Talk:Tool-assisted speedrun

Wiki Education Foundation-supported course assignment

This article was the subject of a Wiki Education Foundation-supported course assignment, between 15 June 2021 and 3 August 2021. Further details are available on the course page. Student editor(s): MasterYoshi5.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 11:30, 17 January 2022 (UTC)

Tool-assistance outline

o History
o Why and how
----o Goals
----o Emulation
o Controversy

Some links that should be linked to or otherwise used

• Slashdot | 'Perfect' Zelda NES Speed Record Beaten
• Interview of Phil for September 2005 issue of GEE
• Interview of Bisqwit for September 2005 issue of GEE
• Interview of Bisqwit for GameCritics
• MTV News - Gamers Divided Over Freakish Feats Achieved With Tool-Assisted Speed Runs
• Slashdot | Nonsense with Google's AdSense?
  • Huh? What does this link have to do with an encyclopediac article of speedruns? --Bisqwit 11:58, 22 May 2006 (UTC)
    • Absolutely nothing. Striking it out now. Nifboy 17:54, 22 May 2006 (UTC)
      • I was collecting links of NESVideos in the news here. Doesn't have anything to do with speedrunning, but when I was collecting them I didn't have that criteria in mind. It was listed here for completeness. —Michiel Sikma, 06:52, 23 May 2006 (UTC)

Some images that may be used

• Image:Bombmanmap.png – Route planning sheet for the Bomb Man stage in Joel Yliluoma's Rockman tool-assisted speedrun
• Image:Bombmanmap2.png – A more definitive version of the above image; redundant routes have been scrapped and the most feasible one has been noted down in greater detail

Some resources for citation

Citation is important for this article. See Wikipedia:References to find out how to use the <ref> tags.

• NESVideos / Glossary
• NESVideos / Guidelines
• NESVideos / Why And How

Sandbox

<ref name="ToolsAssistedSpeedrun">Doom tool-assisted speedrunning is sometimes referred to as "tools-assisted speedrunning", with the plural of "tool". This was the name of the site about Doom tool-assisted speedrunning by Esko Koskimaa, Peo Sjoblom and Yonatan Donner. A news post after the creation of the site, however, read "Indeed, I was wrong and the site should be called 'Tool-Assisted Speedruns' rather than 'Tools-Assisted Speedruns'. I'm not going to redo the logo though."</ref>

Checking In

Hi, I got your "invitation" to work on this page. I don't know what kind of time I hae this weekend but I had to point one thing that struck me right away by the opening: As such, the makers of these speedruns are not competing in terms of playing skill due to its complete absense. - this seems to imply that the filmmakers have no playing skill at all. This is obviously not the case! I'm going to try and reword it. --Do Not Talk About Feitclub (contributions) 16:01, 8 April 2006 (UTC)

Yeah, it was difficult to word that sentence. I meant to imply that something such as "skill" isn't even relevant, but it's wrong to make it seem like the TAS authors have no skill. Quite the opposite if you look at who started Doom tool-assisted speedrunning. I'm very glad that you're helping out, by the way, since you know a lot about this and you're also probably the most active Wikipedian around here. Maybe you could also shed some light on how active the Japanese speedrun community is. It seems to me that they're pretty enthusiastic about it, since a lot of visitors of Bisqwit's site are Japanese, and they've produced some amazing speedruns (Takeshi's MGS2 run!) —Michiel Sikma, 19:02, 8 April 2006 (UTC) PS: I always get "humanly" wrong.

To say that tool-assisted runners don't need any skill does no justice. They need to be above-average players of the game they're running to be even able to pull off some of the most difficult tricks that are needed in TASes. Wikipedian06 09:00, 11 April 2006 (UTC)

You're right, and I never meant to imply this. Tool-assisted speedrunners actually usually are good players. The starters of the Doom tool-assisted speedrun community actually were seasoned veterans from the larger speedrun projects, for example. That's something that should be added in... —Michiel Sikma, 14:35, 11 April 2006 (UTC)

Re-recording

I made a short article about re-recording. It might be worth checking it out to correct any mistakes, even though I doubt there will be any, seen as how it's such a small article. —Michiel Sikma, 09:55, 9 April 2006 (UTC)

I think this should have been made a section of the tas article. Bob A 18:56, 11 April 2006 (UTC)

Maybe. I think that it's best to briefly explain re-recording here and leave the article around for more detailed explanation. You forgot to add the {{mergefrom}} tag, by the way. —Michiel Sikma, 06:02, 12 April 2006 (UTC)

Yeah, I've just thought it over, and I think it's best to explain it in this article in addition to re-recording having its own article. My rationale is that re-recording is or can be linked to from other places as well where it would be inappropriate to write an inline explanation. In this article, it's necessary to mention it (under a "techniques" paragraph, most likely) but it would not be very useful to link all pages to Tool-assisted speedrun#Re-recording.

In short, I think it's best to keep the article, but also put useful information in this article about re-recording. —Michiel Sikma, 18:49, 12 April 2006 (UTC)

I don't get it. Why wouldn't it be useful to have it redirect to Tool-assisted speedrun#Re-recording? Bob A 01:22, 14 April 2006 (UTC)

Because those links are confusing, most of the time. Browsers don't jump to the anchor until the page is fully loaded. You'll be shown a page that's about something completely different, and when the page finishes loading, you'll scroll to a small part of that page, emerged in unrelated context, that explains what you've been trying to find. I personally find that very ambiguous, and I therefore like to mirror things (not verbatim, but depending on where the information is located). —Michiel Sikma, 10:04, 15 April 2006 (UTC)

Endurance

Wikipedian06, please stop adding in "endurance" in the top paragraph. There's good reason why it's not in; obviously, endurance is a factor when speedrunning which is eliminated with the use of tools, but it's untrue that it's important enough to mention. Afterall, there has never been made a tool-assisted speedrun that's longer than about 3 and a half hours, but there have been gigantic single-segment speedruns (such as the Ocarina of Time run by TSA). It's untrue that one of the main points of tools for tool-assisted speedrunning is to overcome human endurance. —Michiel Sikma, 14:34, 11 April 2006 (UTC)

NPOV/Weasel

There seems to be a lot of negativity towards TASes in this article.

• There is skill in TASes. Some tricks are just very insanely difficult to pull off. You don't see the average n00b trying to make a TAS, do you? Most TASers, especially those who're speedrunning FPSes, are above-average, if not the best players, in those games. In fact, the reason the 16:17 TAS record in Doom II wasn't beaten for so long was that after Marijo Sedlic retired, nearly no one had the skill to perform the insanely difficult rocket jumps on some of the levels. It was asked on the Doomworld message boards.

• What the hell is "official world record keeping?" There is no "official" recordkeeper as far as I know. The closest to an official WR authority on video games is Twin Galaxies. The point of having TASes go in a separate category is so that there can be two world records for every game -- the fastest unassisted time and the fastest tool-assisted time.

• "today's tool-assisted record holders have less fame and association with their run." Not true at all. Most speedrunners don't become "famous," period. One of the most well-known speedrunners is Mike "TSA" Damiani, but that's because he holds impressive unassisted records in many LoZ games. If one TASer broke as many records as TSA did, I'm sure he'd be well-known as well. Marijo Sedlic and Andrey Budko are two DOOM TASers who are well-known in their community.

• "Some purists believe this is the only real way to do a speedrun, even going so far as to describe tool-assisted speedruns as "fake" or "cheating"." This is a narrow-minded, conservative viewpoint that is fading away. Many are beginning to recognize TASes simply as another category that deserves as much respect as traditional, unassisted runs.

• The positive aspects of TASes need to be brought out. People appreciate TASes because they allow one to see the theoretical human limit in a game, without having to waste time watching errors in gameplay. Unassisted runners, when appropriate, often use TASes as standards for comparison (of course, this depends on the game and how wide the differences are between the unassisted and TAS routes).

• It should be mentioned that the two communities often help and cooperate with each other. For example, recently the TASer for Mario 64 discovered a new trick via emulator that allowed one to get a star without using a cannon. It was later discovered that the trick was reproducible on a console, and now the unassisted speedrunners have a new trick up their sleeves too.

69.181.79.72 01:01, 18 June 2006 (UTC)

As a dude who has three TAS's published, I feel compelled to reply:

• With proper tools (particularly frame advance), any amount of skill can be reduced to knowledge and patience; The knowledge to know how to do the tricks, and the patience to pull them off. That the most skillful players are generally the ones making the TASs is a result of having the knowledge to do so.
• I'm tempted to say there is no "official world" anything, given that it's nearly impossible to get the entire world to agree on anything. Merely a bad word choice, I'd replace "official" with some other adjective. "Recognized world records", perhaps.
• "The positive aspects of TASes need to be brought out." To me, this says, "I want to replace the article's current bias with my own."
• Statements about fame and "what (xyz group) believe" are better off cited than removed, but either is better than uncited.

I haven't been watching the article too carefully, but I do sort of see what you're talking about, though. I'll tag a few choice phrases with {{fact}} Nifboy 04:49, 18 June 2006 (UTC)

I'm sorry. I haven't worked on this article much. I agree with some of the things you say. I'm a big fan of speedruns, both tool-assisted and unassisted, and would like this article to become a featured article after I'm done with speedrun. I will need the help of other people, though... I hope there can be another main contributor on this article. —Michiel Sikma (Kijken maar niet aanraken) 14:45, 19 June 2006 (UTC)

PS: I'll be busy with speedrun for quite some time. I do suggest that if anybody is interested in this article at all, then please go ahead and edit! I've been the only actual active contributor to speedrun articles for a while now, and it feels lonely. —Michiel Sikma (Kijken maar niet aanraken) 06:35, 27 June 2006 (UTC)

• The NPOV / weasel tags have been tainting this page for a long time. Do they still apply? Should they be removed? --Bisqwit 08:41, 2 April 2007 (UTC)

Glossary

Do we really need a glossary like that? Wouldn't linking to articles describing the subject be better? Amaurea 17:58, 2 July 2006 (UTC)

I intend to write more elaborately about some of the items mentioned in the glossary later, similar to Speedrun, § Common techniques. What I think is that it's probably very useful to keep a quick glossary in the article as well for easy reference; I want to do this for the Speedrun article at some point, too. Whatever the case may be, I personally don't think that it's a good thing to just link to other articles in these cases; some of the techniques don't really need their own article nor should they be explained in a section such as § Common techniques. —Michiel Sikma (Kijken maar niet aanraken) 18:40, 2 July 2006 (UTC)

Goal of TAS

I do not like this sentance: "The goal of tool-assisted speedrunning is to create entertaining movies." This is no more or less true for unassisted runs as for assisted ones. The real goal is usually to complete the game as fast as possible, sometimes with a few restrictions. This happenes to be quite entertaining, but a run with lots of interesting antics that make it 3 times as slow as what is possible would not be much of a speed run. I'm replacing that paragraph with the definition from the method section. Amaurea 11:03, 3 August 2006 (UTC)

Actually, the preceeding section seems sufficient to me, so I just removed the troublesome paragraph. Amaurea 11:10, 3 August 2006 (UTC)

There is a lot of disagreement about this. That gives us something to write about. :) —msikma <user_talk:msikma> 15:17, 3 August 2006 (UTC)

I have removed a similar section from the "seperation from unassisted runs"-section:

While speed is desired in these runs, the focus is entertainment. For example, there is a famous tool-assisted speedrun of Super Mario Bros. 3 by "Morimoto" in which the game is finished in approximately 11 minutes — the fun is in the breathtaking speed at which the game is being played, according to its author, rather than his presumed game playing skills.[1] The video has since been "obsoleted" by a faster version,[2] but this video is still the best-known example of its kind due to how often it was posted on forums on the Internet. (See the Super Mario series paragraph in the Speedrun article.)

That entertainment is more important than speed is no more the case for assisted runs than unassisted ones. I have never seen a slower run obsolete a faster one in either category. It is true that entertainment is desired, but the focus, in both cases, is speed. That section made TASing seem like a less competitive and less serious activity than unassisted speedrunning. Amaurea 11:24, 17 August 2006 (UTC)

But there's still a sense of truth in that paragraph. We should rewrite it, since it should not go unnoticed that there are indeed a lot of TAS makers that think it's all about entertainment rather than speed. Also, please note that just because obsoletion movies have always been faster than the old versions shouldn't imply that the focus of these movies is definitely not entertainment. Me, I like to think that it is mostly about entertainment, but it just so happens that speed is the primary (not only) entertaining aspect of TAS videos. —msikma <user_talk:msikma> 15:59, 17 August 2006 (UTC)

I personally see this "entertainment factor" as a POV guideline imposed by Bisqwit and his site. Speedruns are about speed, and tool-assisted ones are no different. Otherwise, why is it that whenever a run with a faster time is submitted, it obsoletes the slower one?

Also, note that runners are ADVISED (but not required) to entertain viewers during autoscrolling sections. Many highly-regarded speedrunners do not go out of their way to entertain viewers at all. It's really a personal choice.

While it's true that some speedrunners waste a few frames over the course of a long run to entertain viewers, such practices are under great controversy at the site. Look at the "speed vs. entertainment" debates on the forums. Wikipedian06 18:56, 20 August 2006 (UTC)

But regardless, it is true that a lot (some?) speedrunners do these small tricks to try and entertain viewers a bit more, even if it costs a few frames. I don't think that tool-assisted speedruns are all about speed. I think that they're all about entertainment, but it just so happens that speed is the absolute best entertainment factor. Having said that, they're still speedruns, like you say, and it's only natural that faster speedruns obsolete older ones. However, it doesn't really matter what you and I think, to be honest! All that matters is that there are separate opinions about what a tool-assisted speedrun is really about; this is something we can write about. But since the biggest tool-assisted speedrun site (Bisqwit's) still states that the runs are for entertainment purposes (among others), we still must write the article stating that entertainment is said to be the purpose, with some people disputing this. —msikma <user_talk:msikma> 20:36, 20 August 2006 (UTC)

Although Bisqwit's site is the largest TAS-related one on the 'Net, I don't think his words and guidelines have to be treated as gospel. Nonetheless, some runners who do sacrifice frames for the sake of entertainment have justified their actions using Bisqwit's very subjective guidelines. But yes, I do agree with you---we should present both sides of the debate. Wikipedian06 05:55, 21 August 2006 (UTC)

We don't need to treat it as gospel, but we still have to adhere to the fact that we cannot make our own research. We cannot claim that Bisqwit says something that is biased if we are the only ones to think that it is so. Unless another external (and relatively important) party says so, we cannot really report on it. —msikma <user_talk:msikma> 06:04, 21 August 2006 (UTC)

Am I the only one who thinks that the obvious solution to this issue is to make a note that the goals of TASes can vary considerably within the guideline of frame-precision? <Upthorn> 17:07, 1 November 2006 (UTC)

Example image in the introduction

While using a program to automatically search a short optimal path (ie a robot) is cool, it's by no means a common occurrence in tool-assisted speedrunning. There are probably less than 10 runs (from the well over 300) at tasvideos.org where robots have been used in any way. The problem with presenting the robot example image in the introduction is that it can easily give the wrong impression that such programs are used in all or most runs and that it's a very common technique. This may give a rather wrong impression of what tool-assisted speedrunning is. The example image is well worth preserving in the article, but it may be a good idea to move it towards the end of it. In the introduction some other, more representative image could be used. The snapshot of an emulator program (perhaps showing some TAS features) could be a good idea. Wopr 12:53, 29 October 2006 (UTC)

I agree. Having it at the top makes it seem ubiquitous, which it isn't. Maybe we should put it further down after better explanation of what the phenomenon is all about, while mentioning that this isn't very often done. I'm not going to do it for a while, though... —msikma <user_talk:msikma> 14:59, 29 October 2006 (UTC)

New SMB Record

For example, the fastest Super Mario Bros. TAS currently stands at 4'59.6", while the fastest unassisted run stands at 5'00"

Is this according to Twin Galaxies Rules or SDA's? The new speedrun uses the vine glitch, so I don't think it's valid according to TG. --Ouzo 15:59, 16 April 2007 (UTC)

It should be noted that Twin Galaxies is not the standard for speedrun times in the speedrun article. All times there are from Speed Demos Archive. This is mainly due to the speedrun community being most active at that site, rather than at Twin Galaxies, and because records from Twin Galaxies are not verifiable by a third party. —msikma (user, talk) 22:41, 1 May 2007 (UTC)

I compared and it saw a video but the TAS is earlier when measuring in the same place and attempting to let out in 60fps.

Together, both results by 60fps are TAS after using individually and attempting to measure and the made public video is 58 in 4 minutes and 58 seconds. Because it is 11 in 5 minutes and 3 seconds at the video of the just in 5 minutes. TAS is earlier.--Naohiro19(Talk Page/Contributions/Do you send mail for me?) 11:24, 1 May 2007 (UTC)

New SMB1 time and other notes

The TAS for SMB currently stands at 04:58.18. I changed it, but I'm not sure if this changes the "compared to Speed-Demos Archive is 0:47:00" part of that sentence.

I would recommend changing the same section's notes about "a non-TAS could be faster than a TAS," at least on Bisqwit's tasvideos.org site. Perhaps back during the site's inception, but the community is large enough now that this isn't feasable. Before a TAS for any game could be published, this would be considered. This was a problem when a TAS for the game Jaws for the NES was beaten by the world record and this now is a fundamental guideline for publication.

From the rules for submission on tasvideos.org:

If your tool-assisted movie is slower than the non-tool-assisted world record of the game you play, your movie will almost certainly be rejected. This only matters if the goals are directly comparable between the TAS and unassisted record, though.

Before a movie is published, it is viewed by many members of the community and discussed. Even if a suspected faster time is considered, the movie isn't published, and a TAS that looks slower than frame-advancement, let alone human reflexes, wouldn't make it through the forum. If a decent looking movie for the game can't be created by a member, the site goes without a movie for that game.

It would be near impossible for a real-time speedrun to trump a TAS, at least with a final goal of fastest time.

TAS-san

A small paragraph about TAS-san was recently removed, because the editor could not find the term on a google search. This is probably because he searched in english. TASさん yields 500k+ results, and so is probably notable. The more recently invented full name, however, gave < 10 hits, and I had never heard of it before. I will therefore put the paragraph back, but with the full name removed for now. Amaurea (talk) 06:38, 20 September 2008 (UTC)

Clarify/correct statement in "Separation of Unassisted runs"

A trick in A Link to the Past allowing for walking through walls has allowed for an extremely short 3'45" TAS, but because the trick is impossible to reproduce in real time using a standard controller, the fastest unassisted run is over an hour long.

There seem to be a couple of problems with what that sentence may imply to someone unfamiliar with what falls under the scope of being a TAS.

There is also a factual problem with the segment; according to this video, the standard controller variant of the glitch can be done in a much shorter time than "over an hour": http://www.youtube.com/watch?v=MGinhCZ2rO8

Additionally, with a non-standard controller, the same technique as used in the tool-assisted run can be executed in real-time, albeit with less precision overall when compared to the final time of TAS. At least, it can be in an emulator. But perhaps it can also be on the console. The particulars of verifying this through the correct WP procedures and sorting out what can, cannot, and should be in this part of the article are just beyond what I felt was my place to do, editing-wise.

76.5.89.160 (talk) 16:28, 23 February 2009 (UTC)

Is this statement accurate?

"Tool-assisted runs are timed by input, i.e. from game power-on to the last input necessary such that the ending scene and/or game credits can not be prevented. Any introductory cutscenes, game-loading screens, and trailing dialogues after the last boss battle (if input is necessary to scroll through the text) are included in the final times."

If I recall correctly, there is indeed several runs that break this 'rule', such as Gimmik! which has the final boss be killed with nearly no input, yet I am certain that if human control took over AFTER the runs input is ended, the viewer can potentially break the run. Flygon (talk) 03:31, 16 October 2009 (UTC)

Sometimes it is possible to "end input early" such that with no further input the final boss is defeated, and the ending sequence plays. Trailing dialogues generally make this impossible, though sometimes exceptions are made for very long ending sequences that have forced button presses only near the end. 50.21.206.23 (talk) 00:49, 22 March 2018 (UTC)

Serious error in introduction

The introduction says that a tool-assisted speedrun can do tricks that are impossible or very difficult. That's not true. A tool assisted speedrun lets you do very hard tricks but won't let you do impossible ones or it's not called a TAS but rather uses a gameshark like device. For instance, you can't release 4 bombchus at a time in a TAS of Majora's Mask. Blackbombchu (talk) 14:57, 17 June 2013 (UTC)

It means that some of the things done in a TAS are impossible for a human to do because a human can't press the buttons quickly and accurately enough. It doesn't mean that one can do things that the game doesn't allow. Amaurea (talk) 08:12, 21 April 2014 (UTC)

BizHawk

While I recognize that the BizHawk section was reverted because the person who posted it is a known sock puppet master, that does not invalidate the content of the section. That particular emulator is a notable achievement, and should probably be mentioned alongside the other emulators. Someone please consider adding that emulator to the list of emulators. Invariel (talk) 20:03, 20 July 2015 (UTC)

"timeattack" vs. "time attack"

“You are misunderstanding; there is a difference between the time attack mode featured in some video games and the synonym of "TAS" used years ago when it all started: "timeattack".”

You’re missing the part where that paragraph does not strictly refer to TASing. Try reading it, you’ll see “this term can also be seen in numerous games that have a lowest completion time mode”. I stand by my edit.--ALAKTORN (talk) 23:36, 19 November 2015 (UTC)

It does refer to TASing. You have to take into account that it's an item in the glossary section of an article about tool-assisted speedruns. The part you are quoting is only there to give some context as to why "timeattack" is not used nowadays by the English community to refer to TASes. Zeupar (talk) 21:52, 29 December 2015 (UTC)

Either remove that phrase, reword it to specify that games use the different “time attack” term, or go back to school.--83.211.186.59 (talk) 21:56, 29 December 2015 (UTC)

Last input?

I am unsure of the statement, times are timed to the last input. Could be relevant to several kinds of games possibly, but I am thinking of racing games in particular. If one is racing toward the final finish line, and is far enough in front not to be overtaken, one could technically stop all input on the last straight before the race is actually finished, stop the timing clock, then allow the car to cruise over the finish line under it's own momentum (so long as one doesn't need input to scroll through text or something after the race). This would mean the time of the "input" itself would be slightly shorter as having stopped before the very end of the race, but the time until actual completion would be slightly longer, as the last straight would not be at full speed.

Any game with speed and momentum like say sonic one could point the character in the right direction at high speed, stop input and let the character speed toward the end under their own momentum.

Or any game where things happen without constant direct input like Lemmings, or Age of Empires. One could quickly give out all sufficiently needed commands to the units, stop all input and the clock and let the on screen units like horsemen or lemmings carry out their orders after all input has stopped and let them finish on their own and lead to the game becoming completed a period of time after all input has stopped to game completion. As far as I know this never happens in speedruns, or does it??.

Something as simple as killing a boss with a bomb in a bomberman game could be a difference of a second or two, does the clock stop when the last bomb is laid and input stops, or when the last bomb explodes and the boss is killed?

As far as I always thought time is until game completion, but I don't really know for sure, I only have an interest in watching speedruns, I'm not involved in making them.

Whether I'm right or wrong, the article could clarify this - also is this relevant or not to the regular speedrun parent article?

 Carlwev  16:38, 1 April 2016 (UTC)

The time is always the number of frames in the input file. Usually it's more entertaining to make the end happen sooner, even though the input file is longer, but not always. For example, the Monopoly 4 player TAS ends input long before the ending of the game happens, and yet it's quite entertaining. Or finding a difficult to use safe spot and ending input and letting the final boss time out can be quite entertaining. — Preceding unsigned comment added by 50.21.206.23 (talk) 00:53, 22 March 2018 (UTC)

roblox speedrun for obbys List item — Preceding unsigned comment added by 188.236.194.162 (talk) 15:57, 13 December 2020 (UTC)

TAS/RTA (real-time attack)

I went on the Wiki page of Tool-assisted speedrun precisely to learn the meaning of "RTA", and noticed the initials only appears near the end, unexplained ("Drozdowsky holds the RTA record", etc.). However "real-time" is used a few times.

I eventually found this Reddit thread: https://www.reddit.com/r/speedrun/comments/41dkk9/what_is_rta_timing/ and found the answers informative as they give the origin "real-time attack" but also give some variations on the meaning (including cinematics as opposed to in-game timing, actual play as opposed to TAS). I've also read somewhere else that when the machine's power can affect the game speed, it is important to distinguish real-time count and frames count (the number of frames may even depend on machine power depending on how the game is programmed, but I don't know how they count in this case).

Since "RTA" is used in the article in some places, it may be good to introduce the term at some point before.

(I'm not often editing pages or discussing on them so I don't have an account and cannot monitor this page, so I may not reply quickly. If you think you have a good fix for this, don't wait for my reply and go ahead - or continue the discussion with more active contributors instead) — Preceding unsigned comment added by 2A01:CB14:411:D300:B951:B141:F1FA:14A8 (talk) 23:35, 30 December 2020 (UTC)

Update: it looks like this discussion could be merged with topic "Is this statement accurate?" started by Flygon, as indicating the various ways to measure a run would also remove ambiguity.

I also noticed that https://en.wikipedia.org/wiki/Speedrun mentions RTA means real-time attack, so maybe the TAS article was extracted at some point from the Speedrun article, but the explanation of RTA was lost in the process. — Preceding unsigned comment added by 2A01:CB14:411:D300:B951:B141:F1FA:14A8 (talk) 23:43, 30 December 2020 (UTC)