31–40 of 124 posts
On the video I see pausing of the timer on 0:36 (video time, not timer time). What's going on there?
On the video I see pausing of the timer on 0:36 (video time, not timer time). What's going on there?
As you'll notice by the end of the video the timer doesn't actually display 4:55.913, even though that's his record time - this is because the entire run has been re-timed (more accurately) after the record was set.
>When Kosmic says he's tied with the TAS run at 4-1, that means his inputs are matched up exactly with the program's and he can't physically do any better. It means he can't physically get a better time, but it does not mean his inputs are matched up exactly with the TAS. SMB1 level transitions are based around the "frame rule", which means you don't advance to the next level until the number of elapsed frames from t…
Earlier quoted context omitted.
To an extent I think that streaming has solved this problem, compared to the days of people just submitting videos that they recorded by themselves. Streamers might fake a single run to get a record, but it would be a huge amount of work to fake all the progression they had to go through to get there.
I agree, but furthermore after you get to a certain level of skill, you pretty much have a free pass to cheat if you like. To wit, I don't think many people suspected Billy Mitchell of cheating until there was undeniable proof of that fact.
Not so much. Billy has been suspected of cheating for ages. His live play is completely at odds with his WR runs. As of the early/mid 2000s he just wasn't very good at Donkey Kong when compared with his contemporaries.
There is something amazing about artforms when the performers and spectators have an appreciation for the subtlest details. Mario speedruns are an oddball artform surely, but right now it has enough of a community to appreciate those details. For example, from the article I gather to "clip the wall to warp into the first level of world four" is a well-known and discussed precision move in the genre. It reminds me a l…
Naw, clipping the wall is taking advantage of a well known intentionally created bug where you walk through a wall to get to an area to skip levels. Every kid playing the game back in the 80s knew about it.
super impressive stuff. I'm generally unfamiliar with speedrunning, but I am familiar with the billy mitchell controversies, so I'm surprised that people accept the legitimacy of streamed runs. wouldn't it be trivial-enough to fake? tangental: my personal favorite mario-related feat is sethbling injecting flappy bird into super mario world manually: https://www.youtube.com/watch?v=hB6eY73sLV0
To an extent I think that streaming has solved this problem, compared to the days of people just submitting videos that they recorded by themselves. Streamers might fake a single run to get a record, but it would be a huge amount of work to fake all the progression they had to go through to get there.
Pretty flimsy system (and far from "solved"), and it lets cheaters steal glory for the months or years it takes to catch them which threatens the legitimacy of the sport past a critical mass.
Moderators/judges are on the wrong side of a trapdoor function because of the amount of analysis it may take to verify a single video, and they aren't professionals in the field of video analysis.
I don't see how "all progression" is relevant here since you don't need to be a streamer to speedrun much less to submit a well-doctored video with a time that beats other people.
If anyone is interested in speed runs I really recommend this YouTube channel https://www.youtube.com/channel/UCtUbO6rBht0daVIOGML3c8w Thy're a series of videos on how world records have been lowered over time on various games. It's fascinating watching particularity the one on Super Mario 64 https://www.youtube.com/watch?v=l7ePi38LnrA&t=16s
>When Kosmic says he's tied with the TAS run at 4-1, that means his inputs are matched up exactly with the program's and he can't physically do any better. It means he can't physically get a better time, but it does not mean his inputs are matched up exactly with the TAS. SMB1 level transitions are based around the "frame rule", which means you don't advance to the next level until the number of elapsed frames from t…
What proves that he's not switching to a TAS from times to times?
>When Kosmic says he's tied with the TAS run at 4-1, that means his inputs are matched up exactly with the program's and he can't physically do any better. It means he can't physically get a better time, but it does not mean his inputs are matched up exactly with the TAS. SMB1 level transitions are based around the "frame rule", which means you don't advance to the next level until the number of elapsed frames from t…
What proves that he's not switching to a TAS from times to times?
Splicing isn't always cheating, it can be used for other reasons.
I can't prove that he isn't, strictly speaking, but Kosmic is a highly respected runner and has the skills to make this record.
Earlier quoted context omitted.
To an extent I think that streaming has solved this problem, compared to the days of people just submitting videos that they recorded by themselves. Streamers might fake a single run to get a record, but it would be a huge amount of work to fake all the progression they had to go through to get there.
This just isn't true. Recently there was a Super Meat Boy cheater who was discovered due to analysis of a persistent animation across the game that wasn't lining up due to video cuts. Pretty flimsy system (and far from "solved"), and it lets cheaters steal glory for the months or years it takes to catch them which threatens the legitimacy of the sport past a critical mass. Moderators/judges are on the wrong side of a…