114: Chapter 114 The Replicants

Low Entropy Workshop did not reply to any comments.

It did not explain how non-periodicity mapped into Physics and engineering.

It did not clarify why the phase angle of the six legs did not use a sixty-degree equal division, which was the most stable and easiest for gait calculations.

It did not release even the crudest CAD sketch.

Nor did it humbly add a line below the video description like other uploaders—promising to do control experiments later, begging for coins and follows.

The account was as quiet as a stone.

However, in the ecology of the internet, silence is often far more destructive than endless chattering.

This instead made the comment section argue even more fiercely, even sparking an uncontrollable cross-category explosion.

If the creator of a hardcore tech video immediately jumped out to frantically explain parameters to the camera, show drawings, or argue with doubters, all the debates would quickly collapse into a judgment of the creator's attitude.

Is he making excuses? Is he covering something up?

But if the creator said nothing and cold-bloodedly laid out that pile of warped, failed scrap parts, dozens of brutal limit-test clips, a risk disclosure written like a disclaimer, and that ugly machine defying human aesthetic instincts for symmetry, the comment section would grow wildly like a Petri dish that lost its apex predator.

Those from the mechanics section stared at its hardware.

Analyzing the modulus of those ratchets frame by frame, trying to deduce whether that complex linkage mechanism would suffer a kinematic dead center at extreme values, they built hundreds of reply threads analyzing force distribution over a ratchet pawl tooth shape that appeared for only half a second.

Those from the robotics and algorithm sections looked at its gait and underlying logic.

Staring at the drunken-looking trajectory on screen, they frantically applied their familiar tripod gait, CPG, impedance control, IMU attitude closed-loop, and ZMP stability criteria, only to realize that this machine had none of these things at all.

Those from the math and science popularization sections latched onto the words "non-periodic."

They dug up the Jiangs Brick theory previously answered by Jiang Lin, searching for that faint ghostly link between the mechanical parameters of finite boundaries and the topology of an infinite plane, discussing whether it was a dimension-striking application of cutting-edge math or merely a brilliantly executed clout-chasing marketing scheme.

As for the peripheral general tech audience who couldn't understand formulas with π and K-values, they only saw an ugly iron bug with six purely electrically driven legs that, after having one leg lifted, bizarrely wriggled to dissipate all the error stress, not only failing to collapse on the spot, but continuing to stumble over obstacles.

Thus, within just forty-eight hours of the video's release, inside a narrow yet highly technical core circle where nobody usually conceded to anyone, the video began to be repeatedly forwarded, dissected, and even autopsied frame by frame.

The first group that truly couldn't hold back and rolled up their sleeves to enter the arena were not top-tier university national laboratories equipped with lavish budgets and five-axis machining centers.

It was a wild uploader who had been grinding in Bilibili's mechanics section for five or six years.

His account was named Engineer Zhou.

He had 230,000 followers.

Old Zhou was not a top-tier creator who relied on editing and special effects to gain views, but in the mechanical DIY section, he was highly recognizable.

He had built ultra-high-precision small tracked EOD vehicles using purely electric drives and servo motors, simple four-axis robotic arms capable of threading needles, and even restored vintage Soviet lathes from the last century with his bare hands. He also frequently used cheap FDM 3D printers and aluminum extrusions to manually craft remarkable, bizarre mechanisms.

His viewers all knew he had a distinct trait: he had a harsh tongue, but his skills and craft were just as solid.

He was a practical practitioner who would roast you on the spot if your code was garbage, or track you down through the ethernet cable to curse at you if your solder joints were cold.

After seeing Low Entropy Workshop's video titled Non-Periodic Gait, Old Zhou disappeared for half a day.

It was not until late that night that he posted a text-only update consisting of just a few lines.

[I finished watching Low Entropy Workshop's bug frame by frame.]

[First, my conclusion: I won't comment on whether his so-called non-periodicity is mathematically useful, nor will I judge whether his engineering hypothesis—that parameter asymmetry can resist cascade failure—holds up statistically. That's for academic papers to handle.]

[But strictly looking at the structure of the mechanical body, folks, there's no need to deify this thing. Six-legged linkages, differential rocker arms, torsion springs, one-way ratchets... the lineage behind these components is crystal clear. Frankly, they're all old things handed down from the Industrial Revolution—he just stuffed them all together.]

[Cut the talk. I'm firing up the machines over these next two days to hand-craft a simplified version with regular phases.]

[I'll use a standard 24V brushless DC motor for pure electric direct drive. I'm not claiming a 1:1 replica of the original, I'm just verifying this: what makes it so hard to piece these old parts together and get it to walk?]

As soon as this update—packed with strong hostility and practical spirit—was posted, the comment section below exploded like a kicked hive.

[Go get 'em, Engineer Zhou! Give these problem-solvers who boast about concepts all day a little machinist shock.]

[The fake-busting is finally starting! I felt all along that video was just making a mystery out of nothing.]

[Don't stir up trouble above. Engineer Zhou said 'verification,' not fake-busting. Let's see him make the legs first—it looks simple, but the machining precision required is definitely not low.]

[Old Zhou, take my advice and remember to make the ratchet—definitely remember that one-way locking ratchet pawl. Without it, this thing will forever be a toy that just spins idly and can't climb bumps at all.]

[Low Entropy Workshop: I welcome anyone on the internet to find fault.]

[Engineer Zhou: Find fault my ass, I'll dig up your foundation directly to see what you're made of.]

Amidst the passionate instigations, genuine experts in the comment section also reminded him.

[Engineer Zhou, don't underestimate the timing system that flashed by in the video. The core difficulty of this thing might not be the linkage length or motor torque at all, but at which exact millisecond the ratchet locks. One wrong step, and the whole system freezes.]

Sitting in front of his computer covered in aluminum shavings, Engineer Zhou looked at this comment, lit a cigarette, and replied quickly and arrogantly.

[I know. I'm more familiar with mechanical timing from repairing car transmissions than anyone.]

[So I won't court death. I'll do a regular phase version first and won't touch his weird non-periodic parameters.]

[Six legs, perfect uniform phase angles of sixty degrees, unified spring stiffness coefficient, unified pushrod stride, unified ESC output. Let's first get this structure walking under a completely symmetrical framework that humans can understand.]

This reply was instantly upvoted to the top.

Because it sounded so reasonable—so reasonable that it aligned with the most fundamental materialistic logic of every engineering student.

Regular phases, unified stiffness, unified stride.

This was the cornerstone built over centuries of human engineering.

Symmetry and uniformity.

...

Thus, the first stone of the replication trend hit the water.

...

Three days later, during the final hours of Old Zhou's promised deadline.

Engineer Zhou's first verification video quietly passed review and was published at one o'clock in the morning.

The video title contained no exclamation marks, instead carrying a rare sense of frustration.

"I Tried to Replicate the Unified Version of Low Entropy Workshop's Mechanical Bug, and It Crashed."

At the start of the video, Old Zhou didn't act sarcastic or tell jokes to the camera like he usually did.

The scene was in his garage under cool-toned lighting.

On the workbench lay a messy pile of black PETG 3D-printed parts freshly scraped off the heated bed, several aviation aluminum bars polished extremely smooth along the edges, an oil-stained hex key, an oscilloscope, and a thick main drive shaft already assembled with bearings.

"First, a disclaimer to clarify things."

Dressed in a gray work outfit with dark circles under his eyes like a panda, Engineer Zhou spoke into the camera in a raspy voice.

"I am not replicating Low Entropy Workshop's original version, nor can I replicate it."

"They didn't release drawings, didn't release specific parameters, didn't provide torque curves for the motor, didn't share the cam profile equations, and didn't even fully show the tooth shape of that core ratchet pawl. Anyone online right now claiming they can do a 1:1 replica of the non-periodic original based on the video is purely talking trash, and you can unfollow them right away."

"I barely slept over these three days, making only a simplified version with regular phases."

Old Zhou zoomed the camera in on the rough six-legged metal frame taking shape on the table.

"Six legs. To make inverse kinematics calculations easier, I set an absolutely uniform sixty-degree phase angle. The springs are standard off-the-shelf parts with completely uniform stiffness coefficient K, uniform stride, and on the drive end, I used a standard 24V brushless motor plus gearbox driving a central main shaft. For the ratchet system, I cut a corner—I didn't make the complex dynamic feedback shown in his video, only the simplest one-way locking for the stance phase."

"My sole purpose in doing this was to strip away the metaphysics of non-periodicity and see where the most basic engineering difficulties lie in these purely mechanical linkage mechanisms that look like products from the last century."

At this point, the bullet comments were still very friendly, filled with an atmosphere of technical discussion.

[Engineer Zhou's attitude is great, very rigorous.]

[Finally, someone who doesn't just talk smack. Engineers should let physical builds do the talking.]

[The aluminum alloy chamfers, the layer lines on these PETG prints... Engineer Zhou's craftsmanship is as pleasing as ever.]

The first half of the video, namely the assembly and no-load test phase, went very smoothly.

Accompanied by fast-forwarded BGM, Old Zhou deftly tightened screws, wired connections, and applied lubricant.

When the brushless motor was connected to the 24V power supply and the linkages began operating precisely, six mechanical legs lifted perfectly into the air.

Once the main shaft spun, the trajectory of each foot tip drew a closed curve in the air: lift—swing forward—drop—push off.

It was so textbook-standard that it could serve as an illustration in Principles of Mechanisms.

Leaning against the workbench, Engineer Zhou looked at the suspended machine whose six legs stepped into thin air in unison like an honor guard. He blew out a long plume of smoke, his expression relatively relaxed.

"See? Forward kinematics for six legs isn't actually hard. As long as your trigonometry is right, it can walk with a perfect gait."

However, just as he cut the power and placed this regular, symmetrical bug shimmering with metallic luster onto the tabletop covered with an anti-slip mat to prepare for its first actual loaded walk—

The next second, Old Zhou pushed the start throttle.

The motor emitted a high-frequency buzz.

Step one: the front-left and rear-right legs touched down simultaneously, supporting the body.

Perfect.

Step two: the two middle legs followed sequentially, smoothly propelling the body forward by two centimeters.

Absolutely perfect.

Just as the machine prepared to take its third step and the third leg was about to touch down—

There was a heart-stopping metallic clack.

The ratchet pawl, which was supposed to engage stably only in the latter half of the stance phase after the foot firmly bore weight, unexpectedly bit into the rack prematurely.

It was like a sprinter sprinting at high speed having their knee violently shattered by an iron rod.

Between the motor's massive torque and the suddenly locked mechanical structure, an opposing force field generated along the entire right kinetic chain.

The entire machine seized up and went rigid on the table, as if someone had grabbed it by the neck.

The motor stalled instantly with a dull, dangerous hum, while the power supply's overload red light flashed frantically.

The aluminum alloy frame supporting the body even visibly twisted and deformed.

"Holy shit!"

Old Zhou's expression changed instantly, and he hurriedly yanked out the power plug.

The garage fell silent in an instant, leaving only the whirring of the electric fan and the residual trembling of the machine frame.

He stared at the machine frozen in a rigid pose on the desk, remaining silent for ten full seconds.

Then he looked up at the camera and let out a bitter smile.

"Alright, boys, the first massive pitfall of this machine has arrived."

The video frame instantly cut to a black-and-white slow-motion replay, complete with red magnified circles highlighting key areas.

Under ten-times slowed footage, everyone could clearly see it.

Due to an extremely tiny irregularity on the tabletop, the body developed a millimeter-level tilt angle during motion.

It was precisely this tilt angle that altered the relative drop distance of the front-right leg.

Before the foot tip had firmly stepped on the table or begun bearing the machine's weight, the tooth tip of the ratchet pawl touched the rack less than a fraction of a phase early, simply due to the shift in linkage angle.

It was just a tiny graze, light as a breeze passing over water.

Yet this single brush pulled the leg—which was still at the end of its swing phase—into stance constraint prematurely.

Once the constraint failed, all subsequent linkage movements were thrown off.

Propagating upstream like a virus, it threw off every subsequent motion on the main shaft, ultimately causing a complete mechanical dead-lock.

In the video, Engineer Zhou picked up a marker and tapped heavily on the edge of the PETG rack where a chip had snapped off.

[part:gemini-3.5-flash-lite]

"In the original video by Low Entropy Workshop, the subtitles at 02:14 casually mentioned that the right rear pawl engaged prematurely during testing, which was due to insufficient non-periodic parameter matching and required subsequent correction."

Old Zhou's voice became unprecedentedly solemn.

"I used to think this was a localized assembly issue caused by machining tolerances that were too large on his piece of junk machine."

"But now, even after using high-precision bearings, cutting aviation aluminum, and keeping tolerances within a tenth of a millimeter, I still inevitably triggered this problem. I understand now—it's not that at all."

"Brothers, the timing fault tolerance of this thing's pawl is appallingly low. As long as the ground is even slightly uneven, the regular-phase linkage will relentlessly amplify this error, leading to premature locking. This is one of the core difficulties of this kind of pure mechanical feedback system."

At this moment, the video's danmu began to flood in, and the atmosphere started to change.

["Hahaha, Engineer Zhou, I could hear that familiar scraping sound right through the screen; it's identical to the sound made by the failed part in the original video."]

["Low Entropy Workshop: I already filmed the failed part where I fell into the trap and showed it to you, why are you refusing to learn your lesson?"]

["I suddenly got goosebumps. Low Entropy Workshop isn't withholding the blueprints to show off; it's because even if he slaps the blueprints right across your face and you know the answer, you still might not be able to tune that timing window."]

["This is the cruelty of pure machinery. If code goes wrong, the worst it does is throw a warning; if machinery goes wrong, stress will teach you how to behave on the spot."]

The video continued, and Old Zhou did not give up.

With a flash of the progress bar, Old Zhou used a file to polish a new version of the pawl's angle, sacrificing some locking strength in exchange for greater tolerance.

This time, the machine could walk.

It wobbled half a meter across the desk.

But a new wave of despair followed closely behind.

To test the load, Old Zhou taped a ballpoint pen with a diameter of about one centimeter onto the desk to simulate a tiny protruding obstacle.

The machine walked over with neat strides.

The exact moment the left front leg stepped onto a fixed 10mm nylon round rod—because Old Zhou used a standard industrial spring with an extremely high stiffness coefficient K—that leg did not produce enough compliant yielding.

Then, a counter-intuitive scene appeared.

The leg stepping on the pen was forcefully pushed high.

But because all six legs of the entire machine were locked together by the spindle's phase, the body did not tilt along with the force to yield to it. Instead, like a zombie under an immobilization spell, it pushed the two support points on the opposite diagonal side into a semi-suspended state.

Like a beetle tripped up by its own stiff legs, on the verge of flipping over at any moment, its posture was twisted and ugly.

The spindle motor spun helplessly in mid-air, consuming a huge amount of electric current, yet unable to push the machine forward by even a single millimeter.

Scrap Iron Old Zhou stood in front of the workbench, staring at the machine suspended in the air because the spring was too stiff, his brows knitting deeper and deeper.

"The second trap."

His voice carried exhaustion and awe.

"In the field of multi-legged robots, the stiffness of the support structure is definitely not better the harder it is."

"This spring is too stiff. When complex terrain pushes it up, its rule system doesn't know how to compromise; it doesn't yield."

Old Zhou turned his head, looked at the camera, and uttered a sentence that made the entire mechanical community look sideways.

"Now that I think about it, in Low Entropy Workshop's original video, that machine pieced together from scrap parts always looked so slow, even somewhat feeble."

"That probably wasn't because the second-hand motor he used lacked torque and couldn't go fast."

"Rather, in its system design, it had to be slow."

"It had to be slow enough, the series compliance had to be sufficient, and the locking had to be late enough to leave time for this brainless mechanical structure to transfer loads."

After this sentence came out, the usually noisy comment section fell noticeably quiet for a long time.

Because it was no longer the complaining of keyboard warriors, nor was it armchair theoretical deduction.

Instead, it was an epiphany bought with failure by a replicator whose hands were stained with engine oil, after truly touching that ceiling of mechanical logic with his own hands.

At the end of the video, Scrap Iron Old Zhou took a wrench and silently disassembled the expensive first-edition regular-phase simplified insect—which he had spent three all-nighters making—into a pile of parts.

"In the dynamic post at the beginning of the video, I said that there is no need to deify this thing."

Old Zhou lowered his head to take apart the screws without looking at the camera.

"Now I take back the second half of that sentence. The first half still holds: it is indeed not otherworldly technology fabricated and invented out of thin air by someone; its structural pedigree and lineage from predecessors are very clear."

"But right now, facing this pile of scrap metal, I must admit one thing: the truly difficult part is not making the six linkage legs move gracefully in the air."

"It is how you can ensure that when the machine is in an extremely harsh and unknown environment, if a certain leg does not bear force the way you imagined, or even deforms and collapses, the remaining five legs won't go crazy along with it."

Old Zhou raised his head, a fanatical gleam burning in his eyes.

"In the next video, I'm going to make two versions."

"One is the perfect regular-phase version with optimized springs and timing."

"The other, I'll try to probe Low Entropy Workshop's bottom line, and make a low-commensurability phase version with scrambled phases."

"I just want to see which of these two versions will go crazy and deadlock first when placed on the same plank covered in scattered rocks."

At the end of the video, there was no emotional rhetoric asking for a triple-combo like usual.

On the black screen, a line of white text slowly emerged.

["That bug from Low Entropy Workshop runs pretty deep."]

Within hours of its release, this twenty-minute video filled with mechanical assembly and harsh crashing sounds savagely pushed the name Low Entropy Workshop once again into the spotlight.

In the engineering world, one clear failure is far more convincing and lethal than ten thousand empty praises.

...

If the first wave to take the field were lone rangers fighting single-handedly, then the second wave was a fully armed regular army.

The RoboMaster team and Robot Innovation Laboratory of an established 985 engineering university in central China suddenly spoke up.

This team participated in the RoboMaster competition year-round, enjoying stable sponsorship every year with a complete division of labor in machinery, electronic control, and computer vision.

They simultaneously published an eight-thousand-word hardcore academic essay packed with charts, scatter plots, and formula derivations on the academic section of their campus BBS and a mechanical engineering column on Zhihu.

The title was as rigid as an experimental report.

《Regular Phase vs. Low-Commensurability Phase: A Minimalist Kinematic Comparative Analysis Based on Purely Electric-Driven Multi-Legged Platforms and Passive Elastic Joints》

The first paragraph of the article cut straight through all the fan-versus-hater debates on the internet with the meticulousness and arrogance unique to students from elite universities.

["Disclaimer: Our team has not attempted, nor do we feel it necessary, to replicate Low Entropy Workshop's original verification machine."]

["The reason is simple: In terms of engineering ethics and academic norms, the other party has not open-sourced the complete parameter matrix. At the same time, we do not believe that a creator showcasing engineering concepts on a public video platform has any obligation to open-source their core blueprints to the public."]

["The purpose of this article is not to judge the merits of the original machine, but merely to conduct a minimalist controlled-variable comparison: under the conditions of identical linkage topology, identical aviation aluminum materials, identical total mass distribution, identical 24V brushless DC motors, identical reduction mechanisms, and identical electronic speed controller control methods, strictly compare the failure propagation performance of regular symmetrical phase versus artificially set low-commensurability asymmetrical phase in several simplified extreme terrains."]

If Old Zhou was crossing the river by feeling the stones with intuition, then this group of 985 students was modeling the river with data.

The main body of the article detailed three sets of comparative tests.

Group One: Flat ground constant-speed test.

Without any suspense, the regular-phase version won clearly.

Through motion capture from high-speed cameras and current curve analysis on the motor control board, the regularly symmetrical hexapod robot had a steadier speed, with smaller fluctuations in body height and yaw.

Most importantly, its overall energy consumption efficiency was extremely high.

Meanwhile, that control version, whose gait was forcibly scrambled and set to a low-commensurability phase, appeared on flat ground like an extremely awkward clown.

Its body swayed slightly, and the rhythm of its footing was completely unaligned. To maintain this asymmetrical balance, the motors frequently had to output redundant reactive current.

In the conclusion section, the team captain wrote very bluntly, even carrying a hint of defense for classical mechanics.

["On flat ground and predictable regular terrain, symmetrical structures are God's masterpiece. The performance of the low-commensurability phase in such environments is completely an unnecessary waste of energy."]

After this sentence was screenshotted and posted to Tieba and Weibo, it immediately excited some people who couldn't stand Low Entropy Workshop from the very beginning and thought he was putting on an act.

["Hahaha, is that it? Have you seen the elite regular army's report?"]

["Facts speak louder than words, slapping Low Entropy Workshop in the face, slapping non-periodicity in the face."]

["I told you so. With mechanical industry developing for hundreds of years, regular symmetrical structures are definitely the optimal solution of natural selection. That blogger forcibly messing up the legs was purely to stand out and collect an IQ tax."]

But this carnival lasted less than three hours.

As readers continued reading downward and saw the second set of tests begin, the sneers in all the forums came to an abrupt halt, as if their throats had been choked.

Because they replaced the terrain board with a random obstacle board.

The height difference was extreme, the spacing had no pattern whatsoever, the slope angles were completely random, and the materials varied in hardness.

This was the true hell these students prepared for the two machines.

The GIF of the test footage showed that in the first third of the start, the regular-phase version still walked very fast relying on its powerful drive.

But when it reached the complex middle section, disaster struck.

When the second and third legs on the left stepped consecutively twice onto bumps with the same spacing in an extremely short period, a Physics phenomenon emerged.

The symmetrical structure's perfect geometric proportions turned into a fatal poison in this environment full of random disturbances.

The body began to show extremely regular synchronous swaying.

The frequency and amplitude of each sway became more and more similar to the previous one.

The body began to show rhythmic synchronous swaying.

Merely reaching the seventeenth step.

The right rear leg experienced an almost imperceptible slip and jam on a tiny slope.

Because all the legs meshed with each other within the exact same cycle, this minor slip was instantly transmitted to the gear disc.

The pawl engaged prematurely once again when it shouldn't have meshed.

The protection circuit triggered, and the system shut down.

Obvious chipping appeared on the tooth face of the right rear pawl, and the motor driver board's overcurrent protection record was maxed out.

The regular phase showed repetitive accumulation of phase errors in this random obstacle field, and was judged as a failure.

It died of its own perfect symmetry, and died of the infinite cascading amplification of errors.

On the other side, that slow and ugly low-commensurability phase version, although walking even slower...

...was at first so slow that it made people anxious, and its sprawling posture lacked any sense of security whatsoever.

However, every single body sway of its failed to form any clear physical rhythm.

When a certain leg stepped into a pit, just as a bad posture that might lead to resonance was about to rear its head, the next leg within the system—with a completely different temperament and a completely staggered cycle—would drop down extremely rigidly, forcefully breaking this harmonic with a discordant force-bearing angle.

Bumping and stumbling across the obstacle board with all sorts of counter-intuitive distortions, it dragged itself across the finish line.

Its final posture was ugly to the extreme.

But it didn't burn any boards or deadlock.

As for the third set of tests, which was also the most mysterious aperiodic obstacle board...

...these top students from the 985 university acted very cautiously, even wrapping their phrasing in a thick layer of academic disclaimers.

["Note: Limited by processing conditions and computing power, we have not, nor are we able to, use the strict Jiangs Brick algorithm to construct an infinite plane aperiodic test field. This paper does not claim the validity of this terrain board in the strict topological mathematical sense."]

["We merely generated an obstacle field with no simple repeating distance distribution within a finite window via the Monte Carlo algorithm."]

This sentence was written so seamlessly that it made the notoriously picky big shots in the mathematics section nod in slight satisfaction.

The test results were far more complex than anyone imagined, and it wasn't simply a matter of who won and who lost.

Facing this nearly perverted terrain, the perfect regular-phase version wasn't actually completely unable to get through.

As long as there was manual intervention to carefully select a golden starting footing for it, it could actually use brute motor force to crush its way through.

But it was extremely sensitive to the initial state, exhibiting characteristics of a chaotic system.

If the landing point of the first step deviated by just two millimeters, its subsequent strides would frantically amplify this same type of misalignment within a specific set of phase cycles until the system collapsed.

What about the low-commensurate version carrying the aperiodic expectation?

In the actual footage, it experienced two thrilling high-point bumps scraping its chassis, and one of them even resulted in a severe roll and rollover of nearly 45 degrees.

But the most incredible thing was that, judging from the records of the foot pressure sensors and the onboard IMU, its points of failure were discrete.

In other words, it would also make mistakes, but its errors were dispersed by that messy parameter network, and they would never be continuously amplified for more than three steps concentrated on any specific mechanical beat.

At the end of this paper-grade column, the leading second-year graduate student wrote a resounding yet deafening conclusion.

[ Summary: The low-commensurate parameter system is not a universal mechanical improvement. ]

[ On most regular terrains and flat ground in reality, its performance is often worse and consumes more power. It is a disaster for engineers; it is extremely difficult to establish an inverse kinematics model, extremely difficult to tune parameters through code, has lower machining efficiency, and the later mechanical wear and maintenance costs are immeasurable. ]

[ But— ]

[ In extreme terrains that disrupt the inherent rhythm, and in harsh scenarios subject to strong, uncontrollable physical disturbances on a single leg, this system indeed exhibits a failure characteristic distribution completely different from modern classical robotics: its physical layer can spontaneously interrupt the cascading amplification effect, stopping errors locally. ]

[ The rough and simplified experiment we conducted by no means proves that Low Entropy Workshop's grand conjecture about aperiodic parameters being a once-and-for-all anti-failure solution is the absolute truth. ]

[ But this data is enough to illustrate one thing: ]

[ His seemingly extremely crazy conjecture is worth more robotics mechanics laboratories conducting stricter replication experiments. ]

As soon as this article came out, the debate across the entire internet regarding Low Entropy Workshop suddenly underwent a qualitative leap.

Previously, the focus of everyone's infighting remained at a low level.

Was Low Entropy Workshop trying to clout-chase the recently viral mathematical problem?

Was it clickbait just to latch onto trending topics?

Was that rust-covered thing a toy picked up from a scrapyard?

But now, in the face of these rational control data, those debates purely revolving around clout-chasing and clickbait were pushed to a secondary position for the first time.

The core issue turned into cutting-edge academic discussion.

Those who previously questioned the structure compromised.

Fine, the predecessors' lineage is indeed clear, but does this aperiodic parameterized variable assignment method constitute a completely new design degree of freedom that humans have never explored before?

Those who were previously just watching the spectacle were dumbfounded.

To what extent has this matter developed now, and what variables are still missing control in the experimental matrix?

Someone majoring in materials science jumped out and shouted that this asymmetric stress would inevitably lead to local component fatigue.

It is necessary to add long-term decay data on parameter drift after the system runs for 1,000 hours and metallic material fatigue produces deformation.

An engineer working on extreme-environment robots asked whether low temperatures, dust, or loose threads would trigger a cascading collapse.

Big shots in supercomputing also inquired whether they should directly build a full-scale three-dimensional aperiodic continuous field model in simulation software to run a test.

The best communication effect a hardcore tech video can achieve in the human intellectual network is never to make everyone kneel on the ground and fanatically praise it as an inviolable oracle.

Rather, it is like a depth charge dropped into the center of a lake, making everyone capable of thinking feel a prickling sensation on their backs, and starting to spontaneously revolve around the origin it proposed to mend experiments, prove, overthrow, and expand boundaries.

...

This raging fire spreading from the mechanical circle to the academic circle finally made the traditionally aloof popular mathematics science section unable to sit still anymore either.

The first to jump out and speak up was a top-tier UP host on Bilibili who specialized in combinatorial mathematics and the history of mathematics.

His day job was a university lecturer.

He released an urgently produced blackboard-explanation video with a very straightforward title carrying strong bait-like implications:

"Is Low Entropy Workshop's so-called aperiodicity actually an insult to mathematics?"

Clicking into the video, right at the opening, this UP host wearing heavy black-rimmed glasses tapped the whiteboard behind him, his tone very restrained yet piercing straight to the soul.

"Let's state the conclusion first, everyone."

"If you are a stubborn pure math hound and understand Low Entropy Workshop's clanking hexapod machine as an aperiodic system in the strict mathematical topological sense, then you will definitely feel offended."

"Why?"

He wrote a few massive formulas on the whiteboard.

"Because six legs, what does that mean? A finite structure. No matter how many types of spring stiffness it has, they are still finite parameters."

"In mathematics, how could a finite physical system of yours be lumped together with non-repeating aperiodic tiling on an infinite plane (such as the famous Penrose tiles, or the recently viral Jiangs Brick)? Talking about infinite aperiodicity within a finite set is a substitution of concepts."

The UP host changed the subject, and the chalk in his hand drew two rows of dot matrix diagrams on the whiteboard.

In the first row, the distances between dots were completely equal and neat.

In the second row, the distances between dots were uneven, but upon closer inspection, they were definitely not the kind of randomly scattered mess without any pattern.

"The aperiodicity defined in mathematics has an extremely high threshold and is extremely strict; it requires you to exhaust infinite space."

"But in rough real-world engineering full of friction, gravity, and material tolerances, most of the time, what engineers borrow is merely a tiny tip-of-the-iceberg idea of it."

The lecturer circled that uneven dot matrix with a red pen.

"This idea is: do not let your local physical rules collapse into a simple repetition just for the sake of saving trouble."

"Do not let the entire complex system contain only a single frequency beat that is too neat and too harmonious."

"Everyone, this is the truly interesting part of this matter."

The lecturer looked at the camera, his eyes full of appreciation and yearning.

"Low Entropy Workshop's UP host—whoever divine sacred being he might be—his truly awesome part is by no means that he solved a world-class mathematical problem with a hammer and a wrench; that's unrealistic."

"He did something that only a lunatic would try: he forcefully crammed the abstract concept of aperiodicity—which has stayed on paper recently and exploded in the academic world due to the proof of Jiangs Brick—like kneading dough into a real mechanical body that has been replicated and tested by countless people."

"He pulled down that out-of-reach mathematical dark cloud in the sky and turned it into a physical bulletproof vest that, although rough, could withstand bullets. This is what you call the aesthetic of engineering."

Beneath the comment section of this video, problem-solvers from the mathematics section and engineers from the mechanical section unprecedentedly put aside their mutual professional chain of contempt, and for the first time, an extremely peaceful and in-depth dialogue appeared.

[ Those studying mathematics say they understand. So it does not strictly apply Jiangs Brick's structure? ]

[ Those in mechanics reply to you: Correct, it does not directly apply mathematical results; it is more like a dimensional-reduction reference at the underlying thinking level. It turns mathematical bricks that tile space into parameter bricks that tile the time of mechanical motion. ]

[ Then does this count as clickbait clout-chasing? ]

[ It depends on how you define clout-chasing. If he finishes posting this conceptual video and directly starts selling aperiodic-style keyboards next episode, that's shameless clout-chasing. If he dares to shoulder all the curses, continue doing it, and complete the control experiments and algorithm logic, then he has pioneered a new cross-disciplinary school of thought. ]

[ Brothers, stop arguing. The key to everything lies in Low Entropy Workshop's second video. ]

Yes, just as everyone defaulted.

Low Entropy Workshop still had not replied to anyone.

But invisibly, a hand had already put him over the fire.

Because the first episode had already thrown out that worldview-shattering conjecture.

If he did not continue proving it, he would be completely swallowed by this backlash wave he himself had stirred up.

If he continued doing it, once the experiment collapsed, the academic controversy he faced would be too large for this private account of his to bear.

...

At the same time in Beijing.

As night fell, a small graduate student group from a mobile platform research group at Tsinghua University.

Inside a WeChat group of graduate students named [Mobile Platform Anti-Damage and Extreme Environment Group], which usually mostly just synchronized group meeting notices, paper links, experimental shifts, sensor procurement progress, and test bench data, suddenly vibrated continuously a few times.

The person sending the message was a second-year Ph.D. senior student who had just published two top-conference papers and was usually arrogant.

He forwarded the link to the Zhihu column titled "Minimalist Control Analysis" by the central region 985 robotics team.

Immediately afterward, against his usual habits, he attached a long voice-to-text message.

[This video by Low Entropy Workshop, that aperiodic hexapod that was very popular on Bilibili the other day, has any of you looked at it carefully?]

[At first I thought it was folk scientists and folk philosophers doing performance art. Now, undergraduates from a brother institution have done a small-scale simplified control with controlled variables; although the experimental conditions are very crude, the result distribution curve running out of it is quite interesting.]

Three o'clock in the morning was precisely the active period of the brain for people doing scientific research.

After a few minutes, someone in the group replied immediately.

[Watched it. To be honest, the design of its hexapod body is too crude. Even if it is purely electric-driven, the transmission efficiency is also very low. In a real ruins environment, the wear on the linkage dead points must be shockingly large, and it would have to be scrapped after walking less than a few kilometers.]

Another person immediately refuted.

[Hey, do not always stare at their crude hexapod hardware. Hardware can be solved by spending money on better five-axis precision carving and titanium alloys. You need to look at their underlying asymmetric parameterized thinking.]

The third group member silently sent a screenshot, which was precisely the final conclusion marked with a red line in that 985 robotics team's article: "The mechanical physical layer can spontaneously interrupt the cascading amplification effect, stopping errors locally."

[This passive compliance and anti-damage capability achieved purely through mechanical parameter asymmetry—if it could be transplanted to the disaster-rescue quadruped robot or outer space probe our department is currently working on, how much underlying error-correction computing power could be saved? Computing power is electricity, and electricity is life in outer space.]

The group fell silent for a moment.

The depth of the discussion had clearly surpassed the scope of mere spectator gossip.

Finally, someone could not help asking a question.

[By the way, just gossiping, who on earth is the person in charge behind this account Low Entropy Workshop? Is it a retired old man from some research institute, or a new small scientific research team coming out of Harbin Institute of Technology? Has not anyone dug into it?]

[Do not know, the account is as clean as a blank sheet of paper.]

[IP location does not matter; even if he is a Martian, it does not matter. The key is that this topic of using aperiodic Physics to disrupt system resonance is worth studying seriously.]

The group fell quiet again, as if everyone was facing their phone screens, pondering the paper potential behind this conjecture.

About ten minutes passed.

Finally, an account that rarely spoke in the group popped up.

The profile picture was a somewhat blurry photo of a lunar rover ground test.

The display name was very simple.

[Xu Huaiyuan]

An old professor from the Automation Department of Tsinghua University who had spent half his life doing mobile robots and extreme-environment teleoperation systems.

In his early years, he followed space robot ground verification projects, and later opened a practical course on intelligent mobile robots for undergraduates across the university.

People like this rarely participated in idle chatter in student groups on usual days.

He only looked at two things: whether the problem was real, and whether a rig could be set up to test it.

The group was quiet for a few seconds.

Then, Xu Huaiyuan sent a sentence.

[Save it first.]

After a while, he added another sentence.

[Tomorrow after the group meeting, take 20,000 yuan of mobile funding to build a minimum control rig. One regular phase unit, one low-commensurate phase unit. Do not make the terrain complicated yet: one regular board, one random board, and one non-repeating spacing board.]

[Do not stack sensors; record the three items—foot pressure, body posture, and spindle current—clearly first.]

[Look at the data.]

No one in the group posted jokes anymore.

...

At the same time, Jiang Lin was quietly sitting in front of his computer, listing the experimental outline for Low Entropy Workshop's second video.