For some reason I didn't see this before so I am way late, but looks better than what we had so far. Thank you!

Here is a spreadsheet that keeps track of least required jumps and similar challenges: https://docs.google.com/spreadsheets/d/1jbb_6cPQiEJdJKR9b1CHl2b2EtUedFzCc_GlOONqGGk/edit#gid=358357598

I can't quite see it being a category on the boards though, mostly because some of these rooms are very difficult to complete so it would make for a rather inaccessible run. It would also be a pain to verify how many times a runner has jumped.

https://discord.gg/GUuxtkF

I'm not too familiar with how the Switch capturing works, what exactly would two videos merged into one movie file look like? Do you have an example video? Either way, you are definitely on the safe side with capturing one continuous video with "regular" methods.

Chapter complete screen is required for verification purposes. It shows the version number, if assist/variant mode was used and if Everest was installed.

If the built-in capture doesn't suffice and you don't have access to a capture device you can always just point a camera at the screen and upload that. This is perfectly valid as long as the quality of the footage is good enough to make out the final time.

You add an "Level (On Enter)" or "Level (On Exit)" split and then specify the ID of the room in the field to the right. The ID of a room can be found by going into the debug level browser (press Tab on your keyboard when in debug mode) and then hovering over the room you are looking for, the ID will be displayed in the top left corner. Livesplit will then split when you enter/exit that room.

You can either uninstall and and then reinstall or set up 2 Celeste directories, one with Everest and one without.

All the categories on the main boards right now have clear category objectives that make sense within the game itself and don't use variant or assist mode. Right now they all have good reasons to be on there I believe. Where Bny% offends that is that the objective of the category, while well defined, is very arbitrary and specific compared to the other main board categories. The whole category was build on a joke after we found the B-Side unlocking A-Side exploit, I mean even the name "Bny%" is just a funny reference, it just happens to actually make for a fun and unique run.

I do agree though that it feels weird on the extensions boards as the only real category that doesn't use assist or variant mode, Bny% is in a spot where it doesn't belong on the main boards but doesn't feel right on extensions either. But I think that has more to do with how we perceive and realize the extensions boards right now, kind of what Silver said. Don't think of them as the place for weird meme categories but rather think of them as the place for all categories that don't belong on the main boards. That is what they are supposed to be, quite literally an extension of the main boards.

I also agree with the sentiment that it doesn't matter too much where the category is as long as people can submit to it somewhere. On top of that this is also bad timing to do a move like this quite frankly. With the DLC coming out soon(TM) we will probably have to re-imagine a bunch of categories on the main boards which supposedly will lead to some clutter. It would suck to move Bny% over there just to move it back again after the main boards get bloated by DLC categories.

Current version on PS4 is 1.2.1.5

We apply the same verification standards to all runs regardless of if they are competitive or not, however since the run in question is an IL run showing the version number isn't required. The run was initially rejected because we couldn't tell if it was the version or the platform that was wrong in the submission. However now that the platform is known to be PS4 I corrected and verified the submission.

I think having statistics on input methods would be interesting, but I'm not certain if the leaderboard is the correct place to gather that information. There is a lot of grey area on input devices that would be hard to cover with the options available (variables), and forcing it onto submissions will end up complicating the verification and submission process with little benefit. Optional variables, as in players can choose themselves if they want to select any value (or even define their own value) in that variable, are a thing on speedrun.com and I can see that being a decent solution, but even then it might create more clutter than what it's worth for. Also similar to what Msushi mentioned, tying input methods to a objective ranking like the leaderboards might mislead people into thinking that certain methods are better or worse than others, and I'm not certain if we want to support that kind of culture/mindset of discouraging people from playing the way they want to play.

Every category is arbitrary as it is always defined subjectively, but there is a difference in how arbitrary the objective of each category is. Comparing MRB to All Hearts is simply a false equivalency, a better comparison would be to a 23 Hearts run where you do everything All Hearts does, except you don't complete 8C. You are right saying that MRB is just as arbitrarily defined as any other category, and that we could have subjectively chosen any other definition or objective for those categories. However our current main categories intuitively represent a more cohesive and complete objective/run than MRB would.

Also just to clarify, removing the AP subcategories for every category but Any% is currently being talked about and considered, given the lack of interest in them. You concede yourself that MRB will probably suffer a similar fate ("I can understand if this will never be a thing that people will play"), so at this point it would be added to the leaderboards just for the sake of adding it.

I can see this become a thing on the Extension boards, so I encourage you to move the discussion there. As it is, I don't see MRB suited for the main boards, given the incomplete nature of the categories objective on top of the similarity to ARB that will probably lead to little to no activity in the category.

This Any% speedrun in 00:00.000 IGT by dahlukeh is allowed under the current ruleset.

Right now, the rules don't have any statement about debug mode or using debug features. Modifications to the executable are banned (among cheat and assist mode), however debug mode does not modify the .exe and is therefore not covered by this. I don't think there needs to be much discussion about banning the use of any debug tools, this needs to be explicitly stated though in the ruleset of every category somewhere, which it currently isn't.

Beyond that though, I think there needs to be a decision made on allowing debug mode for run submission in general, as in if runs that were played with debug mode being on (but without using any debug features) should be allowed on the leaderboard or not. Currently, as there is no explicit ban on it, these runs are allowed by default. However, I feel like the overall community consensus has always been that it was banned. There is a legitimate discussion here that never really happened on if it should or shouldn't be allowed.

From a runners perspective, it's more convenient to allow playing on debug mode. It avoids having to switch around every time you go from practicing to doing runs and vice versa. On the other hand, banning debug mode makes verification simpler. When debug mode is off, simply looking at the final IGT and the version number, as well as checking that cheat or assist mode weren't used, is pretty much enough to make sure a run is legitimate. This is not the case if debug mode is on, as you can't rule out the possibility of someone using debug features without looking at the gameplay itself.

Inevitably runs that have debug mode on will take longer to verify, so a sensible middle-ground solution would be to allow debug mode, but with the expectancy that runs submitted with debug mode being on will have to wait longer for verification. Otherwise, a binary decision on allowing or banning debug mode has to be explicitly stated in the rules.

With the recent addition of ARB IL categories comes a bit of unfortunate clutter. Namely, for 6A and 8A, the ARB objective is identical to the Full Clear objective, which makes them essentially the same run and therefore renders one of them redundant. Due to a limitation of speedrun.com, we can't simply remove those 2 categories in particular. The way this is currently treated is that the ARB categories for those 2 levels are left blank, which isn't ideal and just simply doesn't look very nice.

The way this can be fixed in an elegant way in my opinion is to make ARB a subcategory of Full Clear. This way we can most notably get rid of the redundant 6A and 8A categories, since speedrun.com allows to add variables to only specific levels. It's currently done this way for 1A Dashless too, which honestly works out great, given that both Dashless and ARB are arbitrary, in the sense that they are not implicitly defined by the game itself, and consequently aren't tracked in the journal either. Looking at how similliar Dashless and ARB ILs are on a meta level, it would make sense to treat them the same way in how they are represented on the leaderboards.

One other thing this proposal would do is that the IL grid would directly mimic the in-game IL grid shown in the journal, both even using the same timing method. This is how it used to be before ARB ILs were added, and I always found this to be very neat and pretty.

The only downside I see to this is that the naming scheme could become a bit weird. With the way it currently is, there would be a category called "[Level name] - Full Clear - Full Clear", however this could be avoided by renaming the current Full Clear category something like "Collectibles" and then making the 2 subcategories of that "Full Clear" and "ARB".

Overall, this proposal would not only get rid of the redundant 6A and 8A ARB categories, but would also get rid of some general clutter and inconsistencies, as well as making the IL leaderboards better resemble the way the game represents and handles level times.

Alright, apparently we are voting on it now .... ¯_(ツ)_/¯ So for anyone invested in this, please go vote on the Discord in #server_and_srcom_suggestions_and_voting

I'd assume so too, but it was 3 days ago since the developers confirmed that it will stay, and nothing really happened since, which is the reason why I made this thread. Any% isn't so much a rule change, but more making the rules what they should have been in the first place. Also some arbitrary tournament shouldn't influence the integrity of the leaderboards.

This thread is mainly a reminder of the recent leaderboard discussion going on in lights of a new exploit being found, regarding All Hearts for the most part, but also Any% and ARB. As of right now, there hasn't really been any communication or response from the leaderboard moderators regarding this topic, and I'd love to know their stance on it and hopefully push for change. There seems to be a pretty unanimous agreement from the community about this change, along with the developer supporting it and stating, that the related exploit will not be fixed, so this shouldn't be swept under the rug. @TGH @baldjared @seckswrecks

The proposed changes were:

• rename "All Hearts" to "All Levles"
• add a main category "All Hearts" with a ruleset along the lines of "Collect all 24 Hearts in the game"
• change the Any% ruleset to something like "Reach the credits/Beat The Summit from a fresh file" (remove the requirement to beat all A-Sides)
• change the ARB ruleset to something like "Beat Core A-Side while collecting all 175 Strawberries" (remove requirement to beat all A-Sides and collecting 4 hearts)

I don't want to go over the reasoning for all of this again in detail, but to summarize: These modified rules should better reflect the actual intend of all categories, stating only the objective without unnecessary and redundant restrictions, and therefore officially allowing different routing choices than the current standard. The addition of All Hearts would create a unique and interesting category, that is very separate from the other categories in that it doesn't require any A-Sides to be beaten.

For further discussion about these proposals, please refer to the following links: (1) https://www.speedrun.com/Celeste/thread/vlnn9 (2) https://www.speedrun.com/Celeste/thread/866fy (3)

It's staying.

I agree, routing choices depending on skill level isn't a bad thing, it's quite the opposite. If we split Any%, we might as well make a separate category that bans hyperdashes, corner boosts and lake skip for all the players that don't like the risk-reward of those.

Also personally I'd like for both All Hearts and All Levels to be main categories by the way, although as long as both get added somewhere on the boards I'm fine with it. Both are pretty different from each other, and both are super different from the 3 other main categories. If one would be miscellaneous though, I'd vote for All Levels to be the secondary category.

One thing I forgot to touch on in the previous post is what Ben said, which is that I'd also like to get confirmation first that the developers will not fix this before we actually change the leaderboard. Although personally I do not see why this would get fixed, it took us 6 months to find in the first place so it's not likely to happen casually. And even if it were to happen in casual play, that person knows that they haven't finished the A-Side because they manually have to quit out, so it's their choice if they want to play that way or not.

This category talk might almost need a forum post of it's own, but it's slightly related so here are my 2 cents:

Firstly, to add onto what Revo said in terms of a faster Any% route and hopefully give context to TeDeMos' (and others) fear, the theorized route would include beating 5B instead of 5A, which would at best save 15s, probably less. And to even get that timesave, you'd have to play a perfect 5B, which is way harder than 5A with the required 5B keyless strat becoming the hardest trick in the Any% run. Nobody has finished a 5B run that even comes close to being fast enough to saving time in Any% so far. It would if anything be a high-risk, low-reward route for top players. All timesave is speculative until somebody gets a 5B run that is fast enough to save time (beating current IL WR by almost 40s).

Now even if it were to save more time, I don't think it warrants a category split like TeDeMos suggested, because a category should not reflect a route, but rather a set of objectives (with maybe certain restrictions). I think this is the underlying issue here, and from my understanding what Revo is addressing. All main categories except 100% right now have a ruleset that represent a route rather than the actual objective, or alternatively they describe the "wrong" objective, as in an objective that mismatches the implicit objective of the category name.

To give an example of what I mean, ARB currently says that you have to beat every A-Side, and collect 4 hearts, both of which are completely irrelevant to the actual objective. The definition should be "Beat Core A-Side with 175 Berries collected from a fresh file", plus the obvious restrictions of not using cheat or assist mode. As of right now, this happens to represent the fastest route anyway so it might not be seen as a problem, but this definition still rules out any potential alternative route, be it faster or slower, that would still intuitively qualify as an ARB run. To give a hypothetical example of why this definition is terrible, if we ever find a way to skip the heart-door in Core, there would be a run that collects all strawberries and beats Core, that would be minutes faster than the current ARB run, yet not qualify on the leaderboard because of sloppy rule definitions. Similarly, Any% should only be concerned with reaching the credits from a fresh file, opposing on how to actually do that, like Revo said.

To address what Jared said, this is where the current All Hearts definition is a mismatch from the name. The current All Hearts category should be called All Levels to properly reflect the objective stated in the rules. The definitions of Any% and ARB should be adjusted accordingly too, all 3 categories do not accurately represent the intuitive objective of their very name right now. So with the consensus being that current All Hearts is actually All Levels, the discussion should be if an actual All Hearts category should be added, presumably as a main category. Which, assuming the developers will indeed not patch the exploit, I still think is yes, for the reasons initially stated, plus what Birdo said: It would be a cool and unique category, separate and different from all the other ones.

Just today, Revolucion found out that you can unlock 2A by beating 1B, the same goes for all other A-Sides. This means that you can beat the intuitive definition of "All Hearts", which is collect all 24 hearts in the game from a fresh file, without beating a single A-Side, but only collecting the tape and hearts of each A-Side. This would cut out over 10 minutes (to be more exact, I approximate around 15 minutes) of the current All Hearts run.

Now clearly, this goes against the way we currently have been running All Hearts, which is why I suggest to split the current All Hearts category into two categories, those being:

• All Levels, this will be the way we currently played All Hearts, beating all 24 levels
• All Hearts, this will be the category exploiting this, so you are not required to beat A-Sides here

I can see why people would dislike not finishing A-Sides, as it kind of feels fishy to not beat the levels, but thinking about it I personally came to the opinion that the new All Hearts would be a cooler category probably. People run All Hearts over other categories to play the B- and C-Sides, not to run the A-Sides for the 1000th time. So this would actually make for a more condensed category, cutting out a good chunk of the stuff that we have already seen so many times. This might increase activity in All Hearts too, because many people were probably scared away by the length of the category thus far.

I think there is further discussion to be had if we should have both categories on the main boards, and if not, then which of these should become the main one. However, I do think splitting the category into two is the best way to go, as both are completely valid, non-arbitrary and, frankly put, cool categories that should be represented on the leaderboards. Giving people the choice which on they would rather play is the best way forward in my opinion.