Fireside this week! I’m back home now from PDXCON2022 so it is back to work. I know there have been a number of requests to know if the historians panel with Eleanor Janega and myself was recorded; it was and the recording is set to be uploaded shortly, but there are a number of turning wheels there. When it appears, I will let you all know!
For this week’s musing, I wanted to talk in a less carefully organized way about how to ‘read’ the meaning encoded into the mechanics of games. This is something I discussed only very briefly in my PDXCON talk (since that was directed more at an audience of players than at historians interested in methodology) but I though it might be interesting to talk about how I think on this. A caveat at the front that because my training didn’t focus much on reception studies and because I’ve only really started to back my way into games analysis, my own reading of other scholars doing this sort of analysis is not very deep, so it is possible I am doing some wheel-reinventing here.1
The tricky thing with assessing mechanics-driven games is that the experience they provide is not the same for everyone who experiences, far beyond the sort of reader-induced subjectivity of a narrative text. So there is no one ‘golden’ playthrough through which all of the meaning may be assessed, yet at the same time I have argued that we can draw out meaning from the mechanics and the experiences they tend to produce.
That said my first ‘rule’ as it were is that the experience is paramount, which is to say that the thing which matters most is the way the player experiences all of the mechanics and content together. The extreme example of this principle would be cut content: yes it is in the game files, but unlike, say, an appendix in a book, players aren’t going to experience it so it doesn’t usually ‘count’ for an analysis. Even for ‘live’ content and mechanics, one needs to always have in mind what proportion of the players of the game will actually experience it. A lot of games feature play-modes and pathways that were planned for but intended for only a few players or to exist mostly as alternatives to give the most common path extra meaning (the most frequent use of this are ‘evil’ options chosen by only a small minority of players which nevertheless give the majority of player’s ‘good’ path more meaning by offering a fully developed alternative).
I should note that I don’t think, in most cases, this is an unfair standard to hold the work of game developers to. Good developers are very aware of what content is more or less likely to get experienced and many modern games with online features actively feed metrics back to developers (e.g. “how many players got that ending” or “beat that boss“) As I argued with Assassin’s Creed: Valhalla, hiding certain elements of long games in areas few players will reach is a development choice: both something fair to critique and also something that clearly bears on the median or typical player experience.
That in turn impacts how I go about experiencing a game in order to critique it. For fairly single-path narrative experiences one can generally just play the game, but for more mechanics-driven games with greater variability in outcome I think it is also helpful to immerse one’s self a bit in the community of the game if possible in order to get a sense of what outcomes and playstyles are common. Ideally a scholar might try to get access to those developer-harvested metrics (but good luck with that), but in their absence the community or even better global achievement statistics (Steam makes these available) can give a sense of the typical playthrough. For instance about 38% of players on Steam reach the final boss in Elden Ring (fairly high for a game so long); 75% beat the first major boss. Strikingly there, the ‘Age of the Stars’ ending – despite being arguably the most complex and difficult to obtain – is by far the most common; 26% of players (so more than two thirds of the 38% that finish the game) have this ending, compared to just 13% with the ‘Lord of Frenzied Flame’ ending and 19% with the ‘Elden Lord’ endings (a package of four endings which share an achievement).
If I’m thinking about the meaning in Elden Ring – which I think frankly would be even better served by being discussed by myth-and-literature scholars than by historians – I have to be thinking about it in terms of a game where only about half of players that make a solid investment reach the end, but of that half, two-thirds aim for an achieve a complex, difficult ending with a very different tone than the others. You could almost pie-chart the player end-states thus when thinking about the experience of the game: half don’t finish, a third or so get the ‘Age of the Stars’ ending in at least one playthrough and the remaining 17% experience only the more tonally downcast endings (in practice given the structure of the game I suspect most of that 17% only experience the Elden Lord endings, because that’s what happens if you do the bare minimum to reach the end of the game).
When thinking about the actual mechanics, it’s important to think both in terms of how they both describe the game world (which in a historical game is ‘the past’) and also how they motivate player actions. To again take a deceptively simple example, a game with falling damage is both attesting to a realistic depiction of gravity but also pushing players to be cautious around long drops: the mechanic both describes the world and also pushes the player towards certain actions. This duality struck me most vividly when I was thinking about the function of EUIV‘s anarchic political system: it both attempts to describe the past (that is, it contends that the early modern period was a violent episode of interstate anarchy) but also motivates player behavior (pushing them towards conquest).
‘Pushes’ on players come in a range of intensities. Explicit win and loss states are the strongest sort of player-motivations, but there are a lot of more subtle ways for mechanics to shape player behavior that needs to be considered. Consider a choice which moves the player closer to a ‘win-state’ but also locks out large amounts of content – is that an encouraged action? Perhaps counter-intuitively, no; that’s the road to the ‘bad ending’ in most multiple ending games (e.g. the ‘Age of Fracture’ ending in Elden Ring; there’s a reason most players that complete the game do so with the far more complex and difficult ‘Age of Stars’ ending). Getting to see all of the content is often itself a reward and thus a motivation for player behavior. Likewise becoming more powerful, even if it doesn’t lead directly to a win-state, is often a fairly strong motivator, though some games play with this: the classic example being the story-framed ‘evil’ choice resulting in somewhat more ‘power’ than the story-framed ‘good’ choice (a tactic that largely just exists to give the good choice most players will take moral resonance, which is why a large amount of content is typically not hidden behind these choices).
All of this can be tricky, but it is important to avoid mistaking what the player can do compared to what the player is encouraged to do. You can play a Grand Theft Auto game as a good citizen who never breaks traffic laws, but doing so will block progress in the plot; the game pushes you towards violent, destructive play. On the flip-side, you can play Skyrim like it’s Grand Theft (Middle) Ages, but you will lock out a ton of content and the game’s systems push back against this style of play.2 It is more useful to think in terms of ‘motivation,’ or ‘pushes and nudges’ than the more common language of ‘intended play’ because of course if the game is able to respond to your actions at all it is because the developers on some level ‘intended’ your play to be possible. Killing every NPC you meet in a FromSoft game is thus ‘intended’ play but because it locks out much more content than it reveals, I’d argue it is mostly discouraged play as well.
At the same time in most games these mechanics are intended to represent something about the world being described and so it is useful for try to understand what is being represented and how. Often the answer is obvious but in some cases one needs to plumb some of the more complex mechanical interactions. A good example of this are EUIV‘s institutions; it would be easy enough to assume that an institution like ‘the Printing Press’ just represents the emergence of modern moveable-type printers and move on; it is their connection with the technology system and the technology system’s connection to military power that reveals institutions as the means that EUIV attempts to explain the military revolution and the great divergence. At the same time it is important to appreciate the level of abstraction involved in such mechanics; complex things are often simplified heavily in game mechanics.
Of course none of these frameworks are a replacement for simply playing the game; at the same time they are helpful in playing a game with an open eye to how the mechanics of the game itself convey part of its meaning.
On to this week’s recommendations!
I can’t offer my normal recommendations for updates on the War in Ukraine, because events right now are moving fairly quickly, with Ukrainian offensives in both Kherson and Kharkiv oblasts apparently finding meaningful success. Anything I posted here would probably be dated by the time you saw it. I’d suggest keeping an eye out for Michael Kofman to make an appearance on the War on the Rocks podcast. The best place to try to keep track of what is confirmed is probably the ISW’s Russian campaign assessments (which despite their name give a sense of both parties, though prioritize Russia to avoid compromising Ukrainian operational security). If you’re on twitter, look for @KofmanMichael, @RALee85, and @Nrg8000 along with the folks they’re citing and retweeting and you won’t be led astray. But I should be careful to stress at this point that the fog of war is thick and we know relatively little for certain.
Meanwhile in the land of Classics, Pasts Imperfect after a bit of a hiatus earlier in the year continues to deliver updates on all sorts of wonderful antiquity related content available to the public. Of particular note for me was their highlighting of the tweet below by Carol Drinkwater on a Roman fort exposed because of sinking water levels in Spain. The view of the fort and its layout, complete with the partially visible ‘playing card‘ shape is fantastic, but of course the climate-change fueled drought that revealed it is also deeply concerning.
For this week’s book recommendation, I am going to recommend R.H. Spector, At War At Sea: Sailors and Naval Combat in the Twentieth Century (2001), particularly as a good modern introduction both to some of the key events of 20th century naval warfare but also the key concepts and tensions which animate the discussion of modern navies and naval warfare. The tension is, in many ways, signaled in the title: this is a book about sailors and naval combat. A persistent tension in military history is between those who focus on metal and those who focus on ‘meat and minds,’ which is to say the human element. In land warfare, the human approach seems to have been dominant for decades, especially since a ‘war and society’ lens tends to understand materiel superiority as a consequence of social structures and popular will. But naval warfare has long been a field where metal was at its strongest; a ship’s crew cannot flee in terror (though a captain might lose their nerve and retreat) nor can they charge with more boldness than the engines allow.
Nevertheless the human factor matters in naval warfare as well and Spector does a good job bringing this out, arguing for a balance between human concerns and the concerns of the metal, though of course technical explanations still intrude when relevant – and they are frequently relevant. Spector brings out this human element very well, particularly in the repeated focus on how institutional structures and cultures produced officers and sailors who then acted in accordance with the values and culture of those institutions for better and for worse. The reader will quickly recognize the repeated patterns of closed, elite naval hierarchies reinforced by conformity-inducing training and narrow recruitment pools struggling to adapt to naval warfare that increasingly demanded both non-elite technical expertise but also operated on a much greater scale with a greater degree of decision-making needing to be delegated down to junior officers and petty officers.
Spector has structured this approach as a series of case studies focused on particular battles or events, connecting the shaping institutions at the beginning of a chapter to the outcome of the event towards the end. The battles so treated are mostly ones to be expected: Tsushima, Jutland, Leyte, with some that may be less familiar like the British evacuation of Crete or the Flight 655 shootdown in 1988. The book is clearly written for a popular audience, though it does feature (end)notes. Importantly, it is clearly written and doesn’t require a mountain of prior experience with naval terminology to make sense of. As a result the book is both an argument for a more human-centered approach to naval warfare and a solid toehold for those looking to push into some of the more resolutely technical and ‘metal-oriented’ areas of the field.
- Which seems a good time to offer a reminder that these Fireside posts are musings, not something very carefully polished and long-considered.
- This is a good example, by the way of how small changes in mechanics can make for big changes in tone. GTA‘s ‘alert’ level (stars) clears fairly quickly if you evade pursuit, encouraging the player to do whatever crime they will and then escape. By contrast Skyrim‘s bounties do not clear over time; with a few odd exceptions once you have one you are stuck with it until it is paid. That means that just carrying large bounties all the time ends up being a very limited playstyle as most quest NPCs are in cities where guards will make things permanently difficult for a player with a bounty. Thus despite having superficially similar crime/bounty/witness systems, the subtle changes cause each game’s system to incentivize radically different play.
Focusing on what the game encourages or discourages you to do is an interesting midpoint between the affective experience of the game and the hard code. I’ve always found analysis that’s too far on either end to be dissatisfying. I also think it’s generative to consider what choices are even possible – for instance, it’s not an option for me to dissolve the government in Civ 5, or to voluntarily re-locate my capital, etc. Those examples might not be the most interesting, but they do say something about the game’s view of history, as both of those things have happened in the real world. Some player actions are nudged or motivated, and some are completely locked out.
This is what you’d call the proceduralist school of game design and analysis. Mechanics are the rules of the game world. Procedures are unwritten rules about how and when a player interacts with those mechanics and what that does to shape the intended experience.
Historical board games are very interesting cases here, more often than not they would use rule limitations and victory conditions to make (conscious) arguments of history.
For instance, in Twilight Struggle – a game about USSR-USA competitions in the Cold War – both USSR and USA players need to do coups or counter coups every round; if they try to expand their influence entirely through peacefully means they will lost victory points. However, if both players do coups too many times in a game, the DEFCON level will drop to the nuclear war, and the player who initiated the last coup – the player who triggered nuclear war – lost the game.
Both are historical arguments of the Cold War, made very consciously, via game mechanics telling players what they should do and what they should not.
RE: Elden Ring Endings. First of all, thank you for correctly identifying the upper bound of what % of people beat the game (the % who beat the boss right before the final boss), instead of just adding the three ending achievement %s together. Second, I think a large part of why Age of Stars is the most common ending is because (and this is conjecture) the kind of person willing to see the whole game through is *also*, by and large, the kind of person willing to look up the ending requirements, or “how do I get the moonlight greatsword” (a weapon as constant as Patches), or simply “how do you get to that weird plateau in Liurnia” and stumble onto Ranni’s questline. Or, indeed, simply find the questline normally, and see that through to the end. The various ending questlines don’t even lock each other out, besides Frenzied Flame, which should *actually* be seen as the annoying-as-hell secret ending.
Finally, regarding the discussion good and evil options and how few people take the evil options, I’m curious what the ratio is in something like Pathfinder: Wrath of the Righteous, a CRPG with multiple, playthrough-defining mythic paths exclusively for evil characters.
Especially given that some endings are better structurally than others, and sense dictates that the best will probably be the one that got the most attention.
IIRC, Paradox data mentiones that by far the most common Stellaris Ethics combinations are xenophile/egalitarian. So it’s partially a matter of people just liking to play the good guys.
It’s worth remembering that Xenophile/Egalitarian are the ethics Paradox put on humans, who are at the top of the list of presets. “It’s the default” has more explanatory power than “people like the good guys”.
It’s not really treated as “mechanically default,” though. No ethics is the default.
Also, there’s plenty of data indicating that players bias towards “good” moral decisions, often overwhelmingly. For a simple example, metrics released by Telltale show that 80-90% of players would choose “good” choices in their Walking Dead game (source: https://venturebeat.com/games/telltale-games-the-walking-dead-statistics-trailer/).
This bias is especially strong in games where the player thinks they need to decide between a Good Playthrough and an Evil Playthrough early in the campaign. With no information to go on (since they’ve just started the game, and statistically speaking, most players don’t read guides or whatever ahead of time), they default to Good, whether because they think of themselves as good people or because they assume an Evil Playthrough would be in some way less satisfying.
(Which they often are, whether because the Evil side didn’t receive equal development resources, feeling obligated to do things you don’t like to stay Evil, or just the fact that Evil is rarely as internally coherent a moral philosophy as Good.)
Examples include most games centered around moral choice (with neutral/mixed endings usually being less committal/interesting than Good or Evil ones), most Star Wars games which let you choose between the Light and Dark Side, and…well, games where you literally have to decide whether your character is Good, Neutral, or Evil during character creation. Like any D&Derivative games, including video games using the rules system of a spinoff of D&D’s 3.5th edition. Like Pathfinder: Kingmaker.
It might not be the default in the sense of needing to move off of it, but the UNE is one of the recommended first empires to play as. If players start exactly one game, it’s likely to be that one. It also helps that Xenophile is a moderately powerful ethic that also just happens to synergize well with egalitarian.
With that said, the fact that people prefer to be the good guys is very real, even among a community like Stellaris that memes about how often people talk about doing slavery and genocide. (I also take pity on the poor computer empires and wish to extend my benevolent rule over them.)
>(Which they often are,
>whether because the Evil
>side didn’t receive equal
>development resources,
>feeling obligated to do
>things you don’t like to
>stay Evil, or just the fact
>that Evil is rarely as in-
>ternally coherent a moral
>philosophy as Good.)
I will note that the second and third of those basically reduce to “reasons why most players would have a general preference for good over evil.”
I’m not comfortable punting toddlers in a video game because punting toddlers is wrong and I know it. It is not a desirable course to pursue, even in a fictional setting where my character can theoretically get away with punting a toddler and not spend years in jail for it.
On a more abstract level, I’m backing, say, the NCR over Caesar’s Legion in Fallout in part because coherent theories of statecraft and human behavior tell me that the NCR will produce better outcomes for the region in the long run. To really and unironically believe Caesar should win, you have to adopt some willful incoherence.
Now, the first point (lack of development resources) is definitely NOT about innate character preferences.
I’d argue that there’s a distinction between the second/third points are, at the very least, not only about players having a general preference for good over evil.
The third point is all about narrative. Some people play games for the mechanics alone, but most people care about the narrative at least a little. A narrative where the protagonist of the story acts with a consistent drive to help others, even at his own expense, is more narratively satisfying than one where the protagonist just randomly shoots minions or kicks puppies because they’re Evil—whether or not you’d prefer good over evil! After all, if you want to play an evil character, you probably want to play an evil character, not an incoherent mess of meanness.
The second is fuzzier, but I’d argue that a player who prefers evil characters will generally find good actions less objectionable than vise versa. There are plenty of pragmatic reasons why doing Good Deeds could help an evil character, after all! That’s not objectionable, that’s just strategy. A good guy doing blood sacrifice, less so.
Stupid Evil is an alignment for good reason.
@Richard H
>With that said, the fact
>that people prefer to be
>the good guys is very
>real, even among a com-
>munity like Stellaris that
>memes about how often
>people talk about doing
>slavery and genocide.
I think that in the modern Internet, we have somehow worked our way around to a culture where people like to talk about doing things in fiction that would be appalling in real life, and feel comfortable and bold and even deliciously transgressive talking about it… But where the majority of people who don’t actively feel good about doing that don’t feel like they can present themselves that way.
Alice runs around playing her Stellaris games going all “EXTERMINATE” Dalek-style and she can crack jokes about it. Betty plays Stellaris games with a relative minimum of conquest and no enslavement/genocide, but she’s not going to be saying “I don’t like to do genocide in a video game, I think it’s wrong” nearly as often as Alice is going to be making Dalek jokes.
Possibly because Betty doesn’t want to be seen as dictating how Alice is allowed to have fun. Possibly because hey, Betty knows she did have an “EXTERMINATE” playthrough that one time so who is she to talk. With the result that the Alices’ collective post count talking about genocidal playthroughs and the Betties’ post count talking about non-genocidal playthroughs wind up closer to evenly matched than you’d think, given that there are several times more Betties than Alices.
Somehow, the “Murder Empire” stuff gets seen as more memeworthy than the “Wholesome Space People” stuff, by one process or another. And I think this creates a disconnect between how Internet communities dedicated to discussing the game see themselves, versus the actual experiences and preferences of the playerbase as a whole.
@Simon_Jester, you are probably overthinking your Stellaris example.
Online game designers have known for a long time that different players have very different attitudes, and when these collide in multiplayer games the mutual incomprehension leads to much angst and complaining.
A commonly accepted figure is that about half the players identify with their game characters, about half don’t. Two influential articles are 1990 “Lessons of Lucasfilms’s Habitat”, and 1996 “Hearts, Clubs, Diamonds, Spades: Players Who Suit MUDs”. (By this, no there are not several times more “Betties” than “Alices”.)
So Alice in your Stellaris example isn’t thinking that she’s being transgressive or anything that deep. She is *playing a game* and knows it. She’s zapping computer generated icons with no real world consequences, and has no more reason to feel guilty or wicked than the actors playing Lord and Lady MacBeth in a Shakespeare play.
Same goes for punting toddlers in a video game. Why not? Various self-appointed moral guardians have been trying to link video game violence to the real world for decades and failing – there’s quite a nice correlation between increasingly realistic video game violence and *decreasing* violence in the societies that play them.
There can certainly be an internally consistent and narratively satisfying evil character. The Flashman character in Fraser’s Flashman novels is an example: racist, sexist, bullying, and cowardly. And I don’t know about computer game players, but in other settings (e.g,, pornography viewing), people are comfortable with fantasy behaviors that they would not condone in real life.
I’m not saying that internally-consistent, narratively-satisfying evil characters are impossible. They’re just less common, especially in the specific context of video games with binary morality systems. They’re not usually as bad as Fallout 3’s “do you want to nuke Megaton for no reason?” decision, but they’re frequently closer to that end of the spectrum than to Flashman or…um, any compelling villain protagonist from an actual video game.
And yes, people are more willing to murder people in video games than IRL. This is obvious with a brief comparison of Los Santos to Los Angeles; one of them has way more murder, larceny, and speeding. But the statistics still show that people play the hero more often than the villain.
In a game with different species, but humans still present, the developers WILL have to go out of their way to not have them to be the most-played species. I’ve seen this quite a lot in Sword of the Stars (1), a space empire 4X that was actually picked up by Paradox when their former publisher went broke due to the Great Recession (and the influence of which can be seen somewhat in the PDS game Stellaris) :
Humans were designed to be a “medium-advanced” species/faction, with their movement being based around fixed, to be discovered during exploration “nodelines”, which quite complicates the turn-based fuel logistics (especially considering SotS1′ fully 3D maps) as well as the nodepoint-only retreat during the real-time combat.
And despite this being pointed out in both the manual and in the game, you would still see time and time again new players defaulting to Humans, rather than the “newbie-grade” lizard-like Tarkas (with full freedom of movement, except a cheap tech required to change the course of a fleet) or telepathic dolphin Liir (also full freedom, but get slower near star systems).
It might say something about me that in my first playthrough, my credo was “Everything that is not Mushroom is wicked! Everything that is not Mushroom must become Mushroom!”
While Stellaris is not a good 4X –
(in fact the lack of proper fleet management with even basic 4X features like “flag as guard / jump to next idle fleet” missing has put me off trying other PDS games)
– it *does* seem to have beat even the Space Empires series as a great space empire RPG, with a very wide variety of playable empires and memorable stories !
I think a bigger factor is the gap between completionists and casual players. Obviously, these aren’t distinct categories—it’s more of a spectrum—but with how big Elden Ring is, and how it (I would argue) encourages the player to focus more on its open world than its central plot, virtually all players who get around to doing an ending are going to be closer to the completionist end of the spectrum, making them more likely to come across and want to complete relevant sidequests. How many players are going to play Elden Ring diligently enough to reach the endgame, then just take the laziest ending because it’s there?
You could argue that this is just a different way of framing the same argument, but I think focusing on completionists in contrast to (so-called) “casuals” is more enlightening than saying “someone who does X is probably also gonna do Y, Q, and F.”
Considering it also has Good-exclusive mythic paths? It’s probably about the same as other RPGs.
It occurs to me that completionists are also likely to come away with a much more distinct experience of a game than casuals.
If you’re a completionist and you spend 120 hours engaging with every NPC you can find and really learning which strategies are best and finding out how the game ends if you make THIS choice or THAT choice… You are going to have a pretty marked experience coming out of that game. Any messages it has to send, are messages you are likely to have noticed.
If you’re a comparative casual player and you play a game for 12 hours, then set it aside with the vast majority of the content and playthrough options untouched, nothing the game does is going to leave as marked an impact on you. It’s just impossible, because you haven’t put as much time, energy, and thought into it.
So, going just by Steam’s achievement statistics, Angel at 21.2% is the most commonly picked mythic path, followed by Azata (18.3%), Lich (11.5%), Trickster (10%), Aeon (9.1%), and Demon (8.3%) is the least commonly picked of the main mythic paths (Devil, Dragon and Legend are much lower than any of the others, but are only available towards the very end of the game). That said, the game pushes you fairly heavily towards Angel plotwise in the early game, Aeon is easy to miss or lock yourself out of, and most of the paths have a fair amount of wiggle room for roleplaying a good or evil character, with some (Azata, Aeon) having special content for being corrupted away from your intended alignment.
I think the notable one here is that Lich is so high, which seems (anecdotally, from what I’ve seen online) partially due to Lawful Neutral “I did what I had to do” liches being a popular playthrough and partially because the necromancer power fantasy is an underserved niche.
Concerning the book recommendation for R.H. Spector’s work, I do wonder in a tangential way why naval warfare tends to get overlooked, especially in Antiquity. EVERYONE remembers the battle of Cannae and what an utter disaster it was for Rome. The retreat away from Bagradas and losing that fleet in a storm probably cost the Romans more men than Cannes did, but at least as an amateur who talks to other interested amateurs, it is far less well known; similarly, people often don’t seem to grasp just how many men Carthage was willing and able to throw into the meat grinder that was the first Punic war, and I suspect that the biggest battles happening on the water has something to do with that. Any idea why that might be?
Shifting over to game mechanics, I wonder how game devs navigate through competing design principles when they argue for diametrically opposed mechanics. Consider your bog-standard heroic CRPG, something like Baldur’s Gate, Dragon Age Origins, Neverwinter Nights, Mass Effect, etc. Theres usually some crisis going on that you, the Hero, are the only one that can fix it. People are dying, nations or even the world is in danger, and it gets worse every day you’re not killing the BBEG. Verisimilitude would argue this should force mechanics that make you hurry before the bad guys kill your friends and burn down the place you’re basing out of. But hurrying means you miss content, and you usually want to portray the content, so the need to hurry often gets simply dropped into the vat of ludonarrative dissonance. I’ve only ever seen one RPG that really forced you to hurry in the brutal logic it’s story demanded, and that was a fan made Blades of Exile scenario built from an indie developer’s engine.
It is probably a question better asked to a game dev, but I wonder how those conflicts are considered during the crafting of a game.
There’s a couple of examples: Fallout 1 has a set of time limits, and after a point the BBEEG starts killing settlements (and if he kills them all you lose) Star Control 2 has something similar, if you take too long the big bad alien civil war ends nad the winners starts travelling the map killing all other sentient species. (until they *finish* doing so it’s actually possible to still win, by picking up the plot coupons in the now exterminated alien areas)
I’d also point out that upstream from these design decisions are matters like budget and overhead costs, which is why two of the most overt examples of a time limit related to the scenario are from games that are 25 and 30 years old this year- because making a CRPG to market standards is an expensive proposition that requires a lot of nonlinear costs be paid, the ideal game from the producer’s seat is one that has as much content as they can deliver for the budget, so that players feel like they’ve gotten their money’s worth.
In turn, though, factors that deliberately limit content accessibility work against that, not to the point of them totally disappearing from most games, but to the point where urgent scenarios being reflected in the gameplay of a CRPG is absent from traditional developers and rare even among independent developers, who still need to compete in the marketplace traditional developers shape.
So you’d really be limited to deliberate “artsy” games that are explicitly forgoing marketability… but in turn, these usually are also made within the realities of budgets, time, and player response, so CRPGs tend to be a rare genre for them.
I don’t feel like Salamis, the defeat of the Armada, or Trafalgar are overlooked. I haven’t played a computer game in thirty years–they have changed a bit–but naval history still resonates in popular culture generally. There’s a bar in NYC called “Drake’s Drum,” and a leading mutual fund company called the Vanguard Group, for example.
I just completed a play through of Pathfinder:Wrath of the Righteous, which is mainly a traditional CRPG but has an integral side game of running a HOMM3 style turn based strategy game because the concept is your rise to lead a crusade against a demonic invasion of the material plane.
Without getting into spoilers, this clearly created some tensions in game design and plot, but it also created some fridge brilliance (player rage that an NPC wrecked your progress in the campaign game is a major in character justification to go down the evil paths of the plot).
One way around that could be for your sidequests to somehow slow down the enemy’s plan, or improve you or your allies’ productive capacity or ability to resist. That could justify completionism over speedrunning.
That is a beautifully preserved fort! I wouldn’t have thought that those dirt/wooden foundations would still be so wonderfully visible after centuries of laying under moving water.
That square pattern indicates archaeologists have been at work. They’ve cleaned it up quite a bit, making the difference much easier to see. Standard procedure is to go meter-by-meter, excavating a certain depth at a time. Then on well-run sites (the Gray Fossil Site in Tennessee is a good example of this procedure) you store the dirt from each meter excavation in a box for later analysis. It’s really hard to identify things like sesimoid or carpel/tarsal bones in the subsurface, and that stuff matters. It can help with things like identifying minimum possible number of bodies, for example. You also can look at pollen and small to microscopic critters and build a picture of the environment, including microhabitats, if you’re careful enough.
My guess is that the earthworks were somewhat compacted when they were put in place–either heavy stuff was put on them, lots of people walked on them, the dirt was pounded into place, or all of the above. Which makes sense. If you put a wooden stake in loose dirt it’ll just tip over, after all. So you’d want a certain amount of compaction to the soil. Given that these were Romans, I’d assume they’d have had a standardized way of doing this, both because the Roman army doubled as construction workers and because, well, Rome.
Fluvial sediment–stuff in rivers–tends to be loose muck. It’s carried either by saltation–grains bouncing around the bottom–or sediment load–stuff floating in the water. It then gets stirred up by fish, crustaceans, worms, birds, people, plants, and the like (termed bioturbation). Makes for a nice mucky silt. Even when dry, it’s VERY different from non-river-sediment. It’s finer-grained, it’s looser, it’s got different structure to it, it often has different minerals; anyone with a background in sedimentology will instantly see the difference. I remember once finding the bank of a 5-million-year-old river that way, and the difference was still very clear. Archaeologists aren’t any worse at this sort of thing than paleontologists; in fact, in the compliance world (NEPA requires both be monitored for if there’s a chance we’ll find anything) they work for each other. While I’d love to see a sediment description to confirm my speculations, I’ve little doubt that they’re fully qualified to make this assessment. Any competent field worker who routinely plays in the dirt is going to be able to do this fairly well, and there are a number of standard tests that can be conducted if there’s any doubt.
Looking around a bit, it looks likes it was only underwater for a few decades.
“In 1949 it was flooded under the As Conchas Reservoir, property of Fenosa (Northwest Electric Power).”
I remember trying to play Fallout as evil and having a lot of trouble continuing to behave that way, even though the game seems pretty neutral about it in a plot sense. Over various play-throughs I’d grown fond of this one daffy merchant in this one town, and as an evil player I nuked the town. That didn’t kill her; I found her later as an irradiated ghoul (it’s a Fallout thing), seemingly as cheerful and daffy as ever, and yet it pained me deeply.
Which is weird. Unless…are these bits and bytes alive in some sense?
I’ll paraphrase from Terry Pratchett’s “Small Gods”: What it is, is that YOU’RE alive, and YOU’RE a person.
Bits and bytes are just bits and bytes, just as clay is clay, and paint is paint. Living human people arranged clay into the paleolithic Venus figurine; and arranged paint and oil into the countless images of Mary the Virgin; and arranged bits and bytes into that one NPC; and they did so in order to create the concept of a person. And you are correctly perceiving the concept of a person that was created and presented to you.
Will you forgive me for pointing out that the paleolithic Venus figure was, um, lithic?
It doesn’t detract from you point though, and I’d elaborate that the fashioning of representations is done with the intent that it be received by others as an emotional impression. The example from Fallout is an especially complex one: you’re supposed to have feelings about what you’ve done to the town, because as in all video games you are supposed to have empathy for your own player-character (and will even if you’re a person who has difficulty with empathy in general) AND your feelings are then recontextualized by the discovery that you’ve actually granted someone a form of immortality rather than killing them, with the added spin that the immortality is itself kind of cursed.
Versions of this emotional two-and-a-half-step show up in all of the sequels.
While the best-known paleolithic Venus figure, the Venus of Willendorf, is in limestone, and most others are carved from either stone or ivory, there is at least one ceramic example: the Venus of Dolní Věstonice.
https://en.wikipedia.org/wiki/Venus_of_Dolní_Věstonice
There’s also a huge number of female figurines in ceramic from the Cucuteni-Trypillia cultures, which are Neolithic. That’s probably what I got it crossed with.
Ah, I did not realize! Apologies to Rowan.
How do things like Let’s Play videos play into video game analysis? I’m a crappy Dark Souls player, but I’ve enjoyed the game via some LPs by some pretty good players, enabling me to experience content (and thus, given the game’s nature, lore) that would otherwise have been blocked to me. This means that experiencing content and PLAYING the content are not the same.
There’s a literary analog. I’m crap at literary analysis–I rarely bother to really tease out nuances of references in books I read–but thanks to things like “The Lubber’s Hole” I can experience that content via discussions between people who have made a point to explore it. My experience with the books is therefore different from the experience of merely reading the words on the page.
Let’s Plays (which aren’t all videos! A lot of the oldest Let’s Plays were actually done by taking frequent screenshots and presenting them admixed with text) are an interesting case, because they’re an interpreted, non-interactive version of the original work.
To give a good example of this, there was a game that came out a few years ago called Getting Over It With Bennett Foddy, which is a strange mess of a game where you’re a man in a pot trying to climb a mountain of garbage using a hammer. The point of the work is that it’s incredibly frustrating and that that makes “beating” it ultimately more satisfying, to the point where you aren’t supposed to stream or record the ending (since it’s really for the people who struggled to get there).
For an outside observer, however, the point of the game is that you get to see how the player handles stress and frustration. It’s a wildly different (though still completely valid) way of experiencing Getting Over It. The key thing is that watching Getting Over It isn’t frustrating… except for once the player reaches the ending, since you aren’t allowed to see it.
And that’s before you get into the fact that, by their very nature, a Let’s Play is colored by whoever is playing the game. Think about a Let’s Play of, say, a horror game — watching someone play through the game and jump at every shadow isn’t going to feel the same as watching someone play through the same game and constantly crack jokes.
The only real comparison I can think of is that it’s the difference between watching a movie and having a friend describe a movie to you.
My favourites of the screenshotted Let’s Plays were the huge ones of GTA: San Andreas ‘Grove Street 4 Life’ https://lparchive.org/Grand-Theft-Auto-San-Andreas-(Screenshot)/ and Medieval 2: Total War ‘A Scotsman in Egypt’ https://lparchive.org/A-Scotsman-In-Egypt/. They’re both hilarious and take the experience places it wouldn’t normally go.
Getting Over It and Celeste are good counter examples to each other of the differences between games with the same core concept (“succeeding at difficult things is satisfying”) but radically different ideas of who can and should play those games and who games are for – Getting Over It is elitist and hostile, full of nihilistic sarcasm, and ultimately does not care that some people will never be able to play the game – it’s a meme pretending to be philosophy. Celeste cares deeply about the player, and recognizes that the compulsion to pursue difficult challenges can run directly up against constraints outside of the player control – accessibility options are available to all players to use freely at their own discretion, with no judgement, and the game developers do not discourage streaming the entire game, including the ending and post-game content.
I would also disagree that watching Getting Over It isn’t frustrating for viewers. A good community of viewers (whether for Let’s Play or a streamer) will care about the player, and I’ve definitely seen chats that have encouraged a player to stop rather continue to be frustrated, sharing in the player’s frustration rather than being frustrated with the player. Though I do agree that a Let’s Play or stream of a game mediates that game for viewers and provides different experiences depending on the player. Game developers are more and more recognizing that some folks will be experiencing their games in this manner rather than directly playing the game. It will be interesting to see how that affects design choices.
“Getting Over It is elitist and hostile, full of nihilistic sarcasm, and ultimately does not care that some people will never be able to play the game – it’s a meme pretending to be philosophy.”
As I understand it, Getting Over It is a study of Cynicism–not the modern concept, but the Classical Greek concept, the one where you win arguments by throwing plucked chickens at the speaker and live in a pot (see Diogenes). As I understand it, the philosophy encouraged voluntary hardship, such as going barefoot in winter, in order to enhance mental clarity. All of which makes the difficulty and hostility of the game make sense: It’s providing the player with a means to experience life as a Cynic, if through the glass darkly.
More generally, I think a false dichotomy is being set up here: Playing the game vs Watching the game. Most of the gamers I know do both. At least in my experience my experience watching an LP of a game I’ve played is moderated by my experiences with the game itself, and vice versa. For example, Kikoskia and I have very different playstyles for Minecraft–he favors a more aggressive style, I favor a more defensive style. I’m experiencing Kikoskia’s LP in contrast to my own experiences, and I take the information he provides into my gameplay. They add to one another, in other words.
For the first point, I’m indeed aware that Foddy is probably trying to reference the Classical Greeks, him having been a philosophy student and all. I just also think philosophy has rather moved on in 2000 years and that as most players aren’t familiar with it, and from my familiarity with Foddy’s other games, I’m willing to take an, ahem, cynical attitude to it myself.
If you see a dichotomy in my post, it’s not intended. I play many of the games I watch, before and after. But there are many people who do not play the games they watch, whether through will or circumstance. For example, I’m disabled, and there are games I cannot play, even if I want to. So I watch the ones I can’t play. Some folks simply don’t enjoy playing games but enjoy their narratives, they would never play the game anyway. Those are also “experiencing” the games in question, but it’s mediated through someone else, who is usually not the developer of the game, and that’s an interesting new variable for the study of games as well the development of games.
So you *manifestly* talk with your hands, huh? ISWYDT
I had expected a review of all the museums in Stockholm, but this was pretty interesting
That was actually what Patrons got this month – a rundown of the highlights of the museums we got out to see in Stockholm.
Ah, I see, perhaps I should become a patron sometime. By the way that second picture of you is really good!
“a ship’s crew cannot flee in terror (though a captain might lose their nerve and retreat) nor can they charge with more boldness than the engines allow.”
Though even there I’d note that what “the engines allow” on the day can be more influenced by the current morale and boldness of captain and crew than some might first suspect.
As contrasting examples I’d give you:
* HMS Canopus, the old pre-dreadnaught, leading up to the Battle of Coronel — where her chief engineer’s ongoing mental breakdown led to him dramatically underreporting the speed she was still actually capable of. If he’d be willing to give accurate information Craddock’s other ships might have waited for her and not been destroyed.
* USS Samuel B. Roberts in the Battle of Leyte Gulf — where her captain essential decided there was no tomorrow and so might as well burn out her engines today. The engineers pushed them WAY past their design limits; getting her to 28.7 knots (nearly 20% higher than her designed 24 knot top speed! Which, given the vast amount of extra power each additional knot requires at those speed, probably means she was making around twice her designed emergency maximum power!)
Oh, I was thinking this was going to be about the stokers on the Spanish ships at Santiago. They were given brandy to help their performance, but when that wore off, they slowed down quite a bit.
There’s also the Chinese crew fleeing belowdecks at Yalu river, and I think there may have been a bit of that at Santiago. But those were fairly rare cases.
Re that photo of the Roman fort in Galicia: although I don’t doubt for one moment that climate change exists, is happening at an ever-increasing pace, and is pretty much completely man-made, I can’t help but wonder if these changes in climate occur in cycles? The river must have been this low during Roman times for the fort to be built? Then the river rose, either gradually or in a sudden rush, which must have been as alarming then as we find climate change now. But surely, as I say, this points to cyclic changes in climate? The Romans grew wine in Britain, which we are doing again now. Ah well, just a vague hope that climate change may not bring about the complete death and destruction of us all!
If all the data we had was river levels and British wine, you might have a point. But we also have plenty of data indicating that the current changes in climate are non-cyclical, caused primarily by greenhouse gas levels rising ever since the Industrial Revolution.
But hope is not lost! All we have to do is change our economy to A. stop emitting so much greenhouse gas and B. put more resources into various climate-change-reversal projects (like carbon capture). And those changes are already happening; we just need to encourage things to go faster. Collectively, I mean, not individually.
I’ve got a protest sign and a pitchfork, which one do you want?
Ha, guess I’d best take the sign first; save the pitchfork for when I really want to make a point 😂
Or four points, as the case may be.
Boom boom!
The river might have shifted course, cutting or finding an entirely new channel.
As I noted upthread:
Looking around a bit, it looks likes it was only underwater for a few decades.
“In 1949 it was flooded under the As Conchas Reservoir, property of Fenosa (Northwest Electric Power).”
As alluded to in some of the other reply-to-replies…
In this case, the river rose because somebody built a dam and the rising lake waters behind the dam flooded the site. That wasn’t a natural occurrence.
There are natural cycles in climate and gradual natural shifts, including changes in sea level as a result of things like seismic activity. The thing that makes manmade global warming different isn’t that the climate is normally this totally unchanging thing. It’s the pace and the scale. For a good illustration of this, take this comic’s graph here and patiently scroll from top to bottom, reading the timeline gradually as you go.
The end is much more informative if you’ve read all the way from the top to get there.
https://xkcd.com/1732/
You should also consider that the wine the Romans were getting in Britain was likely of a quality that you would literally not be able to give away on the modern market.
Vinegar-quality, I suppose? 😝
There has been a fair amount of effort and ink expended on these sorts of questions–and maybe some progress has been made since I graduated–but in general it is very hard to figure out how ancient foods or beverages actually tasted, given the effects of minuscule chemical changes on sense perceptions. There is also the issue of cultural (or maybe even biological) changes in human perceptions over the millennia.
That said, I tend to agree with the suggestion that we would find garden variety Roman wine undrinkable.
The BBC had a documentary series “Life on the Victorian Farm” a while back. Archaeologists and historians lived as Victorian-era farmers for a year. In a post-experiment interview they asked one of them about the food, and she said she was surprised at what foods tasted really good–the lack of air conditioning, the hard work, the changes in diet, all led her to enjoy foods that she found unpalatable both before and after the event.
It’s experimental evidence supporting the idea that we probably wouldn’t enjoy a lot of stuff our ancestors ate and drank. Even ignoring food production methods, our lifestyles are different enough to have physiological affects on our sense of taste.
Ruth Goodman. Who wrote the *Domestic Revolution* which is particularly good for anyone wanting to learn about how the Industrial Revolution started.
It’s long been said that the best spice is bunger.
Hm. Yeah, I suppose you’d need to explain the intersection of gaming and history very differently to an audience of historians than an audience of gamers.
I’d personally suspect that a significant chunk—perhaps a majority—got an Elden Lord ending for completionism’s sake—partly for a good reason (the most common ending is the hardest one, which makes me think that most players either drop the game before the ending or do as much as they can first), and partly for a bad reason (I spend a lot of time in online communities with the sort of people most likely to be completionist about games, which biases my assumptions about “gamers,” and it’s hard to correct for that sort of thing).
But we’re not likely to know unless someone contacts From Software and convinces them to share the relevant statistics.
Which doubtless makes academic papers about games very frustrating. Books, you can just quote a passage as it exists to illustrate your point; when a film’s script is inadequate, you can bring in a frame to show how details are told visually. Even if something is conveyed through motion, you can get the gist across by putting a few well-chosen frames in sequence.
But how can you show how a creative choice affects the play experience? You can’t. You can only tell, which makes it difficult to parse situations where different academics say different things. If it’s a book or a film, the papers can include a bit of context to help viewers judge whether each is right or wrong; without such context, a reader has to either play the game or just pick one.
To envision how this might play out in practice, look at any YouTube comment section in a gaming video essay and imagine all participants are using properly-spelled formal language.
One complaint I’ve heard about games using some form of karma meter like Knights of the Old Republic or Mass Effect is they give rewards for a pure playthrough. So that tampers with player choice because many are trying to play full Paragon or Renegade, instead of being nice to those who deserve it and punching those who are asking for it.
A mechanic I’d like to see is an extension of GTA’s wanted level, that already exists in TTRPGs. Blades in the Dark calls it Heat and Cthulhu City has Suspicion. It creates a paranoid creeping dread as the opposition close in and become more certain it was you. The player may not even be able to see the stat and just see its effects – like people tell you there were sinister types asking questions, or you start to think you’re being followed. It can have a delayed effect if there were witnesses or evidence left behind, but the authorities haven’t found them yet. You may think you got away with it but days or weeks later they finally put all the pieces together and then there’s a SWAT team outside your house.
For my part, I’d prefer a Faction system–one where you build reputation within individual factions, each with its own goals, morals, and opinions. From what little I’ve played of “Fallout: New Vegas” it seems to have something like that.
Humans don’t have an objective morality. Each group decides its own, and often what’s considered proper–even obligatory–in one is considered improper or even evil in another. To give an example applicable to a post-apocalyptical wasteland, look at food supplies. They are by definition limited, and each faction’s survival requires them to have enough, plus some extra. This means that what’s good for one faction is going to be bad for others. Looting other settlements is going to be a viable system, both of obtaining food and of population control (if you take over the settlement you can afford more people; if not, you have less people to feed). On the flip side, some factions may prefer stability and security, focusing on increasing their own production.
It would be even better if the developers could pick some virtues and vices, and have the player’s rating with each faction be related to that. Raiders would obviously favor bravery and physical strength; more fortified areas may favor intellect and acting methodically.
It would be a lot more difficult to put together, as you’d need to develop different end-games for each faction. Still, I think it would be more realistic. This would allow the player not just to punch those who deserve it, but to CHOOSE who deserves it.
The other thing about video game morality systems is that the “Evil/Renegade/Darkside” options feature a lot of jerkery purely for the sake of jerkery, even when it makes literally no sense.
For example, take Mass Effect. In the second game, your job is to recruit a specialist team to go fight some aliens on a very dangerous mission where it is very important that everyone is on the same page. At one point, while on a mission to get a team member’s head screwed on straight, said team member promises to keep an informant’s identity secret when he tells you who hired the team member’s son to perform an assassination. When you find said person, a local minor crime lord, after you get the necessary info he asks you who ratted him out.
The paragon option, of course, is to tell him to go jump in a lake. The fact that the renegade option, however, is to tell him the name of the informant is dumb. “Why yes, I am going to anger this incredibly deadly person who I will be relying on to watch my back on a very important mission in order to curry favor with this guy who I will probably never see again.” Having the Renegade option be “Don’t tell him in order to not anger the guy you recruited,” and the Paragon option be “Don’t tell him because you told the informant you wouldn’t” would preserve the dichotomy while not having the Renegade option be purely for the evulz, which would help make some of the other, more fraught Paragon/Renegade choices be rather less obvious as to which you should choose.
Ah, the classic “chaotic stupid”!
More Stupid Evil, I suspect.
I remember Planescape: Torment as one of the few games I have played that would allow you to make the same decision while giving different reasons, with the game systems taking account of the distinction.
“…options feature a lot of jerkery purely for the sake of jerkery, even when it makes literally no sense.”
So, I can tell you’ve not spent a lot of time making phone calls to voters, or knocking on doors to talk to people about an upcoming election. In the specific example you give, yeah, the elegant thing would be for the Renegade option to be ‘lie and pin it on somebody you want killed;’ but that’s for making a satisfying narrative. If you’re writing for verisimilitude you have to remember that mean people are mean because it feels good–and because they don’t know how to care about the people they’re hurting.
Some mean people are mean because it feels good.
Other mean people are mean because it gets them what they want.
Being forced to do the former instead of the latter is still not being allowed to play the character you want to play.
Yes, I am quite aware that people can be petty jerks. However, the problem with “writing for verisimilitude” argument is that the sort of person who is going to be your standard epic-quest RPG protagonist might be evil, but they are extraordinarily unlikely to be petty, and they are extremely likely to be smart.
IOTW, their only consideration will be “Does this help me accomplish my objectives?” with the rightness of the action not even being given a consideration. They are not going to be thinking “What’s the best way to make everyone hate my guts?”
One tiny nitpick: the original publication date for Spector’s book (which I first read about ten or fifteen years ago and quite liked) is 2001, not 2021.
Whoops. Silly typo. Fixed.
I think perhaps the most interesting example of what Global achievement stats can tell us about a game’s playerbase comes from Getting Over It with Bennet Foddy.
For those unfamiliar, Getting Over It is a notoriously difficult and brutally punishing game in which you climb a mountain, which a few years ago was a favourite of streamers, because it’s ability to generate moments of pure infuriation on the part of the player makes for very engaging stream content. It is a game almost perfectly crafted to induce you to ragequit.
This game has only three achievements: beat the game, beat the game twice, beat the game 50 times. 8.1% of players have beaten the game, a testament to the difficulty and rage mentioned above. But notably, 2.2% of players have beaten the game *fifty* times. Which means that out of everyone who climbed the mountain once, more than a quarter of those people would go on to do it over another forty-nine times. For the achievement, the satisfaction of a fast climb, or just because it is there? Whatever the reason, clearly the first climb acts as a filter for the player base, only certain kinds of players ever make it to the top, and those kinds of players have a measurably very unusual trait of being willing to complete a game like it over and over.
In any case, I highly recommend Getting Over It, Bret, if you ever decide to go into streaming! 😛
A good game in simulating evil is “Tyranny” where you play the bad guys:
https://www.youtube.com/watch?v=lC3mehnQF0Q
Or the temptation of the “One Ring” in LOTR series.
What I mostly took away from Tyranny is the evil is a terrible government model; the big players spend way more time fighting each other than dealing with the rebels that they are supposedly in the area to deal with!
This is one of those cases where I think, from my reading of the game, this is part of the point. Albeit it has been a while. The evil overlord is very far away and out of sight, you are only told and never shown their attitudes towards everything and thus there’s a lot of room for interpretation.
However the way I see it: the Disfavoured and the Scarlet Chorus both cause Kyros more problems overall than they solve. Their infighting is getting out of hand and their tactics are noted on more than one occasion as having intensified and prolonged the rebellion that they are there to deal with. Add to that that the Chorus are essentially completely uncontrollable and the Disfavoured personally loyal by both social and familial bond to their leader rather than the Overlord and you have a recipe for two forces that frankly the evil overlord has ever incentive to want to dispose of now that their conquest of the world is finished.
In conclusion I think that there’s a reading of Tyranny where the intention of you being sent to the south to read the edit was always meant to end in failure and the destruction of everybody present. It suits the evil overlord’s machinations much better to dispose of minions who have all outlived their usefulness alongside the rebels.
Regarding naval warfare — the Battle off Samar is an excellent example of how meat and minds can function even in naval warfare. Admiral Kurita was beaten by Taffy 3 not because his ships were sunk — I’m not sure if the IJN Yamato was even scratched — but because his will broke. He thought he was facing a much larger force.
But I think it is also metal that broke Kurita’s will, not merely the ferocity of his enemy. Specifically, HIS metal. Kurita was in command of Japan’s biggest and baddest warship, one of only two of its kind, something Japan couldn’t possibly make again with how stretched thin their supplies were. So Kurita treated his battleship like it was a shiny new car that he didn’t want to scratch even once. Even though battleships are specifically built to be scratched. They are called BATTLE ships for a reason.
Kurita must have thought that he was in danger of losing the IJN’s most precious resource. If his ship started to go down, the culture of the Japanese military would have demanded he sink with the ship. So the only way to save his hide, in his mind, was to turn and run.