بقلم Online Games VIP Editorial
·
Niftski improved his Super Mario Bros. any-percent record during a Halloween stream, leaving the run just two frames from the reported TAS target.
Niftski has set another Super Mario Bros. any-percent speedrun world record, according to IGN, improving on a record he already held during a Halloween livestream. The run was completed while he was dressed and performing as Mario, turning an attempt at one of the most technically demanding categories into an unusually theatrical broadcast.
IGN reports that the improvement leaves Niftski two frames, or 33 milliseconds, away from the theoretical perfect run associated with a tool-assisted speedrun, commonly called a TAS. The report also says the lost frames came early in the attempt, during fast accelerations in the first and second rooms.
The achievement is significant because the source describes Niftski as the holder of Super Mario Bros. records in the warpless, glitchless, and any-percent categories. However, the supplied IGN report contains conflicting time figures for the new run, so the exact record time should be treated cautiously until the run footage or an updated leaderboard entry resolves the discrepancy.
IGN reports that Niftski beat his own Super Mario Bros. any-percent record on a Halloween livestream. The attempt is available through Niftski's YouTube channel, according to IGN, although the supplied source does not provide an official Nintendo video or a separate verification statement from a leaderboard administrator.
IGN's account says the run was 0.33 seconds faster than the previous record. The report presents one time as 4:54.299 in the social post describing the achievement. Later in the same article, however, IGN lists the previous record as 4:54:332 and the new time as 4:54:229. Those two passages do not agree with each other, either on the new time or on the size of the improvement.
That inconsistency does not change the central confirmed development: Niftski improved a Super Mario Bros. any-percent record that he previously held. It does mean OnlineGamesVIP cannot responsibly present one of those figures as the definitive official time based only on the supplied source record.
IGN also reports Niftski's explanation for the remaining gap. He made small mistakes in the opening acceleration sequences, losing one frame in each of the first two rooms. In the context of a run measured in minutes, those are tiny errors. In a category already operating near the known theoretical limit, they are the entire difference between this performance and the reported target.
Super Mario Bros. speedrunning depends on consistent movement, precise acceleration, controlled jumps, and decisions that preserve the fastest possible route through the game. IGN's report focuses on the opening rooms because early movement errors can remain visible in the final result even when the rest of the run is executed successfully.
In this case, the two-frame gap is not simply a rounding difference in the displayed timer. IGN describes each frame as roughly 16.7 milliseconds and identifies the total gap as 33 milliseconds. Analysis: at that scale, the challenge is less about finding a broad new route and more about reproducing a narrow sequence of inputs with exceptional consistency.
IGN characterizes the target as a “perfect run” in the sense of matching a theoretical tool-assisted result. A TAS is produced with tools that can help plan and execute inputs with frame-level precision, while a human speedrunner must perform the sequence in real time. The source does not establish that the TAS target is formally unbeatable, nor does it say that the category has been permanently solved. It presents the target as the current theoretical pinnacle described in the report.
IGN further reports that, if no substantially faster strategies are discovered, only two more world records may remain before the human-run time reaches that target. That is an interpretation tied to the source's current assumptions, not a confirmed limit on future Super Mario Bros. research. New techniques, revised route knowledge, or a correction to the timing record could change the picture.
The record concerns Super Mario Bros. “any percent,” a category IGN describes as a race to the end by any available means, including exploits and glitches. That distinguishes it from the warpless and glitchless records that IGN also attributes to Niftski. The report does not describe a new Nintendo release, a hardware update, or a change to the game's rules.
The supplied source is about the original Super Mario Bros. speedrun and its timing categories, not a new Nintendo product announcement. It does not confirm which console, emulator, cartridge setup, capture method, or verification process was used for the livestream. It also does not say that Nintendo has endorsed, reviewed, or formally certified the record.
That platform context matters for readers comparing speedrun times. A record table can distinguish categories and rules that may look similar from the outside, while timing conventions and verification requirements can affect how runs are compared. IGN identifies the game and category, but the supplied article does not provide enough technical detail to independently evaluate the setup behind this particular attempt.
Analysis: the most useful comparison for players is therefore not the headline number alone. The key context is that Niftski is comparing a human-performed run with a highly optimized theoretical target, in a category that permits exploits and glitches. A casual playthrough, a warpless attempt, and an any-percent record attempt are measuring different goals even when they use the same game.
IGN reports that Niftski completed the record while dressed as Mario and spoke in character during the stream. The performance was not presented as a separate gameplay rule or a requirement of the category. It was a livestream choice layered on top of an officially relevant speedrun attempt.
According to IGN, Niftski said the playful performance helped him feel less nervous and made the run feel casual. That is his explanation for the effect of the costume and character work, not evidence that the approach improves speedrun performance generally.
IGN's report also describes the broadcast as a contrast between highly precise gameplay and an intentionally imperfect Mario impression. Analysis: that contrast helps explain why the stream attracted attention beyond the leaderboard itself. The run demonstrates that a serious record attempt can be presented as entertainment without changing the underlying category, although the costume alone does not validate the time.
For viewers, the practical lesson is that presentation and performance should be separated. The costume and livestream format explain how the attempt was shared. The two-frame discussion, category rules, footage, and eventual verification are what determine its competitive importance.
The catalog includes a directly relevant version of the game discussed in this report: Classic NES Series: Super Mario Bros. It provides a way to experience the core platforming challenge behind the record story, though the supplied catalog entry does not establish that it uses the same setup, timing rules, or verification conditions as Niftski's run.
Players interested in trying the category should first decide what kind of challenge they want. An any-percent attempt focuses on reaching the end as quickly as possible and may use the exploits allowed by that category. A personal best does not need to approach the world-record pace to be meaningful, especially when the leading run is separated from its reported theoretical target by only a handful of frames.
The immediate question is the official time. IGN's supplied article gives conflicting figures, including 4:54.299 and 4:54:229, while also describing a 0.33-second improvement. Until the livestream footage, a record database, or a correction from the publisher clarifies the discrepancy, readers should treat the exact number as unconfirmed in this report.
The next question is whether Niftski or another runner can reproduce the two opening frames. IGN says Niftski believes one frame was lost in each of the first and second rooms. If that account is accurate, a future attempt could focus on those early accelerations without requiring a completely different route.
It is also not yet confirmed whether the reported TAS target represents the final practical limit for human runners. IGN presents the possibility that only two records remain if no major new strategies appear, but the source does not establish that no faster techniques exist. The category's future will depend on route research, consistency, timing standards, and verification as much as on the spectacle of another livestream.
For now, the confirmed story is a new Niftski-held Super Mario Bros. any-percent record, achieved during a Mario-themed Halloween stream and described by IGN as two frames from the current theoretical target. The precise time and the long-term meaning of the apparent two-frame gap require further confirmation.