it doesn't take up the space of one and doesn't claim the relevance of one

why does dumpdome keep saying new categories when it will still be one category

Also remember that the new 10x singlestars won't take up any extra space on the category extension misc list, as there will still be just one miscellaneous category extension for those runs just like we have now, but with the new name 10x singlestar instead of 10x cannonless (and subcategories for each interesting enough star).

So while @dumpdome64 is right in that adding an extra category for each interesting enough star would be overkill, there is also no need for that in order for the board to display the new 10x singlestar runs

Agreed, only subcategories in the "10x singlestar" miscellaneous category Extension

...and only after 1-2 runners run a specific singlestar AND the community/modbase approves

@dumpdome64 Ohh that's interesting to know, I always thought it was 10x sockfolder cannonless to see who can do the consistent setup the fastest without sacrificing consistency. So even though I feel like 10x cannonless changed to the better with Sockfolder, that is an interesting new perspective to look at 10x singlestars - being about strats so inconsistent you want to see who gets the highest consistency

Singlestars that could take this role in the present or future might be 10x tsukishima (might even say you have to die if you miss it?), 10x moving reds, and 10x wdw star 4 elevatorless

I also consider 10x singlestar useful for anything where there's an especially fast way for getting the segment once but not consistently or with lots of skill involved in getting consistent, making the singlestar grind less valuable for full-game applications - e.g. Cannonless, Wallkicks will work, WDW elevator, Snowman's Head, Hot-Foot-It, and Pillarless - I do agree that Breezeless, Owlless, wingcapless Log Skip, and LBLJ would be good 10x segments as well though.

I think each segment would probably need 1-2 decent (faster than 10x slow strats) times to be added, and then be considered on a case-by-case basis

It might be reasonable to make a category "10x singlestar" and put "Cannonless", "Owlless", and "Breezeless" as subcategories. I don't think 10x Owlless needs its own full misc CE, and thus I also don't think it needs 12 runners to be a thing

You let Luigi finish the final level for you. I don't think there are any levels where dying twice to unlock Luigi's assist parts would be optimal, unless you die 1-2 times accidentally which actually happens to the best so you might find some use for it

Oh wait, you can still complete levels as Mario but use Luigi's assist parts, can't you? I think the answer was directed at whether you can let Luigi finish a level for you, and I tunneled on that by thinking the question was fully answered with that citation

However, this is pretty much the reason why, given no better alternative like a steep slope, it is often faster to instead jumpdive-rollout in RTA or slidekick-rollout in TAS. Longjumps are easy and fast but rarely optimal because you do land earlier than after a jumpdive and dustless rollout

Definitely not 2 seconds, but you are onto something - the longer you stay in the air, the more speed you accumulate so it might have actually saved you ~2 frames if you held the joystick perfectly forward and the spacing lined up… more like 1 frame if it happened on the wooden bridge because the bounce bought you less extra air time

think you could have that opinion for any current or removed CE more active than All Purple Coins, including 10 captures

The official standard for how much time is given the mods is 3 weeks. Often times, moderaters will work faster, but they don't have to. 48 hours is nothing to be worried about yet

Tbf, I haven't seen a post about the decision on this very forum, only on Discord and on YouTube so I'd say it's okay to miss it

@Tech Like Minimum Captures or Blindfolded, this is a challenge that could also be played with the intent of speedrunning. The deciding factor here is probably that it's not a popular speedrun

It would be skill-heavy because of the coindupe in Cascade, and some practice for the difference between AUM and 100% for that reason. Still, it probably wouldn't be fun

By one input for one input, I don't mean how many possibilities for inputs are being represented by one button. I mean how many inputs are being executed with one button press. If that is one input (e.g. 100,0 on the joystick), and some inputs just can't be executed anymore, then that's currently allowed because it is one input being caused by one input.

Even if you went in the opposite direction and remapped both the L and the A button to cause an A press, each of those buttons would still cause 1 A press, meaning it's still 1 input for 1 input and that remapping would not be a macro. If we allow a2d with the result of two possible ways to input forward on the joystick, a second A button would be a natural thing to allow as well. If you don't want that, probably restrict the a2d remapping in a significant way, e.g. like @minikori suggested

@Anthopants A gameshark modifies the game's software itself, this remapping only modifies where the game receives its inputs from.

A valid and popular stance is that that shouldn't be modified either, but currently the line seems to be drawn between causing just one versus several inputs with one physical action. It is definitely more authentic than gameshark as you are running the unmodified game, something that even TAS also prides itself of.

I heard that we shouldn't jump the gun on banning this modification. But allowing it doesn't really feel less like jumping the gun to me. Unfortunately, this discussion is only happening after many top runners have already taken initiative and run 120 star with the modification so now we're here for the cleanup. I see banning the remapping (and of course allowing the trick on controller, see above for verifyability ideas) as the conservative option here as the remapping is new and atypical enough to warrant the option of a ban, and I see allowing it as the novelty that could even be in the position of jumping the gun - and this gun has already been jumped before @simply made his Video

One person said they don't feel comfortable quickturning with the dpad as it is unfamiliar extra movement of the thumb. That would probably get better with practice. A question I have though, if you can remap to the dpad of the original n64 controller, why do you want to use a boxx or a gamecube controller at all?

A few inputs from an anonymous WR holder in the romhack community:

• It's only 5 seconds. It doesn't buy you a WR or a good placement to purchase an extra controller so it's not really that worth to go for. Way more effective would be to get better at the game to save more than 5 seconds through skill

• If the community wanted to ban a glitch that saves time, they'd be banning BLJs

• If the community wanted to restrict the pool of input devices, they'd be banning the Hori controller

• We already have a clear distinction between what type of inputs are valid and what are not. Anything that is a prerecorded sequence of more than 1 input is not allowed. Anything that is just 1 input if you make 1 physical input is currently allowed

• Framewalking is possible either way, so it's really just QoL

• This discussion is stupid

I still find myself seeing some difficulty in justifying dpad-style inputs over arbitrary a2d inputs, as both are equally unintended ways of sending inputs to the game. You could maybe say it has a traditional aspect, similar to how the SMM2 Super Expert speedrun only has 6 levels (like the first one) rather than a full screen, and it would feel reasonable, but it still has that slight notion of feeling too arbitrary for the main board to me to allow e.g. (100,0) but not (20,14)

I agree that my initial suggestion, a filterable variable, would run the risk of separating the board. Maybe this could be mitigated by hiding runs with the variable by default (which would now be my strong preference over showing them by default), but some risk of some separation would still remain, and the runnerbase might potentially even become toxic about never getting over this discussion (idk, maybe?). As a result, I am no longer really in favour of even a hidden-by-default filter.

minikori suggested another possible rule for this that would effectively keep a2d off the top of the board - one where you could only either use analog or digital inputs for the joystick. I approve of this as it would improve accessibility (to run the game at all) for the Emulator subcategory, and I don't see it being competitive and thus problematic for full-game. For singlestars, however, you might still have to filter or ban a2d

To me, it has a big impact that SM64 was never intended to receive joystick inputs from a dpad. In contrast, most romhacks are designed for emulators, where it has always been quite normal to at least have access to a2d. As a result, romhacks allow a2d inputs for framewalks like in Star Road's course 8, while the main game has a good reason not to