Recently it has come to light that aaaa_5656's WR run utilised the turbo function on a 3rd party controller. As the run is currently by far the fastest run on the leaderboard, it would seem inappropriate to remove the run based off of this usage, as the time turbo saves would not put it faster than any other run on the leaderboard, the decision has been taken to grandfather the run. However a note has been made on the comment on the run, and any future submissions that utilises turbo will be rejected regardless of time. Apologies this has not been dealt with sooner.

Thank you.

In the past week, a glitch was discovered by Geta92, involving stacking two textboxes on top of each other, allowing for interesting effects. These effects that are useful for speedrunning include wrong warps from Birdramon (warping to beneficial places such as Lava Cave and the basement of Grey Lords Mansion), buying glitch items both to "glitch" recruitments into the city that have not been done by regular means as well as setting certain triggers, such as accessing Mt. Infinity before reaching 50 prosperity points, as well as glitching the shops to buy generic items that are coded in the game.

This thread aims to discuss what to do with this glitch with regards to the categories. As mods, we want to hear your opinions, as well as stating our own such that the discussion is completely transparent between everybody. There are going to be three categories on the leaderboards, which are any% (anything goes within the ruleset currently put in place, i.e. no hardware manip/debug), a second category and 100 Prosperity, both of which are to be discussed.

Essentially, the category discussion has boiled down to two different scenarios with this second category and 100 Prosperity.

1. Completely ban the use of the glitch, labelling the categories as "No Major Glitches". This naming convention would avoid confusion as to what is defined as a glitch in the current runs and what isn't, as well as not falling into the trap of calling it "glitchless" and having to ban if another beneficial glitch with slightly less game-breaking properties were to be discovered.

2. Ban certain aspects of the glitch, primarily the setting of the trigger of being able to enter Mt. Infinity and the "fake recruitments", i.e. all 50pp would have to be recruited before entering Mt. Infinity in the second category and all 100 pp would have to be recruited from its source in the 100 Prosperity category.

Please feel free to discuss your opinions on either of these two options, as well as sharing if you have any other ideas on how to approach this that may not fall under either of these two options.

Thanks. Despite this being a discussion on categories (which seems to be notoriously hostile in a lot of speedrunning communities), I hope this can still be kept as civilised as possible!

Okay so a question that frequently comes up is "why is emulator not allowed to save and quit mid-run, whereas console can?". By default, emulator saves and quits far faster than console does and was the primary reason why it was prohibited. However, the use of the "-slowboot" option for ePSXe ends up showing the PS1 boot screen, which reduces the time discrepancy down to roughly 5 seconds, meaning that they are more comparable. So why is it that emu is still not able to save and quit, despite this time difference being rather minor now? This post aims to highlight why.

So currently in the fastest any% route there is only one mid-run reset, and the route before that had 2. So why not just add the time on, that should fix things, right? At least in my personal preference, adding an imaginary amount of time onto a run shouldn't really be a thing, but for the sake of a fair argument I'll ignore that point for less bias. The issue lies with ePSXe itself.

From testing how the console route works on emulator, SydMontague found a quirk which lead to the Kunemon fight (the second fight in the game in that route) to be different, despite having the exact same RNG value. Now ePSXe is not an accurate emulator in any way, but is used as the standard due to the loads being comparable to that of a PS2 console with the fast disk speed option enabled. But enabling save and quit leads into a hypothetical situation where emulator could result in faster RNG manipulations due to this difference in how ePSXe emulates the RNG, meaning that it would be the fastest way to play the game. This is not a situation that should ever occur.

Another issue with emulator is that most emulators are not able to emulate lag correctly for the game, which can speed up certain fights by indeterminate amounts of time depending on what techniques are used, notably the Ogremon 3 fight with WaruSeadramon using Aurora Freeze, a move which generates a lot of slowdown if there are other particle effects happening on screen at the same time. This is a smaller issue but one that should be highlighted regardless.

So why ePSXe? Why not just use another emulator? Well, the issue lies in that the more accurate emulators are only aiming to emulate the PS1 to the highest amount of accuracy, meaning that the loading times are the same as that of a PS1 (very slow compared to PS2 w/ FDS). This is something that might be looked into with some discussion, allowing these sorts of emulators (Mednafen, XEBRA etc.) to save and quit, though the likely scenario is that a route that doesn't save and quit performed on ePSXe may still end up being faster. psxfin is another emulator which is actually completely prohibited, as although it is accurate enough to emulate console RNG, it runs the game faster than console does natively.

The separation of leaderboards has also been brought up, however as mods we do not feel this is an adequate solution. The separation only brings about a split in competition and the idea of an "emulator WR", which would be completely pointless, due to it essentially being a similar run to console only with slightly faster resets, less lag and potentially more favourable RNG manipulations, nothing to do with the players' skill level or, let's be real here: luck. Secondly, Digimon World is not Super Mario 64, it does not have many people submitting to the leaderboards and splitting up competition further seems counter-intuitive when you consider that if you don't care about competition, you can just speedrun with save and quit and not submit to the leaderboards. The third thing is also what emulators to ban and allow, as the best time would require the player to buy a PSP and emulate it that way, as that has the fastest loading times. And if you're having to buy hardware to compete, you could simply buy a PS2 and game? (an NTSC PS2 and copy of Digimon World is not too difficult to obtain if you look around, take this from somebody living in a PAL region who's bought both NTSC-U and J)

If there does happen to be an emulator which can accurate emulate console RNG and does not run the game too fast, then this can be discussed about how to implement the usage of it on the leaderboards while still maintaining fair competition.

Thanks for reading, hope this clears things up.