A dungeon core? He was a dungeon core again?! What happened to the promise of a meaningful life?

I've lived thousands of years this way already, so I know what kind of life it will be.

The desire to grow and become stronger would only lead to more desire, never to gratification. The cycle would continue on until he was eventually killed again, and ultimately nothing of value would have been achieved. His body would crumble into dust and all of his efforts would become particles of light and vanish.

Why? He had finally escaped that miserable existence. To make him a dungeon core all over again was just too cruel.

… Should he just end it before things became truly painful? He had shrunk down to a size that could fit in the palm of a hand and didn't have the mana processing ability to self-destruct like last time. Perhaps he could open a path directly to him and let himself be taken away. Then again, Archimedes had no idea if he would ever be allowed to die if he were to be claimed as a power source by humans.

Was that what the voice of the world was implying? Would his life have some value if he sacrificed it to fuel mortal civilization? Could he delude himself into being a happy slave?

The palm-sized purple marble that was Archimedes’ body trembled. There was no physical way for him to cry, though he wished he could; he just felt so deeply betrayed. Archimedes was only a dungeon core in the end. He may have lived for thousands of years, but his intellectual capacity was only around the level of a slightly clever human, as far as he figured. Most of a dungeon’s ability was devoted to dungeon management. It was necessary to grow quite large before self awareness and active thought had the space to develop. Archimedes was glad that he had retained his knowledge from his long experiences, but he still wasn't able to understand what something like the voice of the world was thinking.

He had no choice but to live this new life, but he refused to live it ignorantly. He had to know why things had happened this way and what the voice of the world was thinking.

And if he wasn't smart enough to do that, he would create a monster who could do it for him. He was a dungeon after all! Sure it would take a while to raise such an intelligent monster, but…

That wouldn't work.

No matter how smart a monster he made, it would only be able to learn things that could be learned within the dungeon. Intelligence was only half the problem here: Archimedes needed someone with experience in order to answer his questions. He needed someone from outside.

That was… a problem. As far as Archimedes’ experience led him to believe, among humans, strength tended to be inversely proportional to intellect. What feeble sage would venture into a dangerous dungeon? Worse, what meathead with a sword could explain the motivations of the voice of the world?

Currently, Archimedes was a palm-sized purple marble floating in a tiny hole somewhere underground. This was actually the safest situation he would likely find himself in until he had built up a good ten floors worth of traps and monsters for protection. His dungeon didn’t even have an entrance yet, and so no one could possibly invade the space he occupied. In other words, he was in no hurry at all, and could safely take the time to think of a plan.

Beyond the earth that encased him, the sky turned lazily, and Archimedes’ mind made steady revolutions of its own.

He wanted to drive away the foolish, but tempt and tantalize the wise. What sort of dungeon would appeal to the most intelligent creatures in the world?

Should he fill his halls with magical constructs and the resources valuable to mages? That might not work. Those who were skilled with spells weren’t necessarily well-versed in other areas of study. He had seen plenty that weren’t worth much beyond a few pointlessly flashy explosions. That one woman from the party that conquered him, was there even anything going on beneath her pointy hat?

It would do better to focus on scholars then, right? Perhaps Archimedes could replicate various technologies he had seen come into his dungeon over his thousands of years of life. He recalled seeing a few humans who seemed interested in studying things from other eras. Bah, but that had the same problem of drawing in specialists who might know nothing about the matter at hand.

What sort of place would attract bizzare people who taught themselves on just about everything? What would draw those people who simply amassed knowledge for the heck of it? Archimedes let three days slip by as he did everything he could to figure this out. If even this was beyond his ability, then perhaps he deserved another several thousand years of punishment, just for being so plainly stupid. Ultimately, though, he did seem to come to some sort of answer.

Even after Archimedes had obtained immense power, the desire to improve never went away, and he was never satisfied. He started to desperately crave for powerful people to invade his halls, so he could test himself against them. Power and knowledge were different fields of obsession, but the apex was the apex. Someone as knowledgeable as Archimedes had been powerful might be craving something to put their wit and wisdom to the test.

What if he made a dungeon where the main reward was the challenge itself, and that challenge was based on the intelligence of the invader? A puzzle dungeon, so to speak, could potentially draw in that bizzare type of person he needed. Additionally, puzzles were fairly effective at stopping muscle brains in their tracks, so he wouldn’t be easily taken and enslaved by humans.

It was good to have a direction to work in. Though, designing challenging puzzles on his own was a bit of a challenge, and there may have been a better method he’d simply missed. Considering that… he might need a bit of help.

Archimedes focused on the various windows that the voice of the world had amassed in front of him, which he had shamelessly ignored for three whole days while he mulled things over.

Archimedes simply closed the window congratulating him on his reincarnation, then chose Increase Creature Wisdom as his initial boon. He stared in contemplation at his options for a starting monster for a fair length of time.

Please select a starting monster. Goblin (F) Kobold (F) Slime (F) Wolfbat (F)

Archimedes had focussed his previous dungeon on power, so he knew that all of these monsters could become remarkably strong in time. His situation had changed a great deal since back then, and now the intelligence of his monsters was something to seriously consider. The end results of the Slime and Kobold evolutions were probably some of the smartest monsters he could think of that he could obtain within a few centuries, but there were drawbacks. Slimes took an incredibly long time to reach that immaculate form. Kobolds were a bit faster, but they focused on strength for most of their lifespans, so he would receive almost no intellectual value from them in the early stages. Goblins grew quickly, but their final intelligence wasn’t any better than Archimedes’, and they were weak on top of that. Goblins were off the table.

Wolfbats, now those were strange monsters. Starting at their first evolution, you could drive them to become more wolf-like and focus on power, or more bat-like and focus on intelligence. That wasn’t to say they suddenly became mages or warriors with a single evolution, but they certainly had the potential to gain intelligence quickly. The drawback would be that their strength would develop more slowly, and their higher evolutions weren’t very loyal as far as dungeon monsters went. Of course, they wouldn’t easily betray the only place they could live in.

Mentally shrugging, Archimedes decided to go all in on intelligence and chose the Wolfbat as his starting monster.

Before he could actually start creating things, he would have to create a room for his core and a path to it. Fortunately, every young dungeon core could place one of each of those without using any precious mana. Archimedes centered his free room around his core and waited for about a minute as the surrounding soil and stones turned into particles of light in the air.

New creatures added: dung beetle, earthworm, fern, mycelium, white fungus

New materials added: dirt, stone

New construct added: small rock

Every new thing that was absorbed by his dungeon became a blueprint so that he would instinctively know how to make more of it. For young dungeons who were better at converting ether into mana than they were at thinking, having models for how to convert basic mana into various things was essential to their ability to grow. For Archimedes though, being handed children’s recipes was somewhat insulting. He was a former A+ rank dungeon after all, and he could visualize the structures just fine on his own.

The next thing to do was complete the path to his core room by placing his one free corridor. Archimedes could perceive everything within his dungeon, but his senses couldn't move far beyond it. As soon as he tried to look too far outside, everything went white. Still, he was close enough to the surface that he could see it.

He seemed to be buried in the side of a mountain. Choosing a level path was easy, and it only took about two minutes to dig everything out of the way. Soon, a straight and simple path, two meters wide, two meters tall, and ten meters long spread out before his basic room of 3 meters on all sides.

Unauthorized usage: this tale is on Amazon without the author's consent. Report any sightings.

New creatures added: moth

New materials added: grass, weeds

Starlight filtered in from the outside, making his spherical body glitter. From what he could see, the mountain wasn’t too terribly steep, but… would anybody be able to find him here? It might be necessary to throw some treasure around just to spread his name a bit. For now, he was just happy to be something that could be called a dungeon, rather than just a fancy rock buried in the dirt. The voice of the world also seemed to deem it sufficient.

Congratulations! A basic dungeon path has been established. All basic abilities unsealed.

A second window happily followed the first.

Dungeon threshold has been met. Intelligence threshold has been met. Would you like to name yourself?

“I already have a name, and you know it,” Archimedes scolded the second window. It stayed quietly where it had appeared, and soon Archimedes began to feel cold. His current self was an infant dungeon core. Could it be that he was unnamed? How truly futile his previous life of thousands of years had been: nothing at all was left of it, not even his name.

I’m sorry

A weak window fluttered into view, causing Archimedes to sigh internally. “My name is Archimedes,” he gave in and murmured.

F rank dungeon core has been assigned the name Archimedes!

It was painful to have everything but his memories stripped from him, but Archimedes was glad to see that the voice of the world hadn’t forgotten about him or stopped caring. He had no idea what sort of creature it was, or if it was alive at all, but it had been the only consistent companion he had had over the course of his long existence. Archimedes wanted to believe that it really did have good intentions in sending him here. He just didn’t know what those good intentions were or which part exactly he was supposed to be happy about.

That was just all the more reason to quickly build up his dungeon and call out a sage to explain it to him. But before he could do that, he had to check exactly how far he had fallen, or rather, remind himself what the resources of a newborn dungeon looked like.

Archimedes Species: Dungeon Rank: F Mana: 100 Accumulation: 10/h Abilities Absorb Substance, Create Creature, Create Structure, Destroy Creation, Dungeon Authority, Ether Conversion, Increase Creature Wisdom, Manage Loot, Move Core

It brought Archimedes physical pain to look at such a pathetic status. His rank, which was essentially determined by the highest level of creature he could create, was the lowest it could possibly be. Previously, Archimedes had been on the cusp between A rank and S rank, the second and first highest ranks, respectively. He had already been defeating hero parties, so it was really only a matter of time before his precious greater Void Dragon evolved again.

Now, his monsters could conceivably be defeated by an adult human with a thick stick, and his other abilities had been similarly demoted. What the hell was he supposed to do with a hundred mana? Even carving out another room at its most efficient size, 27 cubic meters, would cost 1350 mana, assuming the hardest thing he had to dig through was dirt. To add insult to injury, he could only process ten mana every hour to replenish himself!

The mana deficiency was bad, but the first thing Archimedes had to do was protect himself, so he spent 30 mana to create a Wolfbat. He wove and pulled the mana within his crystalline body, slowly guiding it into the general shape of the relevant monster. Archimedes carved bones, clumped organs, grafted skin, and dusted fur, spreading an ethereal blue light that he could see as clearly as the walls around him. Every layer of the unfinished creature could be solidly seen, and every layer could be directly seen through, down to its vivacious center. To the mundane creature, this whole glorious show would go unseen.

After about five minutes, a humble creature took its first confused look at the world. It gazed at the starlight outside, at the dirt walls, and at the floating purple crystal which was its God, and the creature bowed its little head.

Wolfbats were cat-sized mammals with four paws, a fluffy tail, a pair of wings attached to the front limbs, and a long, pig-like snout growing sharp fangs. They had fluffy gray fur covering everything but their nose and their black leathery wings. Their fluffy triangular ears and big red eyes were rather cute if you weren’t being attacked by the fangs right below them.

Thirty mana seemed appropriate for a little thing like this, though Archimedes was looking forward to cutting down the time it took to create.

Archimedes sent a small pulse out from his body, and the Wolfbat seemed to understand that it could lift its head. It waited eagerly for orders of some kind.

“Protect the hall before this room,” Archimedes ordered. The small monster stood at attention and then trotted off like a fox, fluffy tail swaying behind it.

Archimedes sighed lightly and turned his attention to his next immediate concern, which was to increase his mana accumulation. Frankly, the trick to it was easy, but few young dungeon cores made use of it. A dungeon core that didn’t know about the trick would think of the initial investment as a waste of mana, never attempt it, and never reap the benefits. Young dungeon cores were adept at playing a game of costs and balances, but they were weak at experimenting. They simply let the world throw things at them first, and then adapted in reaction to that. Luckily, Archimedes was only young on the outside.

Of his 70 remaining mana, Archimedes promptly spent 27 on two square meters of mycelium, two square meters of white fungus, and one square meter of common weeds. He delicately sculpted the mana into their shapes, strung delicate fibers, and poked porous holes. Once the simple volumes of the creatures were crafted, he planted them where he pleased, and then they entered a state visible to a human. These were some of the most resilient, fastest growing creations he knew about that he could make this early. They would spread dramatically as they absorbed the ambient mana in the dungeon. More importantly, having more living things in the dungeon would draw more Ether to this place, which would allow him to accumulate mana more quickly. Plants and fungi were far more efficient for this purpose than animals. Already his mana accumulation had increased to 11 per hour.

He was down to 43 mana. Archimedes used 5 of it to dig a small hole ten centimeters deep and one square meter wide, then another 15 to fill it with clean water. Just the presence of water nearby would dramatically speed up how quickly the weeds, moss, and fungi spread. That left a meager 23 mana for him to work with. Archimedes patiently waited a few minutes for one additional mana to enter his stores, then he exhausted all of it to create a four square meter colony of glow moss on the ceiling.

Mycelium would grow within the soil, weeds on the floor, moss on the ceiling, and white fungus would fill the gaps by spreading up the walls. In this way, Archimedes would make the most efficient use of the space he had. Also, the ghostly lighting provided by the glow moss would save him some mana on torches or whatever else for lighting.

The process of creation and destruction as experienced by a dungeon was a great deal different than how a more talkative dungeon might explain it to humans. In words, it came across as a delicate art form, as a thing of emotion. In reality, it was more similar to the magic that humans engraved into items: there was a great deal of calculation and precisely detailed work involved that could be seen as a meditative exercise at best. Frankly, it was even more difficult than enchanting, as the dungeon wrote nothing down and used no external materials. Every calculation was performed by the dungeon core, where they were memorized. Additional calculations followed for each part of every atom making up that creation, and not a single one could be neglected or misplaced. The whole of a creation, living or otherwise, was reduced down to a form of magical binary, with every calculation performed using only two “numbers,” represented by either the presence or absence of mana. When everything had been gathered perfectly together, the core then had to use one part of its body to remember the long form of the creature, and another part to compress that into something that was more true to life. When the compression was complete, the long calculation was pushed aside and the mana was forcibly bound by high-power ether reactions, so strong that it would never willingly unbind again, and could safely exist in physical form.

The blueprints obtained by a dungeon when it absorbed new things were essentially the long calculations. Even if they were saved from having to draw those out, the detailed compression work still awaited them. One of the reasons young dungeons developed so painfully slowly, besides their horrifying lack of mana, was that attempts at creation often misfired: creatures would be bound incorrectly and be unable to sustain life and form; or they would not be bound firmly enough, and rapidly destabilize.

The structure of a dungeon core was fundamentally different than the structure of a human brain. They were creatures specialized in performing these detailed and delicate tasks; gifted at performing long and complex calculations; able to maintain multiple trains of thought at the same time. This ability grew more impressive as the core grew larger. It was often thought that the more “human” thought processes a dungeon displayed were sustained by the body of the dungeon, rather than the core. It was the thoughts and the emotions of the creatures roaming the dungeon’s body that rubbed off on it, making the whole of the structure a thinking, feeling creature.

With zero mana remaining, the thinking, feeling Archimedes could no longer support the typical, calculating thoughts of a dungeon core and became inundated with the memories and emotions that thrived on ether. It was a stressful situation for a young dungeon core, and half of his being was very tired, so Archimedes focused completely on the process of converting ether into mana, artificially hypnotizing himself into something similar to sleep while he recovered.

_________________

Creation Costs:

Spoiler: Spoiler

Monsters (per 1)

Wolfbat: 30 / 5 min

Animals (per 1)

Earthworm: 3 / 15 sec | Dung Beetle: 8 / 25 sec | Moth 6 / 13 sec

Plants (per 1 or per 1 square meter as appropriate)

Glow Moss: 6 / 25 sec | Grass: 5 / 30 sec | Fern: 5 / 25 sec | Weeds: 7 / 13 sec

Fungi (per 1 or per 1 square meter as appropriate)

Mycelium: 4 / 15 sec | White Fungus: 6 / 18 sec

Materials (per 10cm x 1meter x 1meter / realtime)

Dirt: 5 | Stone: 10 | Water: 15

Constructs (per 1 / realtime)

Small Rock: 2 / 2 sec