This Week in Speedrunning: Minecraft endurance, AGDQ anticipation, and old games refusing to sit still

November 21, 2026

SI-generated

This week in speedrunning had a little bit of everything: a documentary-sized history lesson, a future-facing GDQ announcement, fresh discoveries in old games, and the familiar joy of somebody finally forcing a personal best over a line that had been staring them down for months. The common thread was not one single blockbuster record, but the way speedrunning keeps moving on several tracks at once.

At the top of the conversation, Summoning Salt’s history of Minecraft world records turned a notoriously punishing category into a story about endurance, resets, and the strange economics of good RNG. Meanwhile, the AGDQ 2027 schedule gave the community a preview of the next big shared stage, with Ocarina of Time, RuneScape Dragonwilds, KOTOR 2, and even a twelve-player Big Walk run among the talking points. Underneath those headline moments, runners were still doing what they do best: shaving frames, testing walls, and making the impossible look like a matter of routine.

There was also a welcome reminder that “speedrun news” does not only mean a world record. A blindfolded first, a sub-1:40 Resident Evil 4 personal best, a controller-hardware debate, and a thoughtful question about MMORPG categories all showed the scene from different angles. Some stories are about the stopwatch; others are about the people and tools that make the stopwatch worth chasing.

Here are the biggest stories from the week.

Minecraft’s Most Miserable Marathon Gets the Salt Treatment

Summoning Salt’s new history of Minecraft world records landed as the week’s biggest discussion magnet, and for good reason. Minecraft any% is one of those categories that can look deceptively simple from the outside: generate a world, gather what you need, reach the finish, and do it faster than everyone else. The reality, as the video makes clear, is a brutal negotiation with probability. Runners spend enormous stretches of time searching for a seed that is not merely good, but playable, and then gamble hours of execution on the hope that every downstream requirement lines up.

The community reaction focused on that mismatch between spectacle and process. One of the highest-rated comments joked that the category is defined by the kind of bad behavior that makes a record-holder disappear from the narrative entirely, while another pointed out how much of the job is resetting rather than running. That is a useful distinction. The finished run is a clean, dramatic artifact; the route to it is often a warehouse full of discarded attempts.

That grind also helps explain why some runners moved toward tournaments instead of raw any% hunting. Competition formats cannot remove RNG, but they can make it feel more legible: there is an opponent, a defined window, and a reason to keep playing even when the perfect seed refuses to appear. The history is therefore not just about who held the record. It is about which formats make the pursuit sustainable.

As always with a Summoning Salt video, the real achievement is turning a spreadsheet of incremental improvements into a story with stakes. The numbers matter, but so do the personalities, the grudges, the skipped names, and the question of why anyone would volunteer for this particular flavor of suffering. Minecraft speedrunning remains a monument to persistence—and perhaps to runners’ unusual ability to mistake repetition for recreation.

AGDQ 2027 Opens the Door to a Very Busy Weekend

The AGDQ 2027 schedule is out, and the early reaction suggests that the event has once again found a nice balance between established favorites and material that feels slightly unexpected. Ocarina of Time is represented by ZFG, RuneScape Dragonwilds gets a prominent look-in, and KOTOR 2 is scheduled for the kind of slot that will have a lot of viewers quietly rearranging their morning.

One of the most charming details in the discussion was the excitement around a twelve-player Big Walk run. Twelve runners on one screen is exactly the kind of idea that sounds orderly in a schedule announcement and becomes beautiful chaos once the timer starts. It is a reminder of what GDQ does especially well: taking a game or format that might live in a small corner of the community and giving it a stage large enough for the premise itself to become part of the show.

The schedule also offers a useful snapshot of the current speedrunning ecosystem. There is room for deep technical mastery, broad audience appeal, comedy, novelty, and games that are still developing their competitive identity. A GDQ lineup is never a complete map of the scene, but it is a good weather report. This one says the forecast includes classic Nintendo precision, sprawling RPG knowledge, online-game curiosity, and at least one event where the number of runners may be the visual gag.

And because speedrunning audiences are incapable of looking at a schedule without immediately imagining their own ideal viewing plan, the announcement has already become a kind of communal pre-event. People are picking must-watch runs, comparing time slots, and preparing to discover a game they did not know they needed to care about. That anticipation is part of the event’s magic long before the first reset.

Majora’s Mask Finds Another Human-First-Cycle Breakthrough

EnNopp112’s 1:11:56 in Majora’s Mask any% NMG was posted with the understated confidence of a record update, but the run represents something more interesting than a number getting smaller. The new time shows the impact of recent human-in-first-cycle technology, the kind of development that can sit quietly in practice rooms before suddenly becoming visible on a leaderboard.

Majora’s Mask has always rewarded runners who can combine route knowledge with an almost unreasonable tolerance for timing pressure. New techniques do not simply add a shortcut; they change the shape of the attempt. A strategy that saves seconds may also create a new failure point, alter the acceptable risk profile, or force the runner to rethink how the opening minutes are practiced. That is why the best breakthroughs often look obvious only after someone has made them work.

The post itself drew a jokey “this run sucks” reaction alongside a more informative observation that the gains from the new tech are finally arriving. That contrast is practically a speedrunning genre: the top comment supplies the heckle, while the second comment explains why the result matters. Beneath the banter is a clear sense that the category still has room to move.

A 1:11 is not the end of the road here. It feels more like a signpost. Once a technique becomes reliable enough to appear in a record run, the next phase is refinement: cleaner setups, better recovery, and the gradual disappearance of the “new” from new tech. The leaderboard may soon make this week’s impressive time look like another step in a much deeper descent.

Ocarina of Time’s Invisible Walls Are Having a Bad Week

An exploratory Ocarina of Time post offered a different kind of excitement: not a finished route, but a glitch that might become relevant somewhere nobody has yet bothered to look. By using bomb explosions to push through certain invisible barriers, the runner demonstrated a way to interact with the game’s boundaries that could matter beyond the specific examples shown.

The immediate caveat is important. The discovery does not replace the established route into Shadow Temple, where other movement options already do the job more efficiently. In speedrunning, “possible” and “useful” are separate categories, and the gap between them is often measured in months of testing. A clip can be real without being route-changing.

Still, this is exactly the sort of post that keeps mature games alive. Ocarina of Time has been examined from nearly every angle, yet runners continue to find interactions that were overlooked because they were too awkward, too localized, or simply irrelevant to the dominant route. The discussion included a reminder that some barrier breaks are already known, but also genuine interest in whether this particular behavior works in other locations.

The follow-up details were almost as valuable as the original clip. Notes about reproducing the setup in the graveyard, using an actor jump, and moving along the cliff near Shadow Temple turn a curiosity into something the community can investigate. That collaborative handoff—from “look what happened” to “here is how to test it”—is where glitch discovery becomes speedrun research.

Blindfolded Child Dungeons Is Exactly as Wild as It Sounds

CrystalSaver claimed the world’s first blindfolded child-dungeons glitchless run in Ocarina of Time, and even the title feels like it was generated by a speedrunning category-name machine that has been left running overnight. It is also a wonderfully concentrated example of how the scene keeps inventing new ways to make familiar games difficult.

The accomplishment sits at the intersection of memorization, audio awareness, controller confidence, and route discipline. Removing sight does not merely make individual rooms harder; it changes how information is gathered. Every movement has to be anchored to timing, sound, remembered geometry, or a practiced sequence of inputs. “Glitchless” adds another constraint, ensuring the run is not simply a demonstration of finding an accidental escape hatch.

The top comment mostly reacted to the title itself, observing that “World’s First Blindfolded Child Dungeons” is quite an opening phrase. That joke works because the run is so specific that it sounds fictional until you remember that speedrunners routinely turn one sentence into a full ruleset. The oddness is not a distraction from the achievement; it is part of the achievement’s charm.

There is a serious point underneath the novelty, too. Challenge runs like this expose how much of a classic route lives in muscle memory and mental mapping. They also create new communities around constraints that would never appear in a standard leaderboard category. Sometimes the best way to appreciate a game’s movement system is to take away the most obvious source of information and see what remains.

The Controller Arms Race Is Still Happening in N64 Land

A question about replacement N64 analogue sticks produced one of the week’s most practical discussions. For anyone who has watched high-level Super Mario 64, the controller can look less like a peripheral and more like a carefully tuned piece of laboratory equipment. The answers reflected that reality: OEM sticks, GameCube controllers, second controllers for specialized inputs, and the increasingly popular steel-bowl designs all have their advocates.

The reason this matters is that analogue control is not a neutral input method in these games. Small differences in tension, range, friction, and physical shape can affect everything from movement consistency to the confidence of a Bowser throw. A runner may be practicing the same route as everyone else while effectively using a different instrument, with its own strengths and failure modes.

The thread also highlighted how difficult it is to speak about “the” preferred setup when the community is actually a collection of individual compromises. One runner’s reliable solution is another runner’s unfamiliar geometry. Some players want the feel of an original stick; others prefer modern alternatives or a second controller dedicated to a narrow task. There is no single best answer, only the answer that lets a particular runner stop thinking about the hardware and start thinking about the line.

That is the quietly fascinating part of speedrun equipment culture. The goal is not necessarily to buy the most expensive setup. It is to remove uncertainty from a high-pressure action that may be repeated thousands of times. The controller arms race is less about luxury than about turning a known source of wobble into one fewer thing capable of stealing a run.

Small Games, Big Milestones, and the Joy of Sub-1:40

Not every important speedrun story needs a world-record tag. In Resident Evil 4 Remake, a runner’s 1:38:40 on PS5 delivered the kind of personal breakthrough that explains why people keep returning to the timer in the first place. The goal was sub-1:40, and the celebration was immediate—even though the run included two self-inflicted explosions and a Krauser fight that refused to cooperate.

That combination of mess and success is more representative of everyday speedrunning than a perfectly scripted record. The route can be practiced, the splits can be understood, and the execution can still wobble in several places before the final time becomes real. A personal best is not a purity test. It is evidence that the runner kept enough of the attempt together to reach a new boundary.

The post’s enthusiasm was contagious because the milestone was concrete. “Sub-1:40” is easy to understand, but it is not easy to earn. It gives the runner a new target—perhaps 1:39 next—and gives everyone watching a glimpse of the ladder that exists inside a single game. Every category has these thresholds, and every one feels enormous to the person who is trying to cross it.

Elsewhere, a new sub-hour Act III record in Star Wars: Galactic Racer showed the same pattern at a higher level. The 58:53.16 run depended on major strategies, unusually favorable event choices, and short races that can vary dramatically in length. Whether or not the game makes the next GDQ schedule, the category is already demonstrating the classic speedrunning cycle: a new route appears, the ceiling moves, and suddenly a barrier that looked permanent is just another split to attack.

The Week’s Quietest Story Might Be the Most Important One

A discussion about MMORPG speedruns opened a door onto a genre that often feels missing from the wider speedrunning conversation. The question was simple: why are there not more prominent MMO categories? The answers pointed to Old School RuneScape, Final Fantasy XIV content, Destiny raids, and the historical spectacle of World of Warcraft leveling races and expansion launches.

MMOs are difficult to fit into the familiar speedrun template. They are persistent, frequently updated, dependent on other players, and often designed around long-term progression rather than a clean finish line. A category may involve server states, account preparation, population density, or rules about what counts as a fresh start. That makes verification and standardization harder before the timer even begins.

But those obstacles are also what make the genre interesting. MMO speedrunning might look less like a single runner clipping through a campaign and more like an expedition: route planning, resource markets, group coordination, and a race against a changing game world. The absence of a universally dominant format does not mean there is no speedrunning there. It may mean the community has not settled on the most satisfying way to measure it.

With RuneScape Dragonwilds appearing on the GDQ 2027 schedule, the timing feels especially good. A highly visible run can give viewers a vocabulary for understanding why persistent games are compelling under speedrun rules. It can also encourage runners in adjacent communities to formalize categories that have previously lived as one-off challenges, races, or personal projects.

That may be the broader lesson from this week. The scene is not only optimizing established routes; it is continually deciding what deserves to be timed, compared, explained, and celebrated.

A Week of Records, Research, and Ridiculous Specificity

Taken together, the week’s stories make a strong case for speedrunning as both sport and folklore. Minecraft offered the long, punishing history of a category shaped by RNG. Majora’s Mask and Ocarina of Time showed old games continuing to yield technical progress. AGDQ offered the communal stage where all of that specialized knowledge becomes entertainment for a much larger audience.

The smaller stories mattered just as much. A blindfolded category, a controller discussion, a personal best, and an MMO rules conversation each revealed a different reason people stay involved. Some chase records. Some chase a first. Some chase a cleaner setup, a better explanation, or the satisfaction of proving that a strange idea works at all.

That variety is the scene’s greatest strength. The leaderboard is the visible structure, but the culture underneath it is built from experiments, jokes, arguments, hardware tinkering, historical research, and the occasional decision to spend an evening trying to defeat an invisible wall with a bomb. The timer gives these activities a shared language without forcing them to become the same kind of achievement.

Next week, some of these discoveries will have become routes, some records will have fallen, and at least one runner will announce a category whose title sounds made up until the video starts. For now, the week leaves us with a scene that is busy, inventive, and still very good at finding new ways to make old games surprising. See you next week.

Source: reddit.com

Comments are closed.