Dispute: Angela Stefanski - NES / FAMICOM / DISK - The Legend of Zelda - NTSC - Fastest Completion [1st Quest] - Player: Rodrigo Lopes - Score: 31:37.0

[Jace Hall]

Here I fixed that for you, sir.

You post is not helpful. I am trying to actually get this matter sorted and this sort of post does not assist and derails the thread. If you continue with this I will ban you from further participation on this thread.

Thank you.

[datagod]

"Lips flapping" is a description of an action, not a description of a person or people. It is a colloquialism for the phrase "just talk." You may want to reevaluate your disappointment if you think I was describing anyone.

I have reevaluated my disappointment as suggested. Turn out, I am still disappointed. Funny thing that. Feelings get hurt no matter how logical the explanation. Humans over-react I suppose. I guess I am more human than machine. I can't keep up with the logic being thrown at me.

I know I keep saying I am done with this dispute....but I am a sucker for punishment.

I hope that you re-evaluate your use of colloquialisms. Until then perhaps I'll start using some of my own (from NFLD):

"Oh me nerves! Ye got me droves!! I knows yer not stun, what odds? "

[Desidious]

You post is not helpful. I am trying to actually get this matter sorted and this sort of post does not assist and derails the thread. If you continue with this I will ban you from further participation on this thread.

Thank you.

Absolutely man. I'm done looking at this circus act anyways. My memes have more weight and value than your smoke and mirrors and answering almost everything with another question responses that YOU give. You lost all respect with that f****** unprofessional lip flapping comment. Do what you must.

[Jace Hall]

I have reevaluated my disappointment as suggested. Turn out, I am still disappointed. Funny thing that. Feelings get hurt no matter how logical the explanation. Humans over-react I suppose. I guess I am more human than machine. I can't keep up with the logic being thrown at me.

I know I keep saying I am done with this dispute....but I am a sucker for punishment.

I hope that you re-evaluate your use of colloquialisms. Until then perhaps I'll start using some of my own (from NFLD):

"Oh me nerves! Ye got me droves!! I knows yer not stun, what odds? "

I see.

For what it is worth, it is easier for dispute participants to be more relaxed or emotional in their determinations about dispute claim matters because they don't necessarily have to be concerned with future legal processes in its regard. Twin Galaxies must be as thorough and as diligent as possible in all these matters. It can not be emotional about it or have an agenda other than getting as close to indisputable absolute fact as much as possible - or at least be able to reasonably demonstrate that it made every effort to do so.

[Jace Hall]

For anyone who felt offended for any reason whatsoever at my use of the phrase "lip flapping" instead of using "just talking" I completely apologize.

At no time did I attempt to describe any individual by that term. I only attempted to describe an action and that is the only context that I used the phrase within.

The entire point that the comment revolved around was that "just talking" about a subject does not necessarily constitute objective and definitive evidence regarding the validity of the subject itself unless the subject is the fact that the talking about the subject took place.

[Lauren Tyler]

For anyone who felt offended for any reason whatsoever at my use of the phrase "lip flapping" instead of using "just talking" I completely apologize.

At no time did I attempt to describe any individual by that term. I only attempted to describe an action and that is the only context that I used the phrase within.

The entire point that the comment revolved around was that "just talking" about a subject does not necessarily constitute objective and definitive evidence regarding the validity of the subject itself unless the subject is the fact that the talking about the subject took place.

I understand what you mean and where you're coming from. Your apology is accepted.

[mrturk]

@mrturk,

Your post was helpful, thank you.

One thing I would like to specifically understand in the finest detail is the notion regarding the enemy positions.



I want to be clear in my understanding of the assertion that it has been definitively proven impossible for the enemy configuration to be as depicted in Rodrigo's run without splicing.

Is this assertion an absolute fact?

So far, what I have been able to discern is the statement that says, "The only other time we've seen this configuration so far is in a spliced run."

Now just because it has only been seen in a "spliced run," that does not definitively prove that there is absolutely no other possibility of the configuration occurrence.

So what I am trying to understand is how is it absolutely known that there is no way possible for the configuration to take place? Have all possible permutations and outcomes been vetted? If so, who did that and where can I see that information?

Also, regarding the historic "spliced run" - is there any documentation about how the splice was determined and definitively proven? Did the person admit to it being spliced? If it is being used as an absolute reference, I want to make sure that the determination on that run was in fact absolute as well.

If it is 100% definitively proven that there is absolutely no programmatic way for the enemy configuration to be as Rodrigo's was without splicing or other unfair manipulation, then the case can easily be closed - so any clarity you can provide on this in terms of hard evidence would be helpful.

1) @starcrytas has done a pretty good job answering the bulk of these questions but let me see if I can add anything else. According to the "optimal" route I defined in post #207 then yes, only TSA's known spliced run puts the enemies in the precise locations as seen in the disputed submission in question. Here is one example from the best current run from the internet:

.
This is a run by lackattack24 in 27:40. This run is much faster than the dispute in question since it uses glitches that are not allowed under TG settings but that does not matter in this analysis. What does matter is what happens at 23:05. At this time you will notice Link face south (down), instead of north (up) like in TSA's run. Lackattack then blows the whistle once to take him to 1 and then follows the "optimal" path as described and then subsequently bombs into the leftmost rock formation to enter 9. When he enters the third room of level 9 you will notice he gets the "common" spawn seen in all other speedruns and noted in the bottom photo of post #28 by @GibGirl . You can choose others to look at but this just results in redundancy after a while.

2). In regards to the possibility of seeing these spawns show up in a non-spliced performance, no run has been found where this has been observed yet. However, as @starcrytas and I showed it would be possible to produce these spawns in 4 "non-optimal" cases as illustrated in posts #36, 37, 52, and finally 115. In each of these cases the time loss can vary from 1:00 to 2:00. An excellent time loss analysis was provided by @starcrytas in post #39 and what the player would have had to sacrifice in order to set the spawns in the locations seen in the disputed submission by following one of these non-optimal paths. It is also important to note that no run has been observed that utilizes any of these 4 non-optimal cases on the internet or on any other site.

3). @starcrytas provided a link which documents the process that SDA used to determine that TSA's run was spliced in post #214. Also, I provide evidence of additional splices including a brief explanation of how the warp whistle works in the top half of post #64. From what I can recall back in 2011 TSA admitted to recording his run in segments then used a "copy and paste" method to only show the segments that he wanted. In other words he would record parts of overworld segments and then record the dungeon parts as part of a different segment and then patch them together until he got the run that he wanted. This resulted in improper enemy spawns, improper enemy drops, and improper warp whistle use throughout his run. He sort of danced around the question of splicing and said "he did not intend" for this to happen but it was clearly obvious and proven that he did. Evidence of this sort of "cut and paste" splicing can be seen starting at 28:04 in TSA's run that I referenced to @Jace Hall in post #207. Unfortunately, I am not really qualified to speak on what kind of software may have been used to achieve this kind of "cut and paste" splice since I have no experience in doing such a thing. The good news is that Mike "TSA" Daimini has been located and an effort will be made to contact him to speak on this matter. It would be great if he does respond or maybe another technical expert could explain the technical aspects of how a splice could be done in such a run. However, this is sort of a touchy matter and it may be difficult to get someone known to have been caught doing this to actually speak on how they did it.

[mrturk]

Here's some more information I was meaning to post in the last post: http://tasvideos.org/3232M.html This link represents the current TAS (tool assisted speedrun) for this game performed by Lord Tom. To reiterate from what I mentioned in an earlier post, a TAS is a computer simulation using an emulator in which a set of precise inputs are used in order to determine the best theoretical solution to a problem in a video game. In most cases these "problems" involve completing a game in the fastest possible way using parameters established for that category (any%, 100%, glitched etc.). These inputs are devised and executed by humans to create the TAS to serve as both a theoretical model as well as to provide entertainment value to the casual viewer. The ability to use save states, frame advance, slowdown features, RAM addresses etc. allow these projects to be completed in a very optimal manner. The TAS community and speedrun/RTA community are mostly considered to be separate entities although ideas and collaboration are frequently done between the 2 groups.

By clinking on the link above you will taken to the most current TAS of the game from 2016 by Lord Tom in 22:17.53. The time is much faster than the disputed time in question mainly due to the fact that glitches are allowed and many items such as additional heart containers, the magical sword, etc that humans would need to complete the game in a timely manner are skipped entirely. Additionally, for reference, if you click on the tab "history of this entry" you can view all of the other iterations of the TAS for the Legend of Zelda (first quest) going all the way back to 2004. For relevancy to this dispute it is then possible to watch the final 5 minutes of each of these iterations and compare the general appearance of the enemies in the final dungeon. In ALL cases dating back to 2004 the enemy spawns in the third room of the final dungeon match the "common" spawn seen in all current RTA runs such as lackattack24's as seen in the preceding post. It is important to note that even a route change in 2016 used in the current iteration of the TAS in which Link walks from dungeon 6 instead of using the optimal route to 9 defined in post #107 does NOT alter the "common" enemy spawn. If you watch all of these iterations the spawns observed in TSA's run and the disputed submission do NOT occur.

[fcoughlin]

@Jace Hall

Above in the thread, you referenced wanting to understand spawns further in Z1, and to understand how things work. I have just local recorded a video discussing the spawns in Z1, and investigating how potentially the spawn count could have been manipulated to get the spawns purported by the accused. It went in depth, so it's 30 minutes; my apologies for rambling for so long.

tl;dr: The following points come up:
- Spawns are based on a counter that's kept throughout the game (ranging from 0-8), and based on the direction you enter the screen. Nothing else affects that (not how long you're on a screen, not movement on a screen or enemy movement, etc.).
- There is exactly 1 value out of those 9 possible values that generates the enemy pattern present in Level 9 in the accused's "WR" video.
- In order to generate that pattern in a single segment play, the only possible way is to bomb open level 9, leave that screen on the side you came in from, and re-enter prior to entering level 9; at least 1 enemy must be left alive on both of the screens for this to happen. That costs you 10 seconds (not to mention possible health loss as well).
- If you were to take a death after bombing open level 9 (and conveniently head to a fairy fountain etc.), then that spawn occurs naturally entering the screen the first time.
- No matter which of the spawn sets you get entering level 9, the spawn counters converge upon exiting the first stair; unlike other cases where a spawn pattern could change a whole level, there is no "long term" benefit that could be argued from the spawn change in level 9. It affects a single screen.

So I can't definitely say with 1000% certainty that the run wasn't single segment (since the runner could have decided to throw away 10-15 seconds minimum there traversing the screen), but I can say, without a shadow of a doubt, that the screen with the entrance to level 9 had to be visited at least twice during the run that ended with the camera recorded level 9 segment. It is 100% impossible for any runner to get that spawn pattern on the first time on the level 9 overworld screen.

[starcrytas]

I found something else last night that might pertain in particular to the handling of this submission, but not necessarily what's on the actual video but that I think still warrants discussion in this thread. As has been discussed in this thread, the traditional method of verification during the time period of this submission was a lone referee. However, there were some titles in particular that involved more than a single referee verification. Games which were considered "popular" and/or "highly competitive" in nature fell into this category. I have found evidence that the original Legend of Zelda fell under this category: http://tasvideos.org/forum/viewtopic...r=asc&start=25. This is an old speedrun topic on tasvideos.org from 2005-2006. If you look at the post about half way down the page by "Sleepz" (real name Richard Ureta), the current #2 on the leaderboard on this site for Legend of Zelda, he says "Triple verification will be done any day now" on 6/25/05. If you scroll down the page to the very bottom on 7/6/2005 Sleepz says "Regarding the time, Mr. Kelley has verified it to be 32:56, and now just waiting on the other 2 judges." I assume the "judge" he is referring to is Kelly Flewin as well as 2 other unnamed referees. The 32:56 is referring to a time that Ureta sent to TG around this time period for verification so there is no doubt when he uses the term "judges" that he is referring to TG referees.

So how does this apply to this submission? If you click the bottom link in post #39 by Matthew Felix in the case of this disputed submission it appears that there was only 1 referee and it was @RTM. It begs the question: Why was one player subjected to a triple verification while the player in question only subjected to a single verification on the same track for the same title? If @RTM was the player in question's referee of choice then fine, but why was this submission not sent to at least 2 other different referees which was the standard for this game at the time? Going back to post #18 in this thread on 7/29/19 by @RTM he says " I sent my cache of tapes that I had to Shawn Cram (via Brian Kuh) back in Nov. 07." I think this would prove the tape never went to a second or a third referee for verification. So, this leads to the following question: If the submission did not meet the minimum requirements for verification established for this title then how could it have ever been entered into the database at all? In other words is this submission even valid at all considering the way it was handled according to the standards that were in place at the time?

According to this link, it took over 9 months for Richard Ureta's speedrun to be verified by two referees, who were Kelly Flewin and RTM in this case.
https://www.twingalaxies.com/showthr...429#post604429

Then according to this link where Ureta's run was announced, Rodrigo claimed a 31:36 time
https://www.twingalaxies.com/showthr...his-OWN!/page2

A month later, Rodrigo's run was verified.
It looks like this speedrun did not go through the required multi referee verification unless something changed, but I couldn't find anything that changed the ruling about multiple referee verification.

So one user had to go through more than one referee to get a run validated, while another only needed one referee.