triangleCZH,

yes, a camera is just fine. we do not care for HD footage, only readable footage. As long as we can make out what is going on, it's fine.

IGT vs RTA... IGT = "In-Game Time" RTA = "Real Time Action"

IGT normally will not count load times, or times where the player may otherwise not be in control of the game and is just waiting, while RTA will count all of the time spent from the start of the run to the end of the run.

In the case of Luigi's Mansion 3, we use RTA for everything except ScareScraper now. For ScareScraper, you'd get your speedrun time by adding together all of the times on the top-left corner of the results screen of each floor of the mode.

If you do not intend to run ScareScraper (terror tower, etc.) then do not worry about this.

Greetings everyone,

As of now, the ScareScraper board is officially going to enforce the rule of using IGT to time the runs. This is the addition of all results screen times into one big time. If you need an easy way to do this, you can simply use Google Sheets, add the times into individual cells, and sum them up.

The top 3-4 times have been adjusted for this and reflect IGT. If your run was not changed to IGT, you may change your run into IGT via the above method and we will verify it to make the change. For ease, we also recommend you paste the results screen times into the run description (as you can see in the altered top 3 runs).

All future runs, regardless of position, will need to submit using IGT from here on.

One important note regarding the conversions of the top player times: Runner ChaserRacer has since privated all of their vods, thus we cannot review his runs. We are attempting to see if we can get them back, but if not, to somewhat preserve the general times on the board, we have opted to instead of removing the runs, to giving an estimate by pretending as if they had the fastest loads possible between floors. This means in reality his real IGT is likely better than what is displayed by a few seconds.

Currently it will not impact 200% and that will continue to use RTA.

Only the IGT will be displayed on the rankings list. As much as we'd like both to be displayed, SRC has the option for IGT vs. RTA as a "all or nothing" deal. So if we enabled it for ScareScraper, it'd enable it for EVERY category in the game.

Heyo!

We would treat this similar to how we treat the ILs. In a 2P~4P team, we will believe in good faith that you are submitting the slowest possible time for yourselves. In the event a WR was tied by a matter of seconds we may re-review the footage to check for player-entry lag that would cause the offset of IGT that you mention, and fix it appropriately if possible.

Obviously, this can't work if the floor spawns people alone instead of as a group, and at that point we can't really do anything. However, in a scenario like this, this would still at the very least reduce the possible offset from "lag/loads" from its potential 4-16 seconds per floor, to just 1 second per floor.

Greetings, everyone.

It's been brought to our attention yet again that ScareScraper load times seem to vary drastically, regardless of if the game is on a digital or physical version. With recent measures to check for timer changes in the IL SS boards, this became more of an eye-sore.

As such, we now have a proposed swapping the full SS boards to IGT just like the ILs. This is actually something we've been debating off-and-on for a very long time. IGT in this case would be the addition of all the results-screen times from each floor into one big time. It would remove all load times and results screen checks from the timer.

The proposed solution to fix the current runs on the board is simple. If we go through with this, we will alter the top 3 runs of each part of ScareScraper to IGT. All future runs will need to use IGT. If your run is not within the top 3 of any given ScareScraper subcat, you can simply edit the run with the IGT and we will review and verify it. This is to keep the workload light while accurately reflecting the top times. The boards will look slightly jank because of this, but it is what it is.

We will accept any feedback or additional ideas regarding this until Wednesday, Dec 15th, 12:00AM EST. After this, we will begin to make the changes and will enforce the new IGT rule if there are no objections.

Nothing else will change regarding ScareScraper at this time.

Howdy.

I conducted some tests with Jennuflect on the matter. I own a Physical copy, they own of a Digital copy. My copy is the on switch itself, their's is on the SD card. Both on version 1.4, without costumes, and the same general geographical location.

Entering the Scarescraper solo, after about 40 total tests (20 for each of us, 10 using local option and 10 using online option, remaining solo), we both came to around the same general loading time.

Timing: As soon as the game screen starts zooming in after the 0:00 countdown, to just as the purple fog begins to lift to reveal the hotel.

For local, mine was an average low of 13 seconds, theirs was an average low of 14 seconds. We both had an average high of 17 seconds.

For online, both of ours were an average of 15 seconds. We both had highs reach as much as 25, as well.

Note, that there could be slight discrepancy in how both of us timed our loads, but we tried to coordinate it as best as possible, I wouldn't expect more than a 0.5-1.0 second discrepancy.

This tells me that there MIGHT be an ever so slight difference between owning a physical vs. digital version of the game, however it seems to be no more than a second or so per floor should this be a fact. Similarly, it would appear being on SD vs. Console shouldn't matter much at all, given the previous concern that SD may be slower, since it appears despite my being Physical+Console, and Jennuflect's Digital+SD, our differences again are merely a second at best. Despite this statement though, I believe there can be another reason for the loads being different at all.

What I am lead to believe is going on regarding higher loading times is various factors, such as simple RNG on the room generation, the health of the console and/or SD card, the connectivity to the Nintendo servers, and if physical, the health of the cartridge. These are all factors that can impede on loading given the nature of ScareScraper. Connectivity is what I presume to be the largest culprit, as whoever has the slowest connection should naturally make everyone else wait longer as well. Jennu and I are very near each other, relatively speaking, so our connection strengths should be similar and thus may have helped lead to the similar load times on the Online testing, but you never know what else may be going on as you try to connect to the servers, and what may be a good connection one day can be bad the next. When you did your tests, was anyone else in your home also using the internet at any point?

Overall, it would seem we have conflicting data. Yours seems to show there is a very clear discrepancy, while mine shows there is practically none. I would be lead to believe that in an isolated environment of just purely Physical over Digital, there is no difference, but whatever else factors that may exist on a person-to-person and console-to-console basis can greatly change things.

Comment test?

Comment test

It has now recently been brought to our attention that there is the chance for in-game time to be inaccurate.

Let me preface by saying that this doesn't matter at all except in the case of a contested WR, so don't even bother thinking much about this beyond as an explanation for why submissions that lead to a tie in WR may potentially have times reworked.

Also hi, I like to give overly long explanations to simple things, rambling is my game.

Anyway, the discovered problem is that during Team Games, players can join floors slightly later than others by a few milliseconds. The in-game timer is not universal and is instead tied to each individual player, starting about 1.2 seconds after they personally spawn into the level. Should a team finish a floor just as a second turns, this has a chance to have the floor come off as a whole second faster than it really was, for the latest-spawning player.

For example, say a team is doing a 4P floor. Player 1 loads into the floor first. Player 4 is recording the attempt and is the last player to spawn into the map at about 150ms later than Player 1 did. If the floor takes 1 min 20 sec 100 ms to complete for Player 1, it will display as 1:20 for Player 1. However, due to Player 4 spawning 150ms later than Player 1, their timer will read as 1 min 19 sec 950 ms, it will display as 1:19 and would make the submitted time technically qualify for a 1:19 instead of 1:20 since Player 4 is the one who is recording their results.

This is a very rare occurrence for 2P~4P teams and generally will only happen if you complete the floor within frames of the timer swapping to the next second. but still is something that can happen that will lead to inaccuracies. We will always go with the slowest time within the team as that is when the earliest action has occurred in the run. We will not sweat this difference except for when it is crucial to decide who truly owns a WR.

Now. How will we handle this? Well, in the case of a tied or near-tied WR (1 second difference), we will check the delay in milliseconds in which the recording player spawned into the floor relative to the 1st player, and add those milliseconds to the in-game timer. In cases like this, the submitted time may be changed from the results screen time.

For example, say Team 1 submits a WR run for 1:19 while Team 2 submits a run for 1:20. We would inspect and see that Team 1's time is, as far as the recording players PoV goes, a 1:19.990, with the recording player having entered the floor 30 ms later than the 1st player, so we'd adjust the time to 1:20(.020). Team 2's time ends up staying as a true 1:20, so the teams in the end are tied.

"Why not just display milliseconds to decide ties?" Because we want to keep true to the format of the in-game timer of SS and keep it to solid seconds, while at the same time wanting to keep it fair, as it would suck to lose WR because your recording player spawned 10 ms faster than another team's recording player that decided a second turning or not for their results.

The question has been answered and an IL for Rare Ghosts has been created

Hello!

Yes, teams are still required to be with friends. Thank you for letting me know I forgot to add that detail in.

As for adding in Rare Ghosts, we have been reconsidering adding it in since making the post. However, our concern with it currently is of the Rare Ghost Types. 1.3 and 1.4 have added new ghost types, and for now we have not verified if those new ghost types require buying the DLC or not to access. We are currently looking for input! (And this question goes out to anyone!!) Do you know of anyone who does NOT own the DLC, and is on Patch 1.4, and have they encountered the new ghost types before?

Greetings!

A change to the board has occured today.

Individual Level categories have been added for ScareScraper. Each mission type in ScareScraper + Boolossus now has its own unique leaderboard that can be submitted to. These will be single-floor submissions rather than full ScareScraper runs. We feel with how popular ScareScraper has remained, this can serve as a nice addition to the boards, as now if a run goes bad there is always a shot at the ILs for times.

The same basic rules from ScareScraper apply there:

• Local co-op (Player 2 as Gooigi) is banned
• Teams must be made with friends only.
• The player count must be consistent (players cannot disconnect)

Additional rules:

• The results screen must be displayed at the end of the run
• The category you submit to must match what your floors mission is.

Timing will use the in-game timer (the time you see on the results screen). In the event of ties, we may recount to check for timer lag, but milliseconds will not be used to break ties.

If you do not wish to submit standalone videos of the floors, you may also submit your pre-existing or future Full ScareScraper runs to these, however your video link must be correctly timestamped to the specific floor you wish to submit for, or the run will be rejected.

If you have any questions and/or desire clarifications feel free to reply below.

As well, it is the very fact you select "you are player x" that changes the order, likely because the site goes thru an extra hoop when you select as it isnt about the position itself but more as a quick way to throw your name in automatically.

So while the order means nothing, if you want it that way, do not have yourself specified as any player, just type your name into the slot

Good day, everyone

As some of you may have noticed, yesterday we have created Category Extensions for the game. With this, a few new categories can be found there; A Rank, All Clear Gems, 200%, and Putt-Putt%. We hope these can serve as fun extra other ways to run the game! We will also be open to any ideas you may have for additional categories within it.

As well, we have moved All Gems, All Boos, and All Collectibles to the Category Extensions. The reason behind the move is we've felt they are not distinct enough from any% or 100% to really warrant being on the main board.

Lastly, No Major Skips has been moved from Misc. to the Main Board, serving as a way to run the game without the need to rely on having a Seed.

The board can be found through this linkas well as the sidebar: https://www.speedrun.com/lm3ce