While Fenn and Grak continued magic item testing, I stared at my character sheet and tried to determine how I was going to spend my points.

What do you want to be when you grow up?

Well, no, that was the wrong question. The real question was the proper way to minmax, given the information I had available to me. Skills had both primary and secondary abilities associated with them, which probably helped to define success and failure in accordance with some hidden rules that I wasn’t privy to. My guess was that chance to Dodge was something like the roll of a die plus skill in Dodge multiplied by the primary stat and also half the secondary stat, but none of that was exposed to me and it seemed like it would be really hard to test, especially if (as I suspected) the rules changed depending on whether I was in combat or not.

One of the core concepts of D&D character optimization was avoiding MAD, multiple ability dependence. The general idea was that if your character could be boiled down to a single ability score, then you could focus all of your various resources on raising that ability, which would make you superior to the guy who spread his resources thinner. This was in part due to how the math and mechanics of D&D worked out; the game made it harder to just infinitely stack bonuses on a single ability in order to combat that hyperspecialization, but it was still one of the best general optimization strategies.

Because there were both primary and secondary abilities for every skill, it was impossible to go truly single ability, because eventually you’d hit the cap on the skill’s secondary ability. However, there were different ways to divide up the various skills:

Skills which have both primary and secondary in the same superclass of either PHY, MEN, or SOC. Those skills can have their cap raised by just putting the points into the superclass ability, which gives a 2:3 payout. (e.g. if One-Handed Weapons has POW and SPD as primary and secondary respectively, and I was only concerned with raising that skill, then I would put points into PHY when the skill was capped on SPD (putting them into POW when it was capped on POW) and attempt to maintain a 5:3 ratio between POW and SPD. If I had done that from the start, then using my lifetime 23 points at level 10 I would have 11 PHY, 17 POW, 10 SPD, and 10 END to show for it, with a cap of 50 on One-Handed Weapons. Going pure PHY instead would result in 14 PHY and 13 for each child stat, with a point left over, resulting in a cap of 39, for a better generalist approach but weaker specialist one.) Skills which have their primary and secondary split between superclasses. This meant that I’d either be pushing points into the superclass abilities on both sides, or I would be “wasting” points on putting them into the specific abilities. (e.g. Parry had SPD as primary and INS as secondary. Going for that skill, and only that skill, the specialist approach would have given me 17 SPD and 10 INS with a cap of 50, but not much else to show for it. The generalist approach would have left me with a 9 in all PHY stats and a 6 in all SOC stats, for a cap of 27 in all skills that didn’t need SOC stats.) Skills that depended on other skills. So far, I only knew of one of those, which was Skin Magic, which required Art to make the tattoos. Skin Magic encompassed more than just the tattoos, since there was apparently scar magic as well, but for all intents and purposes if I wanted to pursue Skin Magic to the exclusion of all other skills, I might be faced with having to raise four different abilities, two each from each of the skills, which could get really complicated depending on which abilities there were and was probably worth avoiding at all costs.

So logically, in order to minmax, I should focus exclusively on those skills that were confined to within a single superclass, which meant that my decision was basically down to three choices, and SOC was a bit of a non-starter, which meant that it came down to either PHY or MEN.

The problem was, the mental stats seemed to be linked mostly to magic, and all the magic I knew sucked. Well, ‘sucked’ was a strong word, but it was blood, bone, skin, and gem magic. Skin magic was the definition of MAD, blood magic was ‘hidden’ MAD because it was mostly a melee-combat oriented magic, bone magic needed a constant supply of bones to drain, and gem magic was not only limited by expensive gems, but also drained mental energy. And worse, I didn’t actually know the secondary abilities for any of those, so they might not even have benefitted from going the superclass ability route (my strongest candidate for Blood Magic’s secondary was END rather than CUN or KNO).

But generally speaking, the conventional wisdom in-universe was that magic was the foundation of civilization and progress. From what I knew of how fights went, there were a variety of mages that you couldn’t go toe-to-toe with in single combat unless you were lucky or smart, and you couldn’t be lucky or smart without diverting points from PHY. With (perhaps) the exception of survivability, MEN seemed like the winner, especially in the long term.

I guess I was also leery of the mental stats for another reason, and that reason was Tom.

The party had been stuck on a puzzle for quite some time, but at least they were working together to figure it out rather than getting bored or frustrated. The puzzle took the form of blocks with lines on top of them, arranged in a particular way and responsive to the touch with displays of lights. There was an underlying, discoverable set of rules that governed the patterns of lights, which they’d slowly figured out and written down, but after that it was a case of beating the puzzle using those rules against an adversarial intelligence. It was, in essence, a boardgame, one that I had developed and then practiced until I was good at it. I had the advantage of that practice, and of not having to discover the rules, but I was pretty sure that I was going to lose, because it was five brains against one.

“Can I roll Intelligence to figure out our next move?” asked Tom. He was playing a gnomish wizard with 22 INT; 18 INT was supposed to be where intelligence maxed out for a human, though that sort of depended on edition. Tom was a great guy, very understanding and funny, but he wasn’t really a thinker, and certainly not on par with his character.

“Eh, no,” I said. “Maybe we’ll do that if you guys can’t beat it on your own.”

Tom shook his fist at the heavens. “Curse my useless gigantic brain!”

“You kind of have to roll with it,” said Craig. “Pun not intended. I mean realistically you’d figure out how to beat the puzzle right away, because you’re going up against Joon, who is decidedly not at 22 INT.”

“Hey,” I said. I mean, it was true, but you don’t just let stuff like that go unchallenged.

“I’m in class with you, you’re like a 15 at best,” said Craig. He held up a finger. “Which is still really good, that’s like a standard deviation above normal.”

“No, it’s not,” said Reimer. “Average for 3d6 is 10.5 with a standard deviation of 3, but that’s for adventurers, not commoners, and if you used 3d6 for commoners you’d get absurd results like one in every 216 people having an IQ of like 145, which is three standard deviations above normal.”

“I didn’t mean literally,” said Craig with a wave of his hand.

“You meant that 15 was figuratively a standard deviation above normal?” asked Reimer. “Do you not know the difference between figurative and literal?”

“We’re saying that Joon would hypothetically have an IQ of 130,” Arthur cut in, “Which would put him in the top 3% of people, or something like that. That doesn’t actually seem that unrealistic to me, there are about 150 people in our class, so he’d be in the top five or six? But the point is that Tom is playing a character who would have an IQ of 160, which would make him Stephen Hawking or Albert Einstein, and while you could fake being as smart as them, in order to come up with the same things that they came up with in the same amount of time, you’d have to actually be as smart as them. Tom could never come up with general relativity just by pretending to be Einstein.”

“Okay,” I said, “But I don’t want to give you a riddle and have you roll some dice and then I tell you the answer to the riddle. That’s lame.”

“I agree,” said Arthur with a shrug. “But it seems like you have to cheat at some point, because you aren’t Stephen Hawking and neither is Tom.”

We never really did figure out a good solution to that problem, which was present not just for intelligence but for other aspects as well. It was a bit easier for wisdom and charisma, but the problem was still there, and Reimer got some painful comeuppance when he tried to play the most charismatic bard in the world and then kept sticking his foot in his mouth.

I was worried on two fronts.

First, while it was clear that the game was actually changing things in my head, I wasn’t exactly clear on the nature of how that was working. Building the rocket had been the most intellectually challenging thing that I had done, and the feeling had been one of practiced ease, like I was used to dealing with those kinds of problems. You know how you do something enough times and it just becomes automatic? That happened with muscle memory, but it also happened with other brain functions, and that’s what engineering was like for me. And how was that happening? Was the game just inserting thoughts into my head in some kind of man-in-the-middle attack assistance? Or was it actually changing the structure of my mind? Would making myself ‘smarter’ or ‘more knowledgeable’ actually do those things, or would it just increase the frequency and severity of thoughts being forced into my brain by the game? (It probably goes without saying that giving more of my thought processes over to the not-quite-unmalicious game was likely a bad thing.)

And second, it was clear that the game was actually changing things in my head. I’d already put points into MEN (and a lesser extent SOC) but still felt like myself. Radically changing my personality wasn’t really something that I was up for, and that seemed like the end result of radically increasing my cunning, knowledge, and wisdom. It was hard to believe that the Juniper that went into the process would resemble the one that came out the other end. I had already started that process and didn’t really feel that much different, but taking the larger leap was frightening, no matter which way the changes were implemented.

So I was cautious. I made a list of problems or questions which I didn’t know how to solve or answer, some of them within the skill domains and some not, trying to spread them around as many different areas of mental faculty as possible, and then including some social and physical problems as well to make sure that I could identify whether there was any spillover. (I would have liked to take some standardized tests, but I couldn’t get the clonal kit to make one for me, let alone two comparable-but-nonidentical ones that I could use as control and experiment.) Here’s a sample of some of the questions:

What is Tsiolkovsky’s rocket equation? (KNO) How do you always win at tic-tac-toe assuming you start first? (CUN) What governs the numbers behind hp increase? (CUN) What governs the numbers behind mental exhaustion? (CUN) Does Amaryllis feel any actual affection for me? (INS) What should I spend my points on? (???) How do I pull social attributes from bones? (Bone Magic, KNO) What film won the Oscar for Best Picture in 1994? (KNO)

Of course, I was just guessing about what might govern those, and I had no real idea how easy or difficult of problems they might be, but I did want a baseline. The list went on for a page and a half, which was helpful to clarify some things for me, even if it might end up being useless.

And with that I stopped dawdling, closed my eyes, and put the first two points into MEN, which increased all of my mental stats by one each. And then I opened my eyes, because part of the experimental procedure I’d designed included doing things in steps as much as possible.

Nothing really seemed all that different to me. I didn’t feel smarter. Then again, I wasn’t really sure whether people ever felt smarter or dumber, even when they were smarter or dumber due to things like lack of sleep, drug use, nootropics, etc. They might recognize some difficulty or ease with problems they were facing, but absent those problems intellect wasn’t something you felt. I did, maybe, feel a little less emotional, which I might have chalked up to WIS, but it might have also been the relief of having set down this path and not immediately been faced with the existential horror of a changed personality at the first step. So I looked down at my sheet of questions.

And yes, some of them I could solve or answer, which I had worked at before and not gotten.

The four values I had seen for hitpoint maximum were 4, 8, 27, and 36. There was a fairly substantial gap between the 8 and 27 observations, which I would have had data for if I had been looking at my health bar when I’d first started bleeding out in the sewers of Silmar City. If hit points were derived from a single attribute, and that attribute was END, then 1=4, 2=8, 3=?, 4=?, 5=27, 6=36, 7=?.

The big clue was that neither increase was prime, and after doing prime factorization, the pattern became 2*2, 2*2*2, ?, ?, 3*3*3, 2*3*3*2, or alternately, 1*2*2, 2*2*2, ?, ?, 3*3*3, 3*3*4, which would put the missing ones as being 2*2*3 and 2*3*3. There were three numbers that incremented one by one until repeating the process. There were other possibilities, but that was the most simple one, and it gave me several testable predictions. The next increase to END would give me a hitpoint maximum of 3*4*4, or 48, and the missing values, which I’d see if I suffered some gruesome fate that once again left me bleeding out, would be 12 and 18.

Mental exhaustion had gone from being out of 12 to being out of 20, which meant that it was following a different trajectory, and one that I couldn’t predict at all with only two data points, even if I could correlate those to my WIS stat. The likely candidate, given what I thought I had figured out about the END to hitpoint connection, was that it was on a similar multiplicative track; 4=12, 5=20, or 4=2*2*3, 5=2*2*5, meaning that the next in the sequence might be either 5*6 (30) or 2*2*7 (28) (among lots of other possibilities). The former was more likely, given that I didn’t expect the curve to be linear, but it was inconclusive. Fortunately, I had seven ability points left, which meant that I could get more data in short order.

First though, I looked at the most-important-for-now question, which was “What should I spend my points on?”. The only new insight I had was that I was capped at 18 for ten different skills, all of which could be raised by just two points into PHY, and while that was a short-term investment, it was a short-term investment that I might need if we were going to be fighting this unicorn, or if I got into any fights along the way. I had raised sixteen skills above 10 and gotten new (minor) Virtues seven times, plus I had unlocked a bunch of stuff for Bone Magic at 10, which kind of counted. And if I were designing a game, then the stuff that didn’t get a virtue at 10 would get one at 20, and “if I were designing the game” still seemed like a good heuristic.

I closed my eyes and put another two points into MEN, then opened my eyes and looked at the sheet in front of me. The mental exhaustion meter went from 20 to 30, which meant that it was probably in the form of (WIS * (WIS - 1)), or maybe just “the increase will increase by two each time”.

But again, I didn’t feel smarter, or wiser, or more knowledgeable. It just felt like there were things I hadn’t spent the time to figure out, or that I had missed for some reason. Like:

“♬ Blackbird singing in the dead of night, take these broken wings and learn to fly ♬”

Skill Unlocked: Music!

I had a book of plays, stories, music, and poetry written by Uther Penndraig, and I had the Art skill, but I just hadn’t thought to try singing to see whether I had gotten a skill unlock. I couldn’t really see how music would be useful, unless it was tied to some magic in a way that I didn’t understand, but once I thought about music it was obvious in retrospect that it was something that I should at least try, even if it was only to fill in the character sheet more. And not only was it a skill that I could probably increase relatively passively by singing a lot while I did other things, there might be virtues to be unlocked that could have benefits not related to music.

I tried poetry too:

There once was a boy from Kans-ass,

Who sat in his Eng-a-lish class,

He came to on a plane,

With no time to explain,

‘fore he was tossed out on his ass

But the game didn’t seem to think this bit of doggerel was worth commenting on, not even to give me a message about Critical Failure!

So after reviewing my questions again and finding nothing really worthy of note, I closed my eyes and put another two points into MEN, increasing my mental stats by one yet again.

WARNING: COGNITIVE ENHANCEMENT HAS CAUSED PERSONALITY NETWORK TO EXCEED HOLISTIC INTEGRITY THRESHOLDS. REVERSING AND REFACTORING EMULATION TO WITHIN HOLISTIC INTEGRITY THRESHOLDS. ENABLING TRI-STRATUM PSEUDO-INTELLIGENCE INJECTION SETTING TO COMPENSATE.

The words blasted across my field of vision as soon as I opened my eyes, bright red, all caps, and in a different font. As soon as I had read them, the words disappeared, and I was left looking at the papers in front of me, not really seeing them.

“Well, fuck,” I said out loud. That was not the direction that I had expected a dose of existential horror to come from. I’d read through the Eclipse Phase sourcebook enough to know what was probably meant by emulation; it meant that my mind had been scanned or taken apart and was running on some non-brain substrate which was pretending to be my brain. If I was an emulation, then I wasn’t even at the level of a brain in a jar, I was a brain being simulated at some level of fidelity on a computer somewhere, which meant that everything I had seen and experienced was probably also a simulation, or something like it.

And maybe it was the fact that I was at 7 WIS now, but I was freaking out about this less than I thought I would. It basically confirmed the simulation theory to me, but that wasn’t actually that surprising. It did explain how I had gone from fifth period English class to Aerb, at least in part. Someone had either scanned in my actual, real-world brain and then made a reconstruction of it on a computer, making some alterations that would give me anterograde amnesia, or maybe they’d constructed me from whole cloth, with no real Juniper Smith on Earth, and possibly no Earth at all.

For what purpose? That was impossible to say. Well, not impossible, because it was a problem with some bounds, but those bounds were huge. Who or what had put me here? Well, it or they had some interest in me specifically, because this world had been created from my thoughts or memories beyond what the dream-skewered seemed to experience. And the message, about exceeding personality bounds, said that they had put in bounds to keep me from becoming not-Juniper.

I had a name for the creator though, one I knew fit as soon as it passed through my mind: the Dungeon Master. I was playing his (or her, or their) game. I had been on the other end of building the worlds, characters, and plot enough times to understand the impulse, if not how or why you would do it like this, without consent or a safety net, besides maybe abject, hedonistic evil.

(There was still the faint possibility that the things I was presented with on the game layer, or this new layer above it, weren’t real, that they were just figments put in place by a unique magic for some reason, but that had become a lot less credible than it had been a few minutes ago, and it had never really been all that credible.)

And really, it didn’t change all that much. I still thought of Fenn and Amaryllis as real, maybe even more real than I’d thought of them before, because now it wouldn’t be that I was real and they were fake, it would be that we were both fake, or maybe just different levels of fake, or the concept of real and fake wasn’t terribly meaningful. Maybe one or both of them had been created to suit me, or to manipulate me by the unseen DM, but that just made them victims. I wasn’t going to pull a Thomas Covenant or anything. The game had given me a warning in the settings screen that if I died I was dead for good. I hadn’t had reason to doubt that.

And Arthur. I was more convinced than ever that he’d been here, on Aerb, somehow, and that if there was ever an end to any of this, if it wasn’t just endless adventuring and leveling up with new companions and better equipment, then the end was going to come when I found the Lost King.

I stood up, feeling like I wanted to go somewhere and do something. I had little doubt that the Dungeon Master had root access to my mind, the ability to wipe away memories or change thoughts. It had been done to me by revision magic, after all, and if it was possible within Aerb then it was possible for whoever controlled the emulation of my mind or the false reality. I hadn’t just caught a glimpse of the layer above the game, I had been allowed to catch that glimpse. It hadn’t been reversed or wiped away. (If there was a message there, I didn’t understand it.)

But the world had rules, and it wasn’t willing to break those rules, or at least it hadn’t shown that it was willing to break them yet. When a person did something, there was a plausible reason for it. Factions and people were driven into conflict by their beliefs and values, not just ‘good’ and ‘evil’. Fenn was Fenn, she wasn’t someone who shifted and twisted to the whims of the game. Even if we had both questioned what the Loyalty number meant, and what impact it might have on her, our growing closeness didn’t seem like it was the work of an unseen Dungeon Master, it seemed like the natural result of our conversations, beliefs, and shared adversities.

“Alright,” I said to the air, keeping my voice low so no one could hear me unless they had a way of monitoring my every word. Most likely, I didn’t even need to speak, because reading my thoughts surely couldn’t have been out of the Dungeon Master’s power, but it was a matter of ceremony. After a moment, I clasped my hands in prayer, which I hadn’t done in a long, long time. “You already know how I feel about God,” I said. “You probably scanned or created the memory of me attacking Victor Clark like a wild animal when he said that God works in mysterious ways.” That had been a week after Arthur died. “You’re inflicting pain on me when you could choose to do otherwise. You’ve made a world of suffering instead of joy. I never found a good argument that eliminated theodicy and I don’t expect that to change on Aerb. So, fuck you.”

I swallowed. “But until and unless I can actually do something about it, I’m not going to waste any more of my life raging against the heavens. I’m going to keep doing what I’ve been doing. I’m going to keep treating it as though it’s real, because you’ve put effort into making it real, the same kind of effort I like to think I always put into making my worlds real.” I hesitated, because I had no leverage, because I didn’t know what the DM wanted, so all I could do was tell him what I wanted and what I hoped for, which he probably already knew.

“I don’t want Fenn to change because her loyalty metric increased. I want her loyalty metric, if she’s going to have one, to just be a reflection of how loyal she is, not an invisible lever controlling her. I want her to be a real person, or at least as real as I am. And … I’m hesitant to want that for everyone else in Aerb, because sometimes existence is pain, but … if they’re not going to be real, or at least as real as I am, then I want them to be real enough that I can’t tell the difference. I want that for the whole of Aerb, alright? I want to poke at the seams and find out that you thought of everything. And at the end of it, I want Arthur back. That’s the only way that this game is ever going to be worth the candle.“

“Amen.”