Category Rule Discussion
9 years ago
Colorado, USA

I guess we should decide on what categories/rules are what. I'm not deadset on the categories that exist so far, just thought I'd get something up there. Right now I've got:

100% Messenger Transcend any%

I suppose 100% would include unlocking the 3 Messengers (A/B/C), All Sigils, Stars, and Items unlocked (fan/block/connector/etc). Only thing I'm not sure on is how to "end" the run. Messenger ending has highest requirements, but you don't really "unlock" anything new just by ending that way. Same with transcending, you use up the sigils you've found, but don't otherwise do anything besides solve another puzzle. So, I guess the fastest ending would be to Eternalize. Thoughts?

Messenger ending category is low% I guess, since you can skip the red sigils for Nexus Floor 2.

Transcend ending would be the same, low% skipping Floor 2 sigils again, as well as any sigils needed for the Platform unlock.

any% I guess we can start off by setting it to "the fastest way you can end the game with one of the 3 endings, Eternalize, Transcend, or Messenger.

Germany

I think 100% should be all sigils, all stars, all Items, all messengers, arbitrary ending. There's a case to be made that 100% should include collecting all terminal messages and audio logs. But I don't think many people would want to run/watch this category if it meant they had to do that.

A few more ideas for categories:

  • All main area sigils: Collect all sigils from the three areas (i.e. without star levels) and complete the game
  • All sigils: Collect all sigils including the ones from star levels (i'm totally not trying to get this included just because i have wr there, no...)
  • 129 sigils: Equivalent to all sigils for now, but maybe theres a way to skip the stars...?

I think trancend = any% = low% for now. any% because, well, it's the fastest we know of. low% because you skip all stars + 9 red sigils. Messenger get's the stars, which disqualifies it for low% imho.

France

Hey guys what about the fast forward feature ? It's been incorporated in the public beta. I'm just out of words. I read the patchnote and i thought : Naaahhh they didn't just throw in a developper cheat as an ingame feature... Well yeah. They f¤¤¤ing did. (On a sidenote they also removed the possibility to pass a jammer through a forcefield with the help of a box. I wasn't aware it was possible in the first place, but it's a shame to see it go.)

Anyway, regarding the rules and the Fast Forward feature, i think it should be prohibited. I mean... it's originally a cheat. Basically it just speeds up the game's rate (if you activate the game's timer in the advanced options, time goes faster when ff is used). It would reduce RealTime but IngameTime would strictly be the same. Speaking of which, what should be used, real time or ingame time ? Personally, my computer is not really a powerhouse, and when you sum up the time i lose for each loading screen in between maps, it adds an extra 3 minutes or something to my final real time. Not ideal i would say.

What do you guys think ?

Colorado, USA

I agree with 100% as far as no message/log completion. We were discussing in Grokken's stream that maybe 100% should include all puzzles, to make it less arbitrary. That means all Tetromino puzzles, Transcendence tower puzzle. This is in addition to the aforementioned requirements.

I've added All Sigils. Not sure about "All main area sigils". Isn' that just Eternalize Ending?

I think THIS_ONE_ASSHOLE has an any% that is not equivalent to Transcendence, though I haven't reviewed the video, but I think Transcendence and low% are equiv for now.

As far as Fast Forward, I vote strictly banned unless a variable is added to the category (Transcendence w/FF), as opposed to appending the "no-FF" variable to all the original categories.

And finally, as far as the RealTime vs. IngameTime goes, I vote RealTime, since that's what most of the runners have been using and splits are just nicer to look at while watching a stream than seeing one number right at the end.

If and when these routes really settle down into the 5-10 minute difference range, we can enable a debug setting that keeps track of load times.

France

The Any% video you talk about might be the Public Test run, or my IL World A run... Anyway, i agree with everything else, sounds fair.

United States

Not much to say, but I agree with pretty much everything.

Sweden

Pretty sure Eternalize ending is faster than Transcendance, my best split is 9:40 from C2 to end. I doubt doing 9 extra puzzles takes more than 9:40, even if theyre slow.

Fast forward is great for casual play so you dont have to wait, terrible for a speedrun because you would be forced to use it pretty much everywhere. It is just way too fast for anyone to control properly. I don't think we need to add "no ff" to the category names, could just make it banned as default.

I don't see how we would measure ingame time? The time you get for restore backup isn't the final time, for transcend its your time at floor 5 of the tower. I think real time is best. Loading times are pretty much non-existent if you put level caching to ultra.

As for categories, people can run whatever they want with whatever rules they want. Transcend/Eternalize/Messenger is well defined, just get the ending as fast as you can. Any% would just be whichever is the fastest (probably eternalize). Low% would be lowest amount of sigils (transcend + platform skip + skipping a1 gate?). 100% would be all sigils, stars, messengers, possibly all journal entries.

British Columbia, Canada

I agree with banning ff, or making a separate category if people want to run it. For measuring time, I think it should be real time since looking at a backup doesn't give you the actual time, like Grokken said. And a set start should be made (starting the time when you press new game).

Colorado, USA

I think starting the time when pressing new game is arbitrary since 1) it can often include the longest load time for pre-caching 2) you can skip the opening scene anyway.

Alternatively we can just time from first user input (initial movement). This allows you to pre-cache the game and if you reset, it brings you back to the same point as if you had skipped the intro.

As far as Fast Forward, I think the consensus is banned by default. The only exception we were talking about in Grokken's stream last night is allowing FF in any% since it's typically the "anything goes" category, beat the game as bitchingly fast as possible.

If everyone agrees, I will indicate this in the rules for the categories, and add a "w/FF" variable to the other categories (e.g. "Transcendence Ending w/FF").

British Columbia, Canada

Starting at the first user input sounds good to me. People were starting at new game and first initial input, so I thought we should all start at the same time.

United States

I just thought of something.

So I've been working on ways to break the elevator as to make the Messenger route the fastest. There might (extremely wishful thinking) be a glitchy way to do so using floor 6, but you'd think you'd need the star on floor 3 to be able get to floor 6. Turns out, not really. If you collected all stars except for floor 3's, you would have 2/3 numbers needed for floor 6's passcode, and you would only have to guess 10 passcodes (at most) to gain access to the floor. Assuming such a glitch exists, this could the fastest method, in spite of having to guess at random codes.

This raises a question. What if someone was to skip floors 2 through 4 by guessing random codes? In a single segment run, its definitely possible that guessing a combination with 1,000 possibilities is faster than obtain the sigils for floors 3 and 4. It seems it would be unlikely, but still very possible.

But I've been wanting to make a segmented run, and being able to guess level 5's passcode could make the run much, much shorter. All one would have to do is figure out the code, reset to last checkpoint, and then type in the correct one.

So do we allow the guessing of passcodes? My vote is that we do not allow guessing for single-segmented runs, because it is not practical to do so and ruins the quality of the speerun. Yet I think we should allow it for segmented runs, as segmented runs are theoretical in nature and guessing the right code on the first try is theoretically possible.

France

I think one of the devs mentionned on the Steam forums that untill you open SacredNumbers and OsirisPasswords, the passwords don't even exist, preventing you from doing any kind of guessing. Now don't quote me on that, i'm not entirely sure about it, but it would make sense i guess. Croteam likes to prevent the kind of abuse we would be attempting, given it's too game breaking. Another thing i'd like to discuss about the rules : which version of the game should be used ? I'd vote for latest, unless a major glitch/skip is patched. For instance : latest version : jammer shenanigans fixed, oubliette skip fixed. I did the upgrade and can't go back. If i would be to try to beat a flawless run using the previously mentionned techniques, i wouldn't even be able to do so. You guys' take on this ?

Sweden

I don't know how we would play on any other version than the latest one. Steam forces you to update the game when you start it. You cannot run it in offline mode if it isn't up to date. I don't think there is a way to start the game outside of steam unless you have a cracked version.

I don't think jammer glitch saves that much time, it's only used in a few places. It should still be possible to do it post-patch anyway, it will just be a bit harder. Oubliette with connector is like 5-10 sec faster than climbing on the wall with the cube, not a huge deal.

The only major thing I'm worried about them patching is elevator skip. If they patch that the current times will be close to unbeatable, since it saves several minutes.

France

Yeah, that were my thoughts as well... The most endangered skip is the elevator glitch, and i'm pretty sure it will be patched eventually. Let's hope i'm wrong :p