Speedrunning thread 3.6: Oh my god ball block

The Something Awful Forums > Discussion > Video Games > Speedrunning thread 3.6: Oh my god ball block

mycophobia: May 7, 2008

Grassy Knowles posted:

If you�re playing in AC it�s totally fine I just think you should be in a different leaderboard than those who can brave the environment and console heat

hurr

# ? Sep 8, 2023 18:21

Adbot: ADBOT LOVES YOU

# ? Jun 6, 2024 03:28

Hobojim: Oct 31, 2011

Grassy Knowles posted:

If you’re playing in AC it’s totally fine I just think you should be in a different leaderboard than those who can brave the environment and console heat

Sure, if it's a "How long can I stay in this room" leaderboard that makes sense to me!

# ? Sep 8, 2023 18:23

Grassy Knowles: Apr 4, 2003; "The original Terminator was a gritty fucking AMAZING piece of sci-fi. Gritty fucking rock-hard MURDER!"

Hobojim posted:

Sure, if it's a "How long can I stay in this room" leaderboard that makes sense to me!

That�s what a no breaks leaderboard is though

# ? Sep 8, 2023 18:24

mycophobia: May 7, 2008

anyways regardless of how dumb that post is, autofire in jp arcade communities has been an accepted thing for decades and at most they separate leaderboards, most of the time it's just indicated whether or not the player used it in a single leaderboard. and given that jp arcade game players are the best and coolest gamers it would behoove others to follow their example. thank you

# ? Sep 8, 2023 18:26

pixaal: Jan 8, 2004; All ice cream is now for all beings, no matter how many legs.

Just do one leader board with tags and let people compete at whatever sub category they want.

# ? Sep 8, 2023 18:38

Jose Valasquez: Apr 8, 2005

Hobojim posted:

I'm all for turbo in games with lots of mashing but I don't know why separating leaderboards for allowing turbo vs not isn't the obviously simple solution.

It's going to depend heavily on the game, but in my experience it's because one or the other becomes the de facto category so splitting is ultimately pointless, either allow it or don't.

Splitting it is the strategy Chrono Trigger took and there are 3 turbo runs total across all SNES categories despite the fact the only benefit turbo is mashing text during cutscenes, it doesn't actually decrease the skill involved in the run at all it would just save people from RSI. But people don't want to be obligated to buy an asciipad for $15 to compete (ignoring the fact that a physical console and CRT, which are both considerably more expensive, are essentially required for top times, there are only 4 emu runs in the top 20) so it stays a de facto "no turbo" game

# ? Sep 8, 2023 18:57

Hobojim: Oct 31, 2011

Jose Valasquez posted:

It's going to depend heavily on the game, but in my experience it's because one or the other becomes the de facto category so splitting is ultimately pointless, either allow it or don't.

Splitting it is the strategy Chrono Trigger took and there are 3 turbo runs total across all SNES categories despite the fact the only benefit turbo is mashing text during cutscenes, it doesn't actually decrease the skill involved in the run at all it would just save people from RSI. But people don't want to be obligated to buy an asciipad for $15 to compete (ignoring the fact that a physical console and CRT, which are both considerably more expensive, are essentially required for top times, there are only 4 emu runs in the top 20) so it stays a de facto "no turbo" game

Thank you for not being a flippant rear end in a top hat about it

Definitely agree it's game by game, I guess I've just never really understood why one category being considered "real" is that big of a deal. Sounds like Chrono trigger doesn't benefit meaningfully from turbo but what is meaningful varies from person to person and scene to scene. Does it seem like a net negative that Chrono trigger split leaderboards to you? It sounds like a non-issue.

Seems to me that people can and will play in the way they like best, and the communities decide which to run more. Why is that bad?

# ? Sep 8, 2023 19:05

pumpinglemma: Apr 28, 2009; DD: Fondly regard abomination.

For a lot of games, I would guess there�s a danger that splitting the leaderboard is going to turn one community that�s big enough to sustain itself into two communities that aren�t. I�m generally pro-turbo but I do think it�s something that should be decided on a game-by-game basis by the runners themselves.

# ? Sep 8, 2023 19:16

Jose Valasquez: Apr 8, 2005

Hobojim posted:

Thank you for not being a flippant rear end in a top hat about it

Definitely agree it's game by game, I guess I've just never really understood why one category being considered "real" is that big of a deal. Sounds like Chrono trigger doesn't benefit meaningfully from turbo but what is meaningful varies from person to person and scene to scene. Does it seem like a net negative that Chrono trigger split leaderboards to you? It sounds like a non-issue.

Seems to me that people can and will play in the way they like best, and the communities decide which to run more. Why is that bad?

I think the split is pointless because nobody does the turbo categories since non-turbo are the "real" records. I speedrun games for the competitive aspect and running turbo is like running a race by yourself and declaring yourself the winner, it isn't very satisfying. I could get a turbo WR in a single attempt considering the sole turbo run is 45m behind my non-turbo PB, but why bother?

I much prefer the route FF5 took which was to just add a Turbo tag but keep them in the same leaderboard.

Granted, I exclusively run JRPGs where the primary benefit of turbo is to take pee breaks during cutscenes you've sat through a million times, I would have a different opinion about something like Ninja Gaiden where the ability to mash at a fast speed is actually an integral part to the run itself

# ? Sep 8, 2023 19:24

heehee: Sep 5, 2012; haha wow i cant believe how lucky we got to win

you can go to the bathroom without pausing the timer

# ? Sep 8, 2023 20:08

Zamujasa: Oct 27, 2010; Bread Liar

Hobojim posted:

Some people can't do a hard thing so nobody should be celebrated for doing the hard thing is certainly a take

I'm all for turbo in games with lots of mashing but I don't know why separating leaderboards for allowing turbo vs not isn't the obviously simple solution.

(Turbo is actively bad for Startropics so I don't really have a dog in this fight regardless, but hitting buttons with correct rhythm is a huge part of the game and I don't think I'd like that being easily automated, makes the run less interesting imo)

I mean, if you don't want to use turbo, be my guest. People can be purists about poo poo and I don't mind because in the end I'm just here to sometimes watch people play games fast. I just don't like seeing people injure themselves over it, and I know enough people who have that I have a strong opinion about it. I'm allowed to be wrong; nobody is going to read my post and decide to magically say turbo is now 100% allowed in all circumstances.

At the end of the day most speedruns are built around wholly arbitrary rules of what is and isn't allowed by the people who run it; the game itself doesn't give a poo poo.

TACD posted:

This is amazing but why the gently caress are people still doing manual splits? Surely we have the technology to like, take a screenshot of the level transition screen and tell the timer to set a mark when that screen comes up?

Running an entire second setup to analyze and monitor your stream for auto-splitting is a lot more complicated than it might seem, especially when you factor in that on analog consoles you're not going to get a 1:1 replica image every time. It's also another maintenance and upkeep issue.

The cost of doing splits manually is cheap and the only time it really comes into play is if you're submitting a run somewhere that needs that kind of strict timing, in which case you throw the video into an editor and manually time it.

# ? Sep 8, 2023 20:10

Hobojim: Oct 31, 2011

Ninja Gaiden was exactly the game I had in mind for this too.

I guess I just don't have that mentality with speedrunning, where the large pool of competitors is important. Smaller games still have thriving communities, and splitting something sizeable like ninja gaiden would still make bigger groups than a lot of those. All it takes is two people who care about it for competition, so if you took the record for Turbo Chrono trigger from your example, and the old record holder wanted it back, that's enough for me.

Makes sense though, thanks.

# ? Sep 8, 2023 20:12

heehee: Sep 5, 2012; haha wow i cant believe how lucky we got to win

ya i wish autosplitters were way more common / easy to implement. even sm64 doesn't have a working autosplitter for VC (gods console)

# ? Sep 8, 2023 22:33

ilmucche: Mar 16, 2016; What did you say the strategy was?

Zamujasa posted:

At the end of the day most speedruns are built around wholly arbitrary rules of what is and isn't allowed by the people who run it; the game itself doesn't give a poo poo.

This reminds me that DOOM (2016, the good reboot) didn't have a glitchless category for a while because there wasn't a consensus as to what that meant. The only recognised speed run for a while was the any% run that spent most of its time out of bounds not killing demons.

It got played at gdq and so many of the comments were along the lines of "wtf? It's doom, rip and tear the huge guts"

Thank goodness zero master was there to make awesome runs anyway.

# ? Sep 9, 2023 02:37

Petiso: Apr 30, 2012

Zamujasa posted:

Running an entire second setup to analyze and monitor your stream for auto-splitting is a lot more complicated than it might seem, especially when you factor in that on analog consoles you're not going to get a 1:1 replica image every time. It's also another maintenance and upkeep issue.

The cost of doing splits manually is cheap and the only time it really comes into play is if you're submitting a run somewhere that needs that kind of strict timing, in which case you throw the video into an editor and manually time it.

See, I don't speedrun games so maybe I'm missing something but this is a thing that has always bugged me, isn't it terribly inaccurate having the players manually start and end the timers? In a long game where different runs usually are minutes/whole seconds apart that's fine and dandy, but isn't this a problem with games where new records consist of scratching fractions of seconds? You'd think the runs would be later carefully analyzed and from what I've read that's kinda what happens but I see runners always start celebrating like it's a given they didn't mess anything up.

Is it because I only watch runs of actual record holders and there are actually dozens of disqualified runs because they started the timer 0.3 seconds too late or stopped it 0.3 seconds too soon that I'm unaware of or what?

# ? Sep 9, 2023 07:17

heehee: Sep 5, 2012; haha wow i cant believe how lucky we got to win

Petiso posted:

See, I don't speedrun games so maybe I'm missing something but this is a thing that has always bugged me, isn't it terribly inaccurate having the players manually start and end the timers? In a long game where different runs usually are minutes/whole seconds apart that's fine and dandy, but isn't this a problem with games where new records consist of scratching fractions of seconds? You'd think the runs would be later carefully analyzed and from what I've read that's kinda what happens but I see runners always start celebrating like it's a given they didn't mess anything up.

Is it because I only watch runs of actual record holders and there are actually dozens of disqualified runs because they started the timer 0.3 seconds too late or stopped it 0.3 seconds too soon that I'm unaware of or what?

if you're good enough at one of those insane millisecond pb games you kinda just know if you did good enough, but ya basically every run of every game is retimed by mods unless it's just some joke meme 1 off run game

# ? Sep 9, 2023 07:22

Jabor: Jul 16, 2010; #1 Loser at SpaceChem

nobody gets disqualified for stopping their timer early, if it's one of those things where frames are gonna matter then you just use the video to measure what the time should be according to the category rules. (nobody accepts unrecorded world records these days - if it wasn't recorded it doesn't count).

the runner's splits and live timing are mostly for the benefit of people watching live and for the runner themselves.

# ? Sep 9, 2023 07:32

Zamujasa: Oct 27, 2010; Bread Liar

Yeah, re-timing is basically "download VOD, open in editor of choice, seek to when the run starts, frame advance until it does, then seek to where it ends, frame advance until it's over, and then it's just End Time - Start Time = Run Time. If you're not in a hotly contested leaderboard, it can be as simple as just "pause about where it begins and pause where it stops", since ±0.5s isn't going to matter a whole lot if people are still making gains of 30+ seconds or multiple minutes.

Of course, you only have to actually do this if you get a run worth submitting... so you don't really do it too often. And the tighter the leaderboard is, the less often you get a run where you have do it. :v:

# ? Sep 9, 2023 12:01

DanielCross: Aug 16, 2013

Yeah, I think the confusion here comes from the idea that the runner's manual splits are THE splits, the only measure by which the record is checked. They are not, they are a generalization to help the runner and audience have an idea of how well a run is going in individual segments, and when the run is actually submitted to a hotly-contested leaderboard, something more precise is generally used to check the timing.

# ? Sep 9, 2023 21:21

Petiso: Apr 30, 2012

DanielCross posted:

Yeah, I think the confusion here comes from the idea that the runner's manual splits are THE splits, the only measure by which the record is checked. They are not, they are a generalization to help the runner and audience have an idea of how well a run is going in individual segments, and when the run is actually submitted to a hotly-contested leaderboard, something more precise is generally used to check the timing.

That sounds logical, what confused me was seeing someone beating a world record by a second or less while live streaming and jumping and crying in celebration before the run is even submitted anywhere, like the possibility of an imprecise timer isn't even on the table but heehee's explanation about people who speedrun games with runs so so optimized can usually know whether they did every frame-perfect action correctly or not clarified it for me, thanks.

Petiso fucked around with this message at 22:33 on Sep 10, 2023

# ? Sep 10, 2023 22:30

TACD: Oct 27, 2000

DanielCross posted:

Yeah, I think the confusion here comes from the idea that the runner's manual splits are THE splits, the only measure by which the record is checked. They are not, they are a generalization to help the runner and audience have an idea of how well a run is going in individual segments, and when the run is actually submitted to a hotly-contested leaderboard, something more precise is generally used to check the timing.

Nah I know runs are properly re-timed later, but TBH I feel like that makes it more perplexing that auto-splits aren�t used. It doesn�t matter if people set them to trigger at different points or if it triggers early or not at all!

I guess I�m imagining myself in the position of doing a speedrun, I absolutely would not want to have to take my hands off the controls to smack a button and potentially hit the wrong one or fumble the controller or lose concentration.

# ? Sep 10, 2023 23:05

Complications: Jun 19, 2014

Auto-splitters are used that both tie into game memory for emulators or into monitor display state, and well trodden games like Super Mario have preconfigured ones basically ready on download. They aren't frame perfect hence manual re-counts, but it's safe to say that any time as large as a third of a second is probably accurate and reason for celebration.

# ? Sep 10, 2023 23:18

Zamujasa: Oct 27, 2010; Bread Liar

If you factor in that the split, done manually, might be a little late (reacting to finishing the game vs. hitting start at the same time you do a run) you can also generally assume that the timer is equal to or higher than your actual time, and if that comes in below a record you have a pretty good chance of having just beaten it

# ? Sep 11, 2023 01:23

DoctorWhat: Nov 18, 2011; A little privacy, please?

The thing was Super Mario Bros players is that you can tell your time from the hammer throw patterns that Bowser gives you. These top players have practiced 8-4 on the WR frame rule and really can eyeball whether they're going to make world record just from that information.

# ? Sep 11, 2023 03:15

Blurb3947: Sep 30, 2022

It must be really disappointing to find out you didn't get a WR a split second before you find out the exact time.

# ? Sep 11, 2023 04:45

Skippy McPants: Mar 19, 2009

Petiso posted:

That sounds logical, what confused me was seeing someone beating a world record by a second or less while live streaming and jumping and crying in celebration before the run is even submitted anywhere, like the possibility of an imprecise timer isn't even on the table but heehee's explanation about people who speedrun games with runs so so optimized can usually know whether they did every frame-perfect action correctly or not clarified it for me, thanks.

Even beyond being optimized, in this specific instance, Niftski knew he was up by a full Frame Rule (about 1/3rd of a second) going into 8-4. So long as he avoided any major time lost, the WR was certain; the only question was by how much.

Skippy McPants fucked around with this message at 16:48 on Sep 11, 2023

# ? Sep 11, 2023 16:45

Dr. Stab: Sep 12, 2010; 👨🏻‍⚕️🩺🔪🙀😱🙀

In SMB, there's a small number of relevant inputs, and it's pretty easy to tell, for an experienced player, what a correctly made input looks like, and what losing a frame looks like. He executed everything he was going for perfectly, and could tell that he got the right inputs. Even if he was off by a frame or two, it was still a WR by a wide margin because of the frame rule save.

# ? Sep 12, 2023 01:42

Carbon dioxide: Oct 9, 2012

From the new Summoning Salt.

# ? Sep 17, 2023 16:59

Jose Valasquez: Apr 8, 2005

https://www.youtube.com/watch?v=i_9Hj--VfbY

Good video about how absurd Niftski is at SMB

# ? Sep 17, 2023 17:32

mycophobia: May 7, 2008

it's baffling to me that no smb runner had come up with that devils spell thing until the late-2010s. i had played a ton of those japanese smb glitch hacks in the 2000s so i might be a bit more primed to see collision-based exploits but even just understanding how smb1/2 ejects you from solid objects you'd think someone would've thought of that long before then

# ? Sep 17, 2023 17:49

neongrey: Feb 28, 2007; Plaguing your posts with incidental music.

Power up with Pride Winter Pride is open for signups until October 7th. We'll be going Dec 14-17 this year. Hoping to see some of y'all in the submissions.