User:Whipmywillows/SMB1 Reference Index

* wrong warp:

* wrong warp:

* Wyrmwood bounty:

* Wyrmwood bounty:

== Inaccuracies ==

* Gach 2022

** Claim: {{tq|"In April 2004, Scott Kessler cut down the known records by another minute. It was the last time more than a few seconds would be saved in a single run."}}

** Assessment: '''False'''. There are several times attested between Xox's 6:11 and Scott Kessler's 5:13, including Scott Kessler's 5:17 which itself saved time off of Cam Allen's 5:20. Kessler's 5:13 (5:10 modern timing) did not save a minute on the current record. In this case, I think this is just lazy journalism. The reporter found a popular YouTube video to turn into an article, took the runs selected for that video as comprehensive when they were never supposed to be and just ran with it. This source should probably be avoided.

*Bell 2026

**Claim: {{tq|"averge11 snatched the crown from, er, averge11. He'd previously managed to finish the game in 4 minutes 54.515 seconds, beating previous champ Niftski's (USA) time of 4 minutes 54.565 seconds."}}

**Assessment: '''False'''. averge took the record from Niftski, who took the record from Niftski, who took the record from averge. I think this is a case where Guinness here was looking at their own records and for whatever reason Niftski never bothered going back to Guinness with his 4:54.482 or 4:54.448. But the only one here who actually care's about Guinness's records is Guinness. This one should be fine, other sources clear up the confusion well enough. It is however definitely [[WP:PRIMARY]].

*Henges 2022

**Claim: {{tq|"As a result, top runners such as Niftski have been shaving frames and milliseconds off their speedruns and bringing their time closer to the 2011 TAS run."}}

**Assessment: '''Misleading'''. The source uncritically use's HappyLee's 4:54.032 as a benchmark for human potential without considering the fact that it breaks real time rules. This is cleared up well enough by other sources.

*Burnes 2012

**Claim: {{tq|"Nintendo's best score was 5 minute and 20 seconds."}}

**Assessment: '''Unknown'''. This is a quote pulled almost directly from [https://web.archive.org/web/20080208071720/http://www.twingalaxies.com/index.aspx?c=12&id=686 Twin Galaxies'] announcement of the run. TG annoucements also have some other fairly dubious things to say about SMB speedrunning. I have no idea what's being referred to here, or when Nintendo would have polled their players about this. I suppose it's possible Nintendo Power might have done something like this at one point but I'm not gonna dig for it. I do know Cam Allen is attested to have gotten a 5:20 by TG which other announcements mention. Worth keeping for now.

*Orland 2021

**Claim: {{tq|"In the early 2000s, many Super Mario Bros. speedrunners struggled to beat the game any faster than in 5:08. That's primarily because they were sticking to Twin Galaxies' rules, which still prohibit the use of unintended glitches to complete the game faster."}}

**Assessment: '''Misleading'''. In the broad strokes correct, but it reasonably should have mentioned that TG and modern runs use different timing methods or coverted that 5:08 to a 5:05. A fairly minor mistake.

**Claim: {{tq|"Wall clip: Moving Mario with sub-pixel precision allows a player to get Mario’s foot partially (and briefly) stuck in a wall. Jumping repeatedly from that position will get Mario fully stuck, at which point he can clip through and run to the other side. This is useful in saving time getting to the World 1-2 warp zone and setting up a "wrong warp" in World 4-2."}}

**Assessment: '''Partially False'''. This is a pretty poor explanation of how wall clips work. Bismuth gives a fairly decent explanation [https://www.youtube.com/watch?v=U7RzoIEoSMY&t=423s here]. The important part is you do not need to do a bunch of extra jumps to get yourself into the wall. One jump is enough as seen with the pipe clip in 1-2. Extra jumps can help you get fully into the wall if you don't manage to clip in far enough with one jump, as was often done for the clip in 4-2, but it's not an essential part of the trick. It looks like Orland was basing most of these glitch explanations on [https://www.youtube.com/watch?v=_FQJEzJ_cQw&t=405s this Bismuth video] which was made before 1-2 pipe clip was implemented and has a poorer explanation. This doesn't discredit the article but it does mean it shouldn't be used as the basis for a wall clip section.

*Brown 2016

**Claim: {{tq|"One of sockfolder's greatest contributions to the community was the discovery of the flagpole glitch, which allows players to bypass the fireworks at the end of the level."}}

**Assessment: '''False'''. Sockfolder didn't discover flagpole glitch, that was discovered by TASers in 2007. He discovered a setup that allowed it to be possible in real time. This just seems like an example of poor research from the author.

**Claim: {{tq|"He gained legend status in 2011, at the age of 21, when he became the first speedrunner to complete the game in under 5 minutes."}}

**Assessment: '''False'''. AndrewG didn't get the sub 5 in 2011, he got the sub 5 on December 24, 2010 as seen on [https://web.archive.org/web/20120227075158/http://speeddemosarchive.com/Mario1.html the original SDA page]. He uploaded the video to YouTube in 2011, which I believe is where this claim comes from. Given this sources poor research habits, it might be best to find ways to replace it in the main article.

*Irwin 2015

**Claim: {{tq|"It only takes a glance at AndrewG’s Youtube page to understand both the depths of a Mario game and the severe patience needed to unlock those secrets. Nine years ago, on March 23 2006, Gardikis uploaded a video recorded on VHS of him beating Super Mario Bros. in five minutes, five seconds, fractions off the record set by Scott Kessler. The following year, he posted this runthrough of five minutes flat. But it’s not until January 2011, four years later, when he finally beats his record and finishes SMB in 4:59.4, the first to ever break 5:00."}}

**Assessment: '''Misleading'''. The dates this article uses are weird and really off from other sources. AndrewG completed his 5:09 (5:05 modern timing) in May 2005 and Twin Galaxies verified in August 2005, here's [https://www.speedrun.com/smb1/runs/mk32gqxz Andrew himself saying so]. So why does it say March 23, 2006? Because that's when he uploaded the run to YouTube. The source itself says this if you're paying attention, but they don't exactly make the distinction clear either. This is also why they give a January 2011 date for the sub 5 even though the accepted date elsewhere is December 24, 2010. He didn't upload the run immediately. Also, Scott Kessler didn't have the record at the time, Trevor Seguin did but I guess it did break Kessler's time so either way. This source can be used but exact language is necessary. In many cases I think that exact language stops it from being very useful.

*Roeder 2016

**Claim: {{tq|"Andrewg was the first to set a world record under five minutes, in 2011"}}

**Assessment: '''False'''. Again, andrewg completed the run in 2010, he uploaded the video to YouTube in 2011. Maybe this does need to be cited seeing as sourcing is so insistent on it. They're wrong but y'know [[WP:WIKIPEDIAISWRONG]]. Otherwise this source is fine.

*Hernandez 2016

**Claim: {{tq|"Super Mario Bros. players have determined that the fastest theoretical time for the game is 4:54.03"}}

**Assessment: '''Misleading'''. Another source uncritically using the 4:54.032 TAS without considering left+right. To be fair, in 2016 no one was really discussing tying the TAS so it's a reasonable mistake.

*Macy 2016

**Claim: {{tq|"The 4-minute, 57.260-second world record...beats the old one by just 0.007 seconds."}}

**Assessment: '''Embarrassingly False'''. The previous record was 4:57.427, so in this one IGN fails at basic math.

**Claim: {{tq|"In fact, one YouTube commenter points out darbian's run is only 15 frames off from a tool-assisted run."}}

**Assessment: ... Do not use this source. It's not good for very much anyway.

*

== References ==

== References ==