal of being "strategic".
For my part, I succeeded in my first game of Dwarf Fortress and am succeeding in my first game of Eschalon for the same reason: When I got to the character creation screen, I stopped and read the forum guides/wiki until I actually understood what decision I was making, rather than making a blind guess. (Again, I do this because I am driven by a love of understanding the underlying mechanics, so stopping to just read the manual is, itself, one of the most enjoyable things about a game for me.)
For reference, currently, I have (as of the writing of this - I didn't realize it would take a week to get registered to these forums, so I just wrote down what I was thinking as I was thinking it,) just gotten to Blackwater in Book I, and am level 5. I first went exploring south to get that sextant so I could train up my cartography skill after I already got the reveal map spell that meant I wouldn't even need the skill in the first place. (I completed the "bar of mithril" quest without any mapping feature at all, incidentally.) I was playing a "pure mage", but now am playing a "sage" (spellcaster with both arcane and divine magic, but no melee capacity), and went to Blackwater early to buy up the "free" skill points from trainers, going through Bordertown to get there. In my haste to get to Blackwater, I came across a lot of thugs that were easily dispatched with precisely 3 level 5 firebolts (average 45 damage against a 40-hp enemy), and was only slightly troubled when I came across a mercenary character, and was forced to use my first potion - a haste I potion that was dirt cheap from the Alchemy trainer. I kited the mercenary to death. Now that I no longer need it, I am going through Grimhold mostly so I can loot it. After that, I'll maybe consider starting on the plot missions, unless there's something else I feel like looting, first. Stat-wise, I am a druid "conjurer" that started with 14 in Intelligence and Perception, and immediately min-maxed Perception to 40 (the extra 2 points of stats for the extra MP per level), and then Intelligence to 20, and will make Wisdom up to 15 as soon as I get levels to do so, so that I can just cast haste myself. MP is not a problem as long as I hide and kite enemies one at a time (just walk back and forth for MP), and there is no reason I couldn't do so in this game.
This, in turn, brings me to the next obvious problem:
Problem 2: Enforced Min-Maxing
To explain this, let me start with a bit of a metaphor...
The most regular complaint people have about BioWare games (at least, before ME3...) was that it was supposedly all about making "tough choices" that were, actually, boring and routine and took no thinking at all. That is, everything boiled down to a "Good" or "Evil" choice where you were rewarded more the more "Good" choices you took as soon as you started traveling down the "Good" path at all. You actually received more rewards the more "Good" you were, so there was utterly no "temptation" to be evil other than "for the lulz". Hence, this series of "tough choices" ultimately boiled down to a single choice made right at the very start of character creation, and then sitting through dialogue trees that constantly congratulated/berated you on this choice for an entire game.
BioWare enforced the choices you made - they created only a single goal, and created only a single route to achieving that goal.
Many games that want to encourage diverse playstyles, and as such, when they give you choices of skills, they have diminishing return upon investment. Games that enforce min-maxing, however, have increasing return upon investment.
Eschalon has enforced min-maxing. That is to say, not only is min-maxing generally a possible way to abuse and beat the game, it's actually the only way to play the game. It has what 3rd ed. D&D min-maxers call "Multiple Attribute Dependency" (MAD) if you try to mix your "classes" of characters - trying to be a wizard or cleric means you need tons of Perception, and don't have points for Strength... but by the time you have that much Perception, you might as well just go full-caster, anyway. In order to be good at bows, you need tons of dexterity, but once you have that, you might as well just use bows all the time, and get the Dex-based rogue skills while you're at it.
Basically put, if an attribute or skill is ever worth putting one point into, (because, opportunity cost-wise, it provides the most benefits compared to any alternative,) then it's basically guaranteed to be worth putting all the rest of the points you have into, as well. The only reason anyone wouldn't min-max is because they're indecisive, which is probably because they are "hedging their bets" due to not understanding the ramifications of their choices because of the Learning Cliff discussed previously.
In fact, the only reason to ever stop putting points into skills is that they often scale horribly. The difference between 36 Perception as a druid and 38 Perception while above ground is tripling the rate of MP regen in Book I, while 33 to 36 is doubling, and 30 to 33 is merely one-and-a-half-times. (Again, increased return upon investment.) But then, you slam into the wall and arbitrarily gain no more serious benefit from Perception so far as MP regen works. (This formula needs serious revision, as it is mechanically backwards.) There is no benefit to many skills over a certain point, while others scale infinitely.
Worst of all, you make the game have "absolute" evasion percentages, (magic resistance and dodging, for example,) which is basically anathema to "strategic" gameplay, since they devolve the game into nothing but a crapshoot of waiting until you get a lucky string of rolls. (Or worse, reloading until you get a lucky roll, because, strategically, the only thing you did "wrong" was to not get a good roll.) Likewise, consider that the air shield spell is basically nothing less than an "invincibility versus archers" spell costing 2 MP and lasting 20 turns. (This deserves its own section to deal with as a problem, however, and is beside the main point of what is most seriously flawed in the game, so I'll leave that problem for another time.)
This, in turn, leads to characters being "One Trick Ponies" that have one move that beats everything the game throws at them.
There are several solutions to this problem. First, you need to start by rebuilding your formulas to scale properly. Use less addition and subtraction, and start using more multiplication or division. (I'll go into this more in a later post.) Next, you need to reformulate skill point costs, and (if you keep linear returns upon investment, like +1 To-Hit every couple weapon skill ranks,) introduce sub-geometric or geometric growth rates on the costs of skills while increasing total skill points per level over levels, similar to how Avernum works. Likewise, similar to Avernum, make the most useful skills cost more than the utility skills. There's no reason skills like lore or spot hidden or cartography should be equal to arcane skills, especially since there are spells that obviate the whole need to use two of those other skills in the first place.
Alternately, you can ditch skill points altogether, and, like Elona, Dwarf Fortress, or The Elder Scrolls, just use "train with use" mechanics. It also obviates the need to "fix" trainers, since you can simply make trainers into people that just let you have, say, the experience of 50 bow shots for 100 gp at the archery range rather than having to grind out that many bow shots normally.
Your entire "class system" itself is something of a misnomer, as well, and you might want to look into cribbing some notes from games like Elona or World of Darkness, and introduce "Potential" or "Affinity" for skills rather than outright classes. (I have tables I can share on this... but that is going off-topic, however, so I'll save it for another post.)
However, these last two problems were just the obfuscating problems that overshadowed what was truly the cause of why the game is so "easy" it's actually boring, and players have to come up with "challenges" just to keep things interesting:
Problem 3: Overly Simplified Rules Creating A "Solved Game"
The reason why there is nothing but a "learning cliff" that turns into a plateau is that all challenges in the game require exactly the same response - there's no "curve" or "slope" past the initial "challenge" of not knowing what the rules of the game are when creating my character.
Functionally speaking, to my mage character, every melee enemy in the game is just a fanged salamander with more hit points. If they resist magic, that's just making combat a bit more of a crap shoot, and maybe I have to use haste to kite the target, or at worst, reload and try again and hope I'm not as unlucky, or just use fleshboil. There's no further strategic thought that has to go into "this enemy is completely immune to the character build this game forced me into min-maxing my character into being, then schizophrenically decided min-maxing was a bad thing, and made immune to elemental magic," that just means you need to use fleshboil, instead. Air shield means I am invincible against archers. Element shield is basically the same thing versus wizards (with enough Intelligence and the rank 6 spell, that's 99% resist, easy).
Mages have the greatest diversity of tactics at their disposal, at that - melee characters are basically faced with always exactly the same "strategy" for everything. That is, just keep punching it until either it dies, or you die.
Yes, you do have some alternative options as a melee fighter, but - and this is the important part - those options have nothing to do with the fact that you are a melee fighter. Those options are almost always either spacial problems or else consumables, available to all characters. So far as melee fighter is concerned, skill-wise, the difference between a "hard" and an "easy" fight is just how big the numbers are, and how many clicks on the icon it takes to make the HPs become 0 and make the peoples fall down.
You also have to realize that, compared to games like Avernum (or D&D), you are severely limiting the number of tactical options a player has by simply having only one "unit on the board" at a time. By comparison to Avernum, (which has very similar character-building mechanics,) you actually have less options for good moves at a time by an order of magnitude for every individual unit you don't have in your party - in a game where every action takes up equal amounts of time, basically all decisions towards achieving a single clear goal - killing the enemy as fast as possible. Because so many of the skills for killing things operate along similar lines in this game, (you are basically just comparing DPS, since there are no situational bonuses or penalties in this game, barring light penalties,) they become little more than "calculations" - what is the most DPS you can get with a move?
To give a far more in-depth example of how much tactical flexibility some really good RPGs can give a player when they advertise many strategic options for solving problems, I'd like to point to the Russian-made game Hammer and Sickle (a spin-off of Silent Storm, and somewhat similar to X-Com if it had far more focus on fewer characters with deeper skill growth). In one event in that game, the villains have taken over the hideout you could have previously used, and are lying in ambush for you. In my game as a sniper character (with a gunner and grenadier support characters), I responded to this by crawling in one of the windows of an abandoned house where one of the enemy snipers was lying in wait for me, and stealth killing him with my sneaky sniper, then crawling through the darkness to snipe one of the other snipers from another house across a narrow alleyway. I started picking off snipers while moving from window to window to keep from being spotted - popping up and letting of a blast before moving on - and some of the incoming "sweeper patrols" that moved as a unit to pick off hiding units like my sniper main character based on the sound of my rifle.
I was on the second floor of this building, and I had my supporting units behind me, watching the stairs. I guess the sweepers that got into my building somehow heard my supporting units, because they weren't willing to charge up the stairwell all nice and single-file so my waiting gunners could mow them down.
It was an impasse - they weren't willing to come up, and I sure couldn't go down those stairs. I could try jumping out a window, but that would expose me in the streets, and there were still other sweepers out there. Then I noticed that parts of the floor were weak, and a thought occurred to me - I still had some C4 left. I planted it down on top of where I could hear the sweepers were on the floor below (facing the stairwell), and then backed up, and detonated it. The destructible environment collapsed, and the explosion stunned the sweepers, at which point I had my grenadier pop up and chuck a grenade down on them before they could recover and run. He and my gunner then hosed the survivors down with their assault rifles.
Compare this to a game that devolves into what essentially amounts to just throwing dice until one side or the other runs out of hit points, and I think you can see which game can offer up a greater selection of strategic choices. (Worth emphasizing is that in Hammer and Sickle, even the enemies were much smarter about how they fought, and forced the player to fight smarter because of that.) Even if you take examples just from D&D, then you can have defensive stances, tripping attacks, disarming attacks, charges, non-lethal attacks, grapples, and more.
Even in single-character games, you can still look at games like Roguelikes, which tend to have many alternatives and very strange, even gimmicky, items, enemies, and obstacles to exploit.
For example, to give weapon skills more diversity in Book II, feats were added... but there's no real tactical choice in using feats, so they keep combat boring - feats are basically always better than a normal attack, have no real downside, and are just "you can use this once per battle" skills. If, however, you had more than one feat to choose from, but could use only one feat per battle, then you start having real tactical choices from the fighter class.
Before I go further on this, I need to define some more terms so that I can be clearly understood, so I want to link two videos from the excellent gaming-analysis group, Extra Credits:
Choice and Conflict - Helps define "real choice" in games, as opposed to "calculations", as well as defining how "incomparables" are key to creating serious choices, and setting up conflicting goals is key to creating strategic choices.
Perfect Imbalance - Less necessary than the first, but this helps set up a better understanding of the sort of "incomparables" that a turn-based game that focuses upon "strategic" choices really needs to deal in. That is, anything that uses the same types of stats are easily compared. Comparing things that have a perfect imbalance, such as stealth bonuses versus bonuses to loot gained if you sneak around enemies creates serious conflict of goals for players of stealthy characters.
To give an example of incomparables, before bringing this back to the main thrust of the argument, look at the issue of weapon weights, for a second. All of your weapons (and materials they are made of) are completely linear in function, not creating any sort of conflict between differing goals. You missed an obvious chance to make a balance of incomparables with your weapons when you went and made ALL the weapon materials successively heavier than any material "less powerful" than the given material - mithril, one of the best weapon materials - is heavier than steel, and functionally virtually identical to adamantium. Even putting aside the whole fact that the whole defining feature of mithril was that it was lighter than steel, it's missing a perfectly good opportunity to make for an actually interesting choice when it comes to weapons: Do you take a stronger weapon, or a lighter weapon? Even if we don't have any more advanced mechanic, like initiative that gets partially determined by weapon weight, just the fact that a magic user can't cast with a weapon heavier than 1/3rd their Strength alone makes light weapons very attractive to magic users. Why didn't you use this existing mechanic to set up an actual strategic choice for the player to consider - using a slightly weaker weapon in exchange for not needing as much Strength, so that they could spend their attribute points elsewhere, probably buffing that magic casting so that they could buff their melee capacity to compensate?
So, to pull this all together for the main argument I am making, this game is "easy" to anyone past the learning cliff, and nowhere near as strategic as it claims, or, for a turn-based game, (where skill is necessarily found in strategic thought, since there is no dexterous skill particularly required,) needs to be in order to be enjoyable. There is only one real challenge in the game - combat - and in that case, there is only one real goal (defeating the enemy), which does not create a tension of conflicting goals, and all tactical choices are comparables; It is an easy measurement of what actions remove the threat of an enemy most quickly, reducing all combat quickly to routine.
Differences in "classes" are true incomparables, and the true choices that separate out how you play one game from another. However, you are kidding yourself if you are saying there are "unlimited character choices", because there are really only 4 choices you can make to solve the only real problem you face in the game (monsters - doors and chests don't require any real strategy and can be skipped) and that is melee weapon (with choice of exact weapon a quibbling difference with little in-game incomparability), ranged weapon, magic (with differences of wizard/cleric being fairly minor), and hiding/running away. You can try to split the difference on these, but except for the "running away" choice, they just don't mix well because of MAD splitting off your attribute points until you are already so well-leveled (because you've been relying upon a "pure" version of a character) that you have enough spare points to be able to splurge on gaining a second specialization. (Marginal exception for cleric-warriors, since taking points out of Str and Dex can be compensated for with "buffs".)
To go back to how fighters have so few tactical options that are actually related to being a melee fighter, let's take a chance to analyze how important consumable items (and classic D&D Vancian Magic) are in RPGs in creating conflicts of goals:
Potions are a perfect object for creating a conflict of a player's short-term and long-term goals. That is, a potion of haste will always be a massive help in any fight... however, then comes that nagging thought. The thought that you can't escape - BUT YOU MIGHT NEED IT LATER. THAT's a conflict of goals, and what creates real strategic choice in a bottle. (And what's the "most powerful" pair of skills in the game? The alchemy skill and ingredient-gathering skills that give you all the potions you'll ever need, so you WON'T "need them later"...)
This is the same as what Vancian Magic did to wizards - you only have a couple level 3 spells as a level 5 or 6 wizard... the DM seems to be ready to throw landscape obstacles at you - you can take a Fly spell with you to bypass obstacles... but then, what if the DM sends a bunch of harpies after you? Better take a fireball, instead. But what if it's a magic-immune golem, instead? You'd be better served with a Haste spell to throw on your fighter, who can use it to deal more damage against nearly any enemy. Those are real strategic choices based upon incomparables. (And come on, you're not going to use that fireball against a bunch of kobolds - you only have one of those, and you'll need it later, so just use a lesser Sleep spell, instead.)
Something worth recognizing is that you also have "Metagame" conflicts of goals, where in-game goals are put in conflict with out-of-game goals, which can actually be terrible for a player enjoying the game. The Extra Credits video showed how the in-game most powerful strategy was also the out-of-game least enjoyable strategy, which created a conflict of goals between doing badly in the game, or not having fun... which generally results in player dissatisfaction no matter how they choose. In this game, save-scumming for better loot or constant re-rolls are a similar form of metagame conflict of goals. You are trading the game going quickly and letting you go through the game at an enjoyable pace for the knowledge that you're getting the best possible loot. (In)comparably, you could just accept a ton of trapped chests with nothing but a femur inside, and lose enjoyment in the game because you feel "cheated" out of a larger haul of loot. Either way, I actually wind up enjoying the game more when treasure chests have deterministic loot so that I don't have to actually feel this conflict.
It's not just "cheating" that suffers from this problem, however. Many of the game's more frustrating aspects (and, generally, most of what leaves me dissatisfied about the game,) is that so many of the in-game choices are basically just this same "best way to play" in-game rewards versus "completely tedious" out-of-game punishments are basically the only choices ever presented in this game.
Take Reveal Map, for example - it has an enemy-tracking function that is the only power in the game that lets me see enemies before they see me on its level 6 casting... and there's no reason to cast anything less than the maximum, because the only reason I DON'T use it all the time is that it's such a pain to continuously cast. I'm trading being able to know how and where all the enemies are in the game for it not being such a godawful slog of casts and camps as it already is.
That's even more the case in what is probably the single most egregiously broken aspect of the game: The "walking the dog" problem. You can quite simply just out-walk a velociraptor. Then, take advantage of zone boundaries to firebolt creatures to death from utter safety. (Incidentally, that's how I got my Citizen's Writ in Book II by killing the Red Wolves in Wolfenwood at level 1...) It's how people do things like beat the game on level 1, if that isn't enough proof of how broken this game is because of relying upon dumb-as-bricks AI and purely melee-based enemies with no running mechanic or actual speed stat.
In other words, if the most powerful strategy in the game that will probably defeat even the most powerful and vile monster you will ever see is probably best set to the tune "Yakkity Sax", there's a problem with the way the game's mechanics works.
And it's not just that this makes the game "too easy" - I can enjoy easy games - it's that it basically means that most of the ways that you have tried to make the game "harder" has actually just made the game more tedious. (Killing a taurax is the same as killing a fanged salamander, but with more HP and so many resists that it takes running away and camping for more MP once or twice to finally grind it down.) And it's always a choice in this game - do I want to play it "right", the way that, strategically, I should play the game if I were thinking about the game the way that you SAY I should be thinking about the game, and playing the best I can... or do I want to have any sort of enjoyment at all?
Another problem that I see people have argued over heavily is the problem of how trainers help enforce min-maxing. The problem is, people haven't really framed this debate rather well, so the debate tends to go nowhere.
What trainers really do is set up badly balanced incomparables. That is, strategically speaking, we have limited resources with which to build our characters, and of these resources, skill points (and to a lesser extent, attribute points,) are the most precious. This is because, in strategic choice, opportunity cost is everything. With enough skill points, you could be skilled in every playstyle at once, so there's always something else you could be spending those skill points on.
Meanwhile, however, trainers are an option to reach the same player goal of a greater diversity of skills, but where the resource you spend isn't your precious skill points, but are, instead, your relatively worthless gold. Yes, relatively worthless, because the actual opportunity cost of spending money on training is negligible. You can always get more than enough money for everything you need, especially with the many tricks for getting more money.
To show how this could actually be remedied into an incomparable choice that actually encourages strategic thought: If you were to take one of the already-suggested solutions to trainers, (that you just plain get 5 uses of the trainer per game, regardless of what your actual skill level,) and then mix in the increasing skill point cost per rank (so that, at its simplest, the first level of a skill costs 1 skill point, second level of a skill costs 2 skill points, third level of a skill costs 3 skill points, etc.) idea, then you create a conflict. That is, buying a skill from a trainer for money at skill level 1 will give you the benefit of having spent 2 skill points to get to level 2 in that skill. Meanwhile, buying that skill from a trainer at level 11 will give you the benefit of 12 skill points worth of actual skill increase. That's a "discount" of 10 whole skill points!
But here's the conflict - the longer you wait to buy those "free skill points" by training, the further in the game you will have already have to be, and higher the level you will need to have been.
Hence, it's a conflict of a player's desire to have the most powerful character they can have, statistically, versus the player's desire to actually use those skills in the game - what's the good of having high skill points on paper, if you've waited until the very end of the game to get them, when you obviously didn't need them, and can't even use them anymore, anyway? (Except, possibly, against some type of "bonus boss".)
When designing a game, you need to recognize what each game element actually adds into the game in terms of what its actual choices or player interactions. Every game element, especially in a game that asks players to "think strategically," has to add a greater selection of real incomparables, and you need to actually be able to understand what choices those incomparables are making players select between from even a "metagame" standpoint.
I have a set of possible solutions to these problems I have laid out (especially the last one), but I would like to first see the response to all that I've written so far. This isn't even starting to get into the problems I see in the magic system, the food system, or how the stealth system is one of those bare-bones affairs where if they can't see you, they just do nothing but sit there and ignore the arrows sticking out of their eyes until they fall over dead.
In a sense, at this point, when Book III is almost finished, and you're just now starting to be at the point where you ask "what sort of series do I start next?" is the best point in time to start asking questions like these. Some changes are best done at the total outset of a game (or series).
Just plain "putting more pieces on the board" alone would add a lot more tactical and strategic depth, and so would diversifying the number and types of skills that comprise "classes" (for example, having more than just two "magic" skills/"schools"), but that gets into the question of just what the next project you're starting will be like? There are a great many games where you can take examples for multi-unit strategic or tactical choices, like Avernum or even Final Fantasy Tactics Advance, whereas games that have only one character tend to demand far more individual tactical options and depth than Eschalon gives, and games like Elona, and some other Roguelikes, are good to point out as examples. If the next game is also single-unit-based, then concepts like stealth, for example, should really see expansion with uses of players purposefully creating sounds (especially through player-built noise-maker "traps" they could trigger remotely, along with actual traps that noise-makers could lure enemies into, which would give rogue character skills actual value), better indication of how stealthy characters are, and other things I could easily fill up a post half as long as this one with that wouldn't even require significantly more sophisticated programming than already exists.
It's just that, as it stands, I want to like this game, but just tend to find it boring since all problems tend to be solvable by repeated firebolting to the face. (Or, if they're immune to firebolts, swap over to fleshboils, which work in exactly the same way, and require absolutely no change in tactics.) The major downside to using magic all the time, the MP cost, is not a serious in-game penalty, but it IS a serious out-of-game penalty to have to go off and constantly camp. And that's one of those metagaming incomparables that wind up hurting the player's enjoyment of the experience as a whole.
In a turn-based game like this, it should be tactically challenging. I should be able to feel clever for overcoming an obstacle or defeating a powerful monster, or else I should feel powerful and viscerally enjoy the destruction, but instead, I just feel like I'm going through the motions. However, it's still not even routine enough to even be a Skinner Box "level grinder", either. I find myself going off to play something else in my Steam bin that's light and casual and fun for five minutes like some random puzzle game to cheer myself up so I can go back to trying to individually kite every last supposedly fearsome cow while in the dark so they never even know I'm there.
This thread is a review/critique of Eschalon: Book I primarily, but I'm putting it here, since I am critiquing the mechanics of the game that a trip to the wiki tells me are largely unchanging throughout the series, and hence, are applicable to Book II, and (probably) Book III, as well. Since Basilisk Wrangler seems to actively read and participate in these forums, I'll largely address this thread to him. (But that is not to say I am excluding anyone from the conversation.)
I am someone who might be called a connoisseur of strategy and RPG games. I bought Eschalon through Steam because I love game mechanics, and love sampling many small indie games for their mechanics before either "digesting" a game (in a superfluously loquacious manner such as this), or just moving on to the next game.
I have read some of this forum's threads, and see that people have complaints about the game, but that they aren't quite arriving at the conclusion to which I have arrived. Some people say this game is "too easy", and others "too hard", but there is a lack of the proper framing of the debate (a lack of the proper terms to define what the problem actually is so that everyone can recognize the problem for what it really is) to make sense of why this is the case.
In this game, I see a main problem, and two smaller problems that serve to obfuscate the main problem:
Problem 1: The Learning Cliff
Simply put, the "Learning Cliff" is the key reason why many newer players think this game is "too hard" when the game is, in fact, too easy. (This is why it is a minor problem that obfuscates the larger problems.)
The simplest way to explain what the Learning Cliff is, I'll say this:
I play Dwarf Fortress.
Dwarf Fortress is frequently described as an "extremely hard" game, and extremely frequently as a "fiendishly complex" game. The second of which is almost true. In truth, the main reason why so many people have trouble with Dwarf Fortress (aside from general interface shittiness) is that the game basically just dumps you out in a field with an interface filled with hundreds of buttons, no manual, no tutorial, and tells you to just do your best to survive, and that, "Losing is Fun". (In other words, keep trying until you get it right.)
In truth, DF has a very simple formula to solving basically every challenge the game will throw at you: excavate an easily-defended area, seal all dangers off with walls, and get a farm running to supply you with infinite resources. After you have established a basic defensible position, (which takes an experienced player only minutes,) you have overcome basically every challenge in the game.
Most players, however, fail in their games not because they are stupid, unskilled, or otherwise incompetent, but because they are simply flooded with tons of junk data and no capacity to make sense of how to figure out what was the actually important information or priorities in the midst of a sea of bloat data. They start worrying about forges or crafting trade goods when only farming is really important.
Comparatively, a regular game has a "learning curve" - Each element of the game is introduced one at a time, explained to the player, and then tested. You start with a little skill, and only need a little skill to advance. Then, you need gradually more and more skill.
This game has a learning cliff - the game's rules are actually very simple, and winning is very easy, but the problem is that players are forced to start out in a character screen that demands they allocate character points at a time when they have no idea how any of the skills they are deciding on buying or not will actually work.
Rather than being a slope, you need to know a ton of information, have many skills, to make any headway on the game at all, but as soon as you do, you have all the tools to conquer every single challenge in the game - it's a flat plateau of difficulty from there on out, and the game is "easy".
This is why IJBall "fails" twice at the game, only to find it "too easy" on his third and every successive playthrough.
Put simply, the game isn't hard, the players just aren't succeeding because you aren't telling them the rules fairly and simply before you ask them to start playing. The only "difficulty" in the game arises from players simply not knowing the rules, and "skill" is defined solely by whether you understand what you are doing or not.
You already tell players who are looking at spells, for example, how much damage fire bolt does, and deep freeze does, (1d3+2 and 1d4+4, respectively,) when they are looking at them in the game so that they can actually compare these spells situationally. (Twice the damage versus three times the cost.) However, you seem to consider it a "spoiler" to tell players that they gain massively more MP regen benefits for the same flat cost the more points they put into Perception, which is something they really need to know in order to make proper decisions.
There is a relatively simple fix for this problem - simply tell players the rules. Give players more information on the character creation screen about what each skill actually does, mechanically. D&D, for example, doesn't consider it a "spoiler" to tell you that a given feat or skill will give you a +1 bonus to a specific list of situations, or how to-hit calculations are made. Tell players fairly and simply what a single point (or rather, every two points) of weapon skill will actually buy a player for their skill points, and then let them compare that to what the exact mechanics of spending their points in other skills will be.
You also need to tell players what challenges they will face (I.E. monsters, and how they will attack you, or in other words, the combat formula, as well as traps and locks,) and then what skills there are that you will need, and what general focus you will need in them to overcome those problems. Critically important is that you list the alternatives to solving problems, because choices are all about opportunity costs. The alternatives to picking a chest's lock (rogue skills), for example, are bashing it (fighter skills) or casting lock melt (wizard skill). Knowing this, and how much focus they have to put into any of the given "paths" in this game they will need to overcome an obstacle, they can actually start making strategic choices about how to build their character even on their first playthrough.
You claim outright on your advertisements for the game that this is a game that demands "strategy" on the part of the players, and that stats are everything... But the basis of strategy is the capacity to predict the ramifications of your choices several moves in advance, and compare opportunity costs rationally. You're asking players to make the most important choice they will ever make in this game completely blind and guessing the first time they play (unless they "cheat" by reading the actual rules on the wiki).
By setting up a learning cliff, you cripple your ability to achieve your game's own stated goal of being "strategic".
For my part, I succeeded in my first game of Dwarf Fortress and am succeeding in my first game of Eschalon for the same reason: When I got to the character creation screen, I stopped and read the forum guides/wiki until I actually understood what decision I was making, rather than making a blind guess. (Again, I do this because I am driven by a love of understanding the underlying mechanics, so stopping to just read the manual is, itself, one of the most enjoyable things about a game for me.)
For reference, currently, I have (as of the writing of this - I didn't realize it would take a week to get registered to these forums, so I just wrote down what I was thinking as I was thinking it,) just gotten to Blackwater in Book I, and am level 5. I first went exploring south to get that sextant so I could train up my cartography skill after I already got the reveal map spell that meant I wouldn't even need the skill in the first place. (I completed the "bar of mithril" quest without any mapping feature at all, incidentally.) I was playing a "pure mage", but now am playing a "sage" (spellcaster with both arcane and divine magic, but no melee capacity), and went to Blackwater early to buy up the "free" skill points from trainers, going through Bordertown to get there. In my haste to get to Blackwater, I came across a lot of thugs that were easily dispatched with precisely 3 level 5 firebolts (average 45 damage against a 40-hp enemy), and was only slightly troubled when I came across a mercenary character, and was forced to use my first potion - a haste I potion that was dirt cheap from the Alchemy trainer. I kited the mercenary to death. Now that I no longer need it, I am going through Grimhold mostly so I can loot it. After that, I'll maybe consider starting on the plot missions, unless there's something else I feel like looting, first. Stat-wise, I am a druid "conjurer" that started with 14 in Intelligence and Perception, and immediately min-maxed Perception to 40 (the extra 2 points of stats for the extra MP per level), and then Intelligence to 20, and will make Wisdom up to 15 as soon as I get levels to do so, so that I can just cast haste myself. MP is not a problem as long as I hide and kite enemies one at a time (just walk back and forth for MP), and there is no reason I couldn't do so in this game.
This, in turn, brings me to the next obvious problem:
Problem 2: Enforced Min-Maxing
To explain this, let me start with a bit of a metaphor...
The most regular complaint people have about BioWare games (at least, before ME3...) was that it was supposedly all about making "tough choices" that were, actually, boring and routine and took no thinking at all. That is, everything boiled down to a "Good" or "Evil" choice where you were rewarded more the more "Good" choices you took as soon as you started traveling down the "Good" path at all. You actually received more rewards the more "Good" you were, so there was utterly no "temptation" to be evil other than "for the lulz". Hence, this series of "tough choices" ultimately boiled down to a single choice made right at the very start of character creation, and then sitting through dialogue trees that constantly congratulated/berated you on this choice for an entire game.
BioWare enforced the choices you made - they created only a single goal, and created only a single route to achieving that goal.
Many games that want to encourage diverse playstyles, and as such, when they give you choices of skills, they have diminishing return upon investment. Games that enforce min-maxing, however, have increasing return upon investment.
Eschalon has enforced min-maxing. That is to say, not only is min-maxing generally a possible way to abuse and beat the game, it's actually the only way to play the game. It has what 3rd ed. D&D min-maxers call "Multiple Attribute Dependency" (MAD) if you try to mix your "classes" of characters - trying to be a wizard or cleric means you need tons of Perception, and don't have points for Strength... but by the time you have that much Perception, you might as well just go full-caster, anyway. In order to be good at bows, you need tons of dexterity, but once you have that, you might as well just use bows all the time, and get the Dex-based rogue skills while you're at it.
Basically put, if an attribute or skill is ever worth putting one point into, (because, opportunity cost-wise, it provides the most benefits compared to any alternative,) then it's basically guaranteed to be worth putting all the rest of the points you have into, as well. The only reason anyone wouldn't min-max is because they're indecisive, which is probably because they are "hedging their bets" due to not understanding the ramifications of their choices because of the Learning Cliff discussed previously.
In fact, the only reason to ever stop putting points into skills is that they often scale horribly. The difference between 36 Perception as a druid and 38 Perception while above ground is tripling the rate of MP regen in Book I, while 33 to 36 is doubling, and 30 to 33 is merely one-and-a-half-times. (Again, increased return upon investment.) But then, you slam into the wall and arbitrarily gain no more serious benefit from Perception so far as MP regen works. (This formula needs serious revision, as it is mechanically backwards.) There is no benefit to many skills over a certain point, while others scale infinitely.
Worst of all, you make the game have "absolute" evasion percentages, (magic resistance and dodging, for example,) which is basically anathema to "strategic" gameplay, since they devolve the game into nothing but a crapshoot of waiting until you get a lucky string of rolls. (Or worse, reloading until you get a lucky roll, because, strategically, the only thing you did "wrong" was to not get a good roll.) Likewise, consider that the air shield spell is basically nothing less than an "invincibility versus archers" spell costing 2 MP and lasting 20 turns. (This deserves its own section to deal with as a problem, however, and is beside the main point of what is most seriously flawed in the game, so I'll leave that problem for another time.)
This, in turn, leads to characters being "One Trick Ponies" that have one move that beats everything the game throws at them.
There are several solutions to this problem. First, you need to start by rebuilding your formulas to scale properly. Use less addition and subtraction, and start using more multiplication or division. (I'll go into this more in a later post.) Next, you need to reformulate skill point costs, and (if you keep linear returns upon investment, like +1 To-Hit every couple weapon skill ranks,) introduce sub-geometric or geometric growth rates on the costs of skills while increasing total skill points per level over levels, similar to how Avernum works. Likewise, similar to Avernum, make the most useful skills cost more than the utility skills. There's no reason skills like lore or spot hidden or cartography should be equal to arcane skills, especially since there are spells that obviate the whole need to use two of those other skills in the first place.
Alternately, you can ditch skill points altogether, and, like Elona, Dwarf Fortress, or The Elder Scrolls, just use "train with use" mechanics. It also obviates the need to "fix" trainers, since you can simply make trainers into people that just let you have, say, the experience of 50 bow shots for 100 gp at the archery range rather than having to grind out that many bow shots normally.
Your entire "class system" itself is something of a misnomer, as well, and you might want to look into cribbing some notes from games like Elona or World of Darkness, and introduce "Potential" or "Affinity" for skills rather than outright classes. (I have tables I can share on this... but that is going off-topic, however, so I'll save it for another post.)
However, these last two problems were just the obfuscating problems that overshadowed what was truly the cause of why the game is so "easy" it's actually boring, and players have to come up with "challenges" just to keep things interesting:
Problem 3: Overly Simplified Rules Creating A "Solved Game"
The reason why there is nothing but a "learning cliff" that turns into a plateau is that all challenges in the game require exactly the same response - there's no "curve" or "slope" past the initial "challenge" of not knowing what the rules of the game are when creating my character.
Functionally speaking, to my mage character, every melee enemy in the game is just a fanged salamander with more hit points. If they resist magic, that's just making combat a bit more of a crap shoot, and maybe I have to use haste to kite the target, or at worst, reload and try again and hope I'm not as unlucky, or just use fleshboil. There's no further strategic thought that has to go into "this enemy is completely immune to the character build this game forced me into min-maxing my character into being, then schizophrenically decided min-maxing was a bad thing, and made immune to elemental magic," that just means you need to use fleshboil, instead. Air shield means I am invincible against archers. Element shield is basically the same thing versus wizards (with enough Intelligence and the rank 6 spell, that's 99% resist, easy).
Mages have the greatest diversity of tactics at their disposal, at that - melee characters are basically faced with always exactly the same "strategy" for everything. That is, just keep punching it until either it dies, or you die.
Yes, you do have some alternative options as a melee fighter, but - and this is the important part - those options have nothing to do with the fact that you are a melee fighter. Those options are almost always either spacial problems or else consumables, available to all characters. So far as melee fighter is concerned, skill-wise, the difference between a "hard" and an "easy" fight is just how big the numbers are, and how many clicks on the icon it takes to make the HPs become 0 and make the peoples fall down.
You also have to realize that, compared to games like Avernum (or D&D), you are severely limiting the number of tactical options a player has by simply having only one "unit on the board" at a time. By comparison to Avernum, (which has very similar character-building mechanics,) you actually have less options for good moves at a time by an order of magnitude for every individual unit you don't have in your party - in a game where every action takes up equal amounts of time, basically all decisions towards achieving a single clear goal - killing the enemy as fast as possible. Because so many of the skills for killing things operate along similar lines in this game, (you are basically just comparing DPS, since there are no situational bonuses or penalties in this game, barring light penalties,) they become little more than "calculations" - what is the most DPS you can get with a move?
To give a far more in-depth example of how much tactical flexibility some really good RPGs can give a player when they advertise many strategic options for solving problems, I'd like to point to the Russian-made game Hammer and Sickle (a spin-off of Silent Storm, and somewhat similar to X-Com if it had far more focus on fewer characters with deeper skill growth). In one event in that game, the villains have taken over the hideout you could have previously used, and are lying in ambush for you. In my game as a sniper character (with a gunner and grenadier support characters), I responded to this by crawling in one of the windows of an abandoned house where one of the enemy snipers was lying in wait for me, and stealth killing him with my sneaky sniper, then crawling through the darkness to snipe one of the other snipers from another house across a narrow alleyway. I started picking off snipers while moving from window to window to keep from being spotted - popping up and letting of a blast before moving on - and some of the incoming "sweeper patrols" that moved as a unit to pick off hiding units like my sniper main character based on the sound of my rifle.
I was on the second floor of this building, and I had my supporting units behind me, watching the stairs. I guess the sweepers that got into my building somehow heard my supporting units, because they weren't willing to charge up the stairwell all nice and single-file so my waiting gunners could mow them down.
It was an impasse - they weren't willing to come up, and I sure couldn't go down those stairs. I could try jumping out a window, but that would expose me in the streets, and there were still other sweepers out there. Then I noticed that parts of the floor were weak, and a thought occurred to me - I still had some C4 left. I planted it down on top of where I could hear the sweepers were on the floor below (facing the stairwell), and then backed up, and detonated it. The destructible environment collapsed, and the explosion stunned the sweepers, at which point I had my grenadier pop up and chuck a grenade down on them before they could recover and run. He and my gunner then hosed the survivors down with their assault rifles.
Compare this to a game that devolves into what essentially amounts to just throwing dice until one side or the other runs out of hit points, and I think you can see which game can offer up a greater selection of strategic choices. (Worth emphasizing is that in Hammer and Sickle, even the enemies were much smarter about how they fought, and forced the player to fight smarter because of that.) Even if you take examples just from D&D, then you can have defensive stances, tripping attacks, disarming attacks, charges, non-lethal attacks, grapples, and more.
Even in single-character games, you can still look at games like Roguelikes, which tend to have many alternatives and very strange, even gimmicky, items, enemies, and obstacles to exploit.
For example, to give weapon skills more diversity in Book II, feats were added... but there's no real tactical choice in using feats, so they keep combat boring - feats are basically always better than a normal attack, have no real downside, and are just "you can use this once per battle" skills. If, however, you had more than one feat to choose from, but could use only one feat per battle, then you start having real tactical choices from the fighter class.
Before I go further on this, I need to define some more terms so that I can be clearly understood, so I want to link two videos from the excellent gaming-analysis group, Extra Credits:
Choice and Conflict - Helps define "real choice" in games, as opposed to "calculations", as well as defining how "incomparables" are key to creating serious choices, and setting up conflicting goals is key to creating strategic choices.
Perfect Imbalance - Less necessary than the first, but this helps set up a better understanding of the sort of "incomparables" that a turn-based game that focuses upon "strategic" choices really needs to deal in. That is, anything that uses the same types of stats are easily compared. Comparing things that have a perfect imbalance, such as stealth bonuses versus bonuses to loot gained if you sneak around enemies creates serious conflict of goals for players of stealthy characters.
To give an example of incomparables, before bringing this back to the main thrust of the argument, look at the issue of weapon weights, for a second. All of your weapons (and materials they are made of) are completely linear in function, not creating any sort of conflict between differing goals. You missed an obvious chance to make a balance of incomparables with your weapons when you went and made ALL the weapon materials successively heavier than any material "less powerful" than the given material - mithril, one of the best weapon materials - is heavier than steel, and functionally virtually identical to adamantium. Even putting aside the whole fact that the whole defining feature of mithril was that it was lighter than steel, it's missing a perfectly good opportunity to make for an actually interesting choice when it comes to weapons: Do you take a stronger weapon, or a lighter weapon? Even if we don't have any more advanced mechanic, like initiative that gets partially determined by weapon weight, just the fact that a magic user can't cast with a weapon heavier than 1/3rd their Strength alone makes light weapons very attractive to magic users. Why didn't you use this existing mechanic to set up an actual strategic choice for the player to consider - using a slightly weaker weapon in exchange for not needing as much Strength, so that they could spend their attribute points elsewhere, probably buffing that magic casting so that they could buff their melee capacity to compensate?
So, to pull this all together for the main argument I am making, this game is "easy" to anyone past the learning cliff, and nowhere near as strategic as it claims, or, for a turn-based game, (where skill is necessarily found in strategic thought, since there is no dexterous skill particularly required,) needs to be in order to be enjoyable. There is only one real challenge in the game - combat - and in that case, there is only one real goal (defeating the enemy), which does not create a tension of conflicting goals, and all tactical choices are comparables; It is an easy measurement of what actions remove the threat of an enemy most quickly, reducing all combat quickly to routine.
Differences in "classes" are true incomparables, and the true choices that separate out how you play one game from another. However, you are kidding yourself if you are saying there are "unlimited character choices", because there are really only 4 choices you can make to solve the only real problem you face in the game (monsters - doors and chests don't require any real strategy and can be skipped) and that is melee weapon (with choice of exact weapon a quibbling difference with little in-game incomparability), ranged weapon, magic (with differences of wizard/cleric being fairly minor), and hiding/running away. You can try to split the difference on these, but except for the "running away" choice, they just don't mix well because of MAD splitting off your attribute points until you are already so well-leveled (because you've been relying upon a "pure" version of a character) that you have enough spare points to be able to splurge on gaining a second specialization. (Marginal exception for cleric-warriors, since taking points out of Str and Dex can be compensated for with "buffs".)
To go back to how fighters have so few tactical options that are actually related to being a melee fighter, let's take a chance to analyze how important consumable items (and classic D&D Vancian Magic) are in RPGs in creating conflicts of goals:
Potions are a perfect object for creating a conflict of a player's short-term and long-term goals. That is, a potion of haste will always be a massive help in any fight... however, then comes that nagging thought. The thought that you can't escape - BUT YOU MIGHT NEED IT LATER. THAT's a conflict of goals, and what creates real strategic choice in a bottle. (And what's the "most powerful" pair of skills in the game? The alchemy skill and ingredient-gathering skills that give you all the potions you'll ever need, so you WON'T "need them later"...)
This is the same as what Vancian Magic did to wizards - you only have a couple level 3 spells as a level 5 or 6 wizard... the DM seems to be ready to throw landscape obstacles at you - you can take a Fly spell with you to bypass obstacles... but then, what if the DM sends a bunch of harpies after you? Better take a fireball, instead. But what if it's a magic-immune golem, instead? You'd be better served with a Haste spell to throw on your fighter, who can use it to deal more damage against nearly any enemy. Those are real strategic choices based upon incomparables. (And come on, you're not going to use that fireball against a bunch of kobolds - you only have one of those, and you'll need it later, so just use a lesser Sleep spell, instead.)
Something worth recognizing is that you also have "Metagame" conflicts of goals, where in-game goals are put in conflict with out-of-game goals, which can actually be terrible for a player enjoying the game. The Extra Credits video showed how the in-game most powerful strategy was also the out-of-game least enjoyable strategy, which created a conflict of goals between doing badly in the game, or not having fun... which generally results in player dissatisfaction no matter how they choose. In this game, save-scumming for better loot or constant re-rolls are a similar form of metagame conflict of goals. You are trading the game going quickly and letting you go through the game at an enjoyable pace for the knowledge that you're getting the best possible loot. (In)comparably, you could just accept a ton of trapped chests with nothing but a femur inside, and lose enjoyment in the game because you feel "cheated" out of a larger haul of loot. Either way, I actually wind up enjoying the game more when treasure chests have deterministic loot so that I don't have to actually feel this conflict.
It's not just "cheating" that suffers from this problem, however. Many of the game's more frustrating aspects (and, generally, most of what leaves me dissatisfied about the game,) is that so many of the in-game choices are basically just this same "best way to play" in-game rewards versus "completely tedious" out-of-game punishments are basically the only choices ever presented in this game.
Take Reveal Map, for example - it has an enemy-tracking function that is the only power in the game that lets me see enemies before they see me on its level 6 casting... and there's no reason to cast anything less than the maximum, because the only reason I DON'T use it all the time is that it's such a pain to continuously cast. I'm trading being able to know how and where all the enemies are in the game for it not being such a godawful slog of casts and camps as it already is.
That's even more the case in what is probably the single most egregiously broken aspect of the game: The "walking the dog" problem. You can quite simply just out-walk a velociraptor. Then, take advantage of zone boundaries to firebolt creatures to death from utter safety. (Incidentally, that's how I got my Citizen's Writ in Book II by killing the Red Wolves in Wolfenwood at level 1...) It's how people do things like beat the game on level 1, if that isn't enough proof of how broken this game is because of relying upon dumb-as-bricks AI and purely melee-based enemies with no running mechanic or actual speed stat.
In other words, if the most powerful strategy in the game that will probably defeat even the most powerful and vile monster you will ever see is probably best set to the tune "Yakkity Sax", there's a problem with the way the game's mechanics works.
And it's not just that this makes the game "too easy" - I can enjoy easy games - it's that it basically means that most of the ways that you have tried to make the game "harder" has actually just made the game more tedious. (Killing a taurax is the same as killing a fanged salamander, but with more HP and so many resists that it takes running away and camping for more MP once or twice to finally grind it down.) And it's always a choice in this game - do I want to play it "right", the way that, strategically, I should play the game if I were thinking about the game the way that you SAY I should be thinking about the game, and playing the best I can... or do I want to have any sort of enjoyment at all?
Another problem that I see people have argued over heavily is the problem of how trainers help enforce min-maxing. The problem is, people haven't really framed this debate rather well, so the debate tends to go nowhere.
What trainers really do is set up badly balanced incomparables. That is, strategically speaking, we have limited resources with which to build our characters, and of these resources, skill points (and to a lesser extent, attribute points,) are the most precious. This is because, in strategic choice, opportunity cost is everything. With enough skill points, you could be skilled in every playstyle at once, so there's always something else you could be spending those skill points on.
Meanwhile, however, trainers are an option to reach the same player goal of a greater diversity of skills, but where the resource you spend isn't your precious skill points, but are, instead, your relatively worthless gold. Yes, relatively worthless, because the actual opportunity cost of spending money on training is negligible. You can always get more than enough money for everything you need, especially with the many tricks for getting more money.
To show how this could actually be remedied into an incomparable choice that actually encourages strategic thought: If you were to take one of the already-suggested solutions to trainers, (that you just plain get 5 uses of the trainer per game, regardless of what your actual skill level,) and then mix in the increasing skill point cost per rank (so that, at its simplest, the first level of a skill costs 1 skill point, second level of a skill costs 2 skill points, third level of a skill costs 3 skill points, etc.) idea, then you create a conflict. That is, buying a skill from a trainer for money at skill level 1 will give you the benefit of having spent 2 skill points to get to level 2 in that skill. Meanwhile, buying that skill from a trainer at level 11 will give you the benefit of 12 skill points worth of actual skill increase. That's a "discount" of 10 whole skill points!
But here's the conflict - the longer you wait to buy those "free skill points" by training, the further in the game you will have already have to be, and higher the level you will need to have been.
Hence, it's a conflict of a player's desire to have the most powerful character they can have, statistically, versus the player's desire to actually use those skills in the game - what's the good of having high skill points on paper, if you've waited until the very end of the game to get them, when you obviously didn't need them, and can't even use them anymore, anyway? (Except, possibly, against some type of "bonus boss".)
When designing a game, you need to recognize what each game element actually adds into the game in terms of what its actual choices or player interactions. Every game element, especially in a game that asks players to "think strategically," has to add a greater selection of real incomparables, and you need to actually be able to understand what choices those incomparables are making players select between from even a "metagame" standpoint.
I have a set of possible solutions to these problems I have laid out (especially the last one), but I would like to first see the response to all that I've written so far. This isn't even starting to get into the problems I see in the magic system, the food system, or how the stealth system is one of those bare-bones affairs where if they can't see you, they just do nothing but sit there and ignore the arrows sticking out of their eyes until they fall over dead.
In a sense, at this point, when Book III is almost finished, and you're just now starting to be at the point where you ask "what sort of series do I start next?" is the best point in time to start asking questions like these. Some changes are best done at the total outset of a game (or series).
Just plain "putting more pieces on the board" alone would add a lot more tactical and strategic depth, and so would diversifying the number and types of skills that comprise "classes" (for example, having more than just two "magic" skills/"schools"), but that gets into the question of just what the next project you're starting will be like? There are a great many games where you can take examples for multi-unit strategic or tactical choices, like Avernum or even Final Fantasy Tactics Advance, whereas games that have only one character tend to demand far more individual tactical options and depth than Eschalon gives, and games like Elona, and some other Roguelikes, are good to point out as examples. If the next game is also single-unit-based, then concepts like stealth, for example, should really see expansion with uses of players purposefully creating sounds (especially through player-built noise-maker "traps" they could trigger remotely, along with actual traps that noise-makers could lure enemies into, which would give rogue character skills actual value), better indication of how stealthy characters are, and other things I could easily fill up a post half as long as this one with that wouldn't even require significantly more sophisticated programming than already exists.
It's just that, as it stands, I want to like this game, but just tend to find it boring since all problems tend to be solvable by repeated firebolting to the face. (Or, if they're immune to firebolts, swap over to fleshboils, which work in exactly the same way, and require absolutely no change in tactics.) The major downside to using magic all the time, the MP cost, is not a serious in-game penalty, but it IS a serious out-of-game penalty to have to go off and constantly camp. And that's one of those metagaming incomparables that wind up hurting the player's enjoyment of the experience as a whole.
In a turn-based game like this, it should be tactically challenging. I should be able to feel clever for overcoming an obstacle or defeating a powerful monster, or else I should feel powerful and viscerally enjoy the destruction, but instead, I just feel like I'm going through the motions. However, it's still not even routine enough to even be a Skinner Box "level grinder", either. I find myself going off to play something else in my Steam bin that's light and casual and fun for five minutes like some random puzzle game to cheer myself up so I can go back to trying to individually kite every last supposedly fearsome cow while in the dark so they never even know I'm there.
I'll have the Bureau of Mage Supremacy on you!