I'm not entirely sure I understand your question. You need 16 stars to unlock world 2, the minimum amount of levels you need to finish is therefore 6. So I do levels 1 to 6 in World 1, and upon finishing the 6th one (and getting 18 stars) I immediately quit out of the level. Since I have enough stars to unlock world 2 the cut scene will start. On my computer I'm not even able to start any new level after having unlocked a new world. If I spam space instead of hitting ESC I still end up being taken back to the over-world where the cut scene will play, it just takes longer.

The only reasonable explanation I can come up with is that you are somehow running on a different version where the number of stars required to unlock each world is different. Are you running on the current steam version?

I don't really have an answer to the first question. World 2 should unlock after 16 stars, so you should only need to finish the first 6 levels (assuming you get 3 stars on all of them). I doubled check on my PC and I get the cut-scene after finishing 1-6 not 1-7 as in your run. So I'm not sure what is going on there.

As far as the second question goes, once the three stars show up (or really the moment the star count goes above the required amount to unlock the next world) after finishing 1-6 (or 2-6, etc) you can simply hit ESC to quit to the over-world and the cut scene will start playing. So you don't need to mash through it or anything like that. Just be careful, if you do it too early it won't register that you finished the level, and you will have to play it again.

I decided to update the script a bit, mostly due to the fact that the script above is pretty much useless for the "No New Save" category. I really like the new additions, and they will hopefully make running this more enjoyable for people. Here is a link to the new script (see the above posts if you have trouble adding it to livesplit):

https://puu.sh/wDN46/0f80675efe.asl

First and foremost, probably the biggest addition is the ability to automatically start the timer, and this should work for all categories. For the main categories the script will start the timer whenever you hit start on the title screen (assuming you have 0 stars), and for No New Save it will start the timer whenever you enter the first level of any of the worlds. To enable it you have to enable both "Automatically Start Timer" in the custom settings, and "Start" in basic settings.

Secondly I've managed to have the script split at the end of any%. If you enable it (should be obvious which settings does this) the script will split on the frame that you lose control of your character when talking to Benji at the end of the game. So in theory you will not have to worry about your timer at all when running any%, which is nice.

Lastly I've made it so the script can split whenever you enter the first level of any of the worlds. This is mainly meant for No New Save obviously, but it will work fine with any of the other categories as well. One thing to keep in mind is that it will split every time you enter the first level, even if you have entered it before. So if you enter the first level, quit back to level select and enter it again, the script will split twice. It is also important to note that it isn't able to differentiate between normal mode and marathon mode, so it will split when you start marathon mode on any of the worlds.

I'm still working on being able to split at the end of the No New Save run. Haven't found a good way to do that yet, but hopefully that will be coming soon.

I'm about to upload a bunch of individual level runs, so I thought I'd share some of the "glitches" I use in some of those. I'm kind of hesitant to call them glitches though, they are more like bugs or hitbox quirks, but glitches sounds better so I'll refer to them as that.

There are really only two glitches (three if you count using both at the same time), and both are fairly straight forward.

The first one I am going to call death glitch, and it is probably the easiest to explain and perform. Simply put, dying will stop the timer, but you will still be able to finish the level. This means you have to have a shuriken in the air when you die, and it has to hit the remaining enemies. Upshot is that you end up with the time at which you died, rather than when the last enemy was killed.

A good level to show this off in is World 6 - Level 10. (I was going to make a comparison video, but I got lazy. Just try it yourself, with and without dying, and you'll see there is about a second difference)

The second one abuses a weird property of the electricity robots' hitbox. It turns out that you are able to hit them without having to face them, if you just stand close enough. This doesn't sound super useful, but it can often save you shurikens.

The video below is a good example of how it can be useful.

Lastly you can combine the two. Not entirely sure how this works, but my best guess is that you have to run into the electric robot and then turn around and shoot all on pretty much the same frame as when you die. It's not a big window, but it doesn't really feel quite frame perfect either.

This shows off how it works (without dying you can only get about 2.27 on this level)

Death glitch is almost entirely useless for anything other than IL, since it takes the same amount of time RTA with or without dying. Though it can help to make sure you get three stars, and for 100% it can help make sure you get a worse time when doing the marathons (to skip the "new best" animation). You can technically use the other trick outside of IL, but it's really finicky and precise so I doubt people will.

Hopefully that will help people get better IL times, or at the very least explain why I die so much in my IL videos ^^

Sounds good ^^ And yeah, I agree on having it me a misc category. So I'll just go ahead and set it up then.

It seems someone did a run of this game about a year ago where they went through all levels required for All Stars, but didn't start from a fresh save. (I had to reject their run unfortunately) So I think we should look into either changing the rules or creating a new category to accommodate this type of run.

I would prefer to make a new category to be honest. I think starting from a fresh saves makes a lot of sense in a speedrun setting, so I'd like to keep the current category as is. I don't see any problem with adding a new category however. I think the run is different enough that it would be justified.

It would be like a New Game+ category, since you don't have to deal with cut-scenes and you also avoid most of the "New Best" screens. I guess we could call it "All Stars - No New Save" or something along those lines. I don't think it makes sense to actually call it New Game+ since the game doesn't really have a mode for that.

As far as rules go I guess it would be almost the same as the current rules for All Star only that timing should start on entering the first level and that you are not required to start from a fresh/empty save.

Those are just my opinions anyway. Love to hear what others think about this before we make any changes.

Due to a couple of bugs, and it being a bit unreliable at times I decided rewrite the auto splitter script from scratch. For the most part the goal was to: A. have it read values from memory rather than the save file (should be more reliable) and B. make it less of a hassle to set it up.

Here is the link to the new version: https://puu.sh/wmx1G/916c6a0b1a.asl

As with last time you will have to save it to an .asl file and add it to LiveSplit through the "Scriptable Auto Splitter" component. There is no need to edit the file any more, since all the different settings can be tweaked within the layout settings in LiveSplit. This does mean that it isn't as customizable as the previous version, but I think the options I've provided are sufficient (if you really do need it to split on a different star count, contact me and I can add it for you)

There are two quirks with this version. The first being that there is no auto splitting on collectible count. I simply was not able to find any memory address that saves this data. Not too big of a loss in my opinion.

The second one is related to how marathon rank is saved in memory. For whatever reason the content of these memory addresses do not get reset when you delete your save. This means that if you have gotten to a point where the script has split because of you getting S or X rank on a marathon you will have to restart the entire game to do another run.

I have for your convenience added some code that will do this automatically for you. If you enable automatic restarting in the settings the script will restart the entire game given these conditions; the star count suddenly decreases and goes to 0 (should only happen when you delete your save), the script has split at least once because of a marathon.

As with last time, if you find any bugs, or have trouble making it work, feel free to contact me and I'll help out as best as I can ^^

Decided to put together an auto splitter for this game, and I thought I'd share it in case someone finds it useful. It's pretty basic, since all it is doing is reading the values from the save file, but it works well enough.

Here are the steps you need to go through to set it up to work with LiveSplit:

1. Download the .asl file from the link below.
2. Edit the layout in LiveSplit and add the "Scriptable Auto Splitter" to your layout.
3. Go to "Layout Settings" and then to the "Scriptable Auto Splitter" tab.
4. Add the downloaded .asl file to the "Script Path", and edit the options as needed (note, this script is not capable of starting or resetting the timer, so those two options are irrelevant)

Here is the file: https://puu.sh/whL8C/4594152950.asl (edit: fixed a bug where the script wouldn't reset properly when you restarted the game, meaning you had to restart LiveSplit itself between attempts. Should be fixed now.)

You will also need to edit the file itself, both to point it to the save file, and to customize it to your needs. There are comments in the file itself to help you do that, but I'll go through it in short here as well.

There is a variable called path that is defined on line 58 that needs to point to your save file, which you should find in your AppData folder under the path "\Local_10_Second_Ninja_X_Test\asyncsavegroup\10SNX.sav".

In addition to that there are three lists you can edit to change when the script should split. One contains a list of numbers that correspond to the star count at which it should split. One is used for splitting when you get S-rank or higher on certain marathon levels. The last one will split when you gain a certain amount of collectibles.

For any% you will have to do the last split yourself, because there is no way for it to tell when you get to that point. For the two other categories though it is possible to have the script stop the timer at the end of the run. You will also need to manually start the timer at the beginning of the run.

Hope that was explained well enough, if not feel free to ask for clarification. If you find any problems, or if there are any features you think are missing, just contact me here or on Discord and I'll be sure to take a look at it ^^

Oh, right, hadn't really thought about that :P Guess I'll upload my current run anyway, and then just replace it once the final cut comes out. Thanks for setting things up though ^^

Hi ^^

Pleasantly surprised to see someone made a page for this game. I was just wondering if you would consider making a separate category for each book? Most episodic games do this, and it makes a lot of sense in my opinion. There could be one category for each book, plus a last category called "All Books" (or something similar)

I'd really like to run this game, but I don't think I have the time/dedication to do a run of the whole game (not at the moment at least) So it would be nice to be able to run each book on it's own ^^

So a warning straight of the bat, this post will most likely be quite lengthy. Also a quick spoiler, I don’t think this skip is actually possible, at least not until something new is found. I am mostly sharing this here to show what I did find and save people the trouble of trying this stuff for themselves to come to the same conclusion as me. And also, just in case someone has some ideas on how to potentially make it work. It’s not entirely unlikely that I missed something obvious.

So the topic of this post is the “big skip”, which I like to call it. I think most people running this game knows what that is, but in case you don’t, the trigger for the final boss is in the first room of the game from the very start. So if there was a way to get up there somehow, it would cut out pretty much all of the game. So it’s a big deal basically.

We know we can get a shot of on the very top ledge, but we would still need to be in a grounded state to take advantage of that. Busker was the first person who I heard suggest a method for getting up there, it involves getting a shot quite high on the wall and then shooting another one as low as possible right before that hits. Now his approach had two problems, there wasn't enough time to get up to where you could shoot the ledge, and even if you did, you wouldn't have enough time to fall, since you would have to fall two whole blocks down.

I figured out how to solve the first problem, you need to run to the right so the shot has to travel further, giving you enough time to get the shot of at the ledge. The second one seemed unsolvable, until Taiyo discovered that Yuri gets into the grounded state when hitting breakable blocks. I figured that if you could break the top one of the two blocks and fall onto the bottom one, you’d half the fall distance, which might make it viable.

So before I started testing the actual trick, I wanted to check how viable it was. I figured how many frames you get to work with if you move as far right as possible before shooting. And also how many frames it takes to fall down to get to the grounded state. Now, I made a mistake here, because in my test it turned out to work out perfectly. I tested the falling when Yuri had quite a lot of downward momentum. When you actually do the shot you have a lot less and falling ends up taking 6 frames too long. Anyway, here is a .gif comparing the two, and showing how perfect it could have been.

Now I didn’t figure this out until after I actually tested out the trick, so I do actually have video of it. I did all of this using Auto Hotkey to essentially TAS the game. Now, that does work, but only kind of. The problem is that AHK runs on milliseconds while the game runs on frames (and almost never at a constant frame-rate, at least not for me…) so it is very unreliable. When I had finalized the script it would often take up to ten times to actually make it run correctly. It de-syncs a lot, and sometimes for reasons I don’t the game doesn’t take the input for “x”.

(Run the video at 0.25 speed if you want to see everything that happens... Also, this isn’t the ideal way to do this, as you can see I shoot the last shot a bit early, that shot should ideally happen at the same frame as you teleport…)

I also had a second (and a third, but it’s even stupider, so I won’t mention it here…) method in mind. Which included getting a shot at the top of the slope instead of falling down to the block. This one is not as close as the other one, but I thought I’d share non the less.

(Weird cut I know, but it’s not that important… the shot at the slope happens way too late… and as you can see there wouldn’t be enough time to get up and get a shot at the ledge in that time anyway…)

So yeah, that’s as far as I’ve gotten with this. Like I said I doubt this is actually possible. Maybe someone else can come up with something clever though, but for the moment I’m all out of ideas for this skip.

I'm intrigued, but also a bit confused and sceptical. I kinda get removing the collectibles from 107%, since we have a category for that, but I don't really understand why it should be turned into a pure HAPTKEYM category. Not sure if you meant we should remove 107% from it, or if you meant we could just as well remove it. Like I mentioned on discord a run without LMOOB is pretty much just as fast, and it probably wouldn't end up giving you the 107% achievement.

If the proposed category is just "Have a Plan" I kinda feel like we would have to add Diplomat% and pure 107% as well, feels somewhat arbitrary to have a category for one specific achievement and not any of the others. In my opinion the thing that would make the most sense is to add diplomat as a requirement. Whether we keep the collectibles or not doesn't really matter in my opinion. Another question to keep in mind is whether or not the changes will make more people interested in the category. I don't really have any idea, but I guess it's something to consider.

I am much more interested in All Achievements to be honest, and feel like that is a better priority. I'm not saying that a change to 107% is a bad thing, it's been discussed before how weird the rules and requirements for the category are. I just feel like adding Diplomat is the only change that would make the category more logical. Making it a pure HAPTKEYM run would turn it into a completely different category in my opinion.

Well, I'm pretty sure Friday is good for me, as long as it's at a reasonable hour, which maybe we should try and nail that down. I think the earliest I can start is around 14:00 UTC, though it be a lot more convenient if it started at 16:00 UTC or later. If there is a strong preference for a an earlier time I'm sure I can get a day off from work.

I do have a couple of questions though. What do you do if you miss something? I'm guessing that if it's something that's fairly easy to get by starting over you can do that, but does the run then end when the last achievement pops up? What if it's something a bit more difficult, like Diplomat? Guessing the run is just dead then?

Also, I'm aware there aren't really any rigid rules yet, but for 107% (category) the rule was to visit every named room. Are we moving away from that and is the only rule really just to get all of the achievements? Cause then there is at least one room you can skip. (Well, there is more, but I can only think of one that saves time, at least with my current route.)

We should probably also advertise the discord, in case you (Tayio), or anyone else who is considering joining, isn't aware of it. It makes communication both before and during races a lot easier. I don't feel like I have enough authority to share the link for it though, so I'll leave that to Busker, or one of the other mods.

In any case, assuming something doesn't come up, I'm in.

It's late so this will be a short response.

There is a slight problem with doing just collectibles% plus the 107% achievement though, at least if you use the steam achievement rule for 107%. If I counted correctly, the current route for colectibles% is one or two rooms shy of getting the achievement. So if you exclude killing everything you end up with collectibles% plus visit two more rooms, which might add a couple of seconds. So really it would turn into collectibles% IGT.

Maybe my opinion on the matter is a bit biased, or maybe it's more informed, I don't know.

1. I did enjoy running 107%, and I have done a handful of runs in the category at this point. I like the idea of more categories, and I will probably do a run in each, but I doubt I'd start running it as much as the main categories. (but I suppose that is the point of misc categories)

I think we do have to come to an agreement about the rule for 107% first though. As it currently stands our definition of it doesn't match with the achievement's. It doesn't really matter for the current 107% category, you have to visit treeman's domain anyway, but it's fully possible to get 107% (achievement) without going down there if that is the only goal. (Just in case anyone isn't aware, our current rule states that you need to visit every named room in the game, while the achievement is only concerned with the number of unique rooms visited, i.e. "OOB"-rooms count towards it)

The proposed change would still wipe the current 107% category (unless too many people complain about adding "diplomat") and in case it was a concern, I have no problem with that whatsoever. I mean, you'd really just make my run eligible for two new categories, even though the time would be abysmal. So there is no need to worry about a wipe of the current 107%, not for my sake at least.

2. Shot answer, yes. Slightly longer answer, it doesn't really make much sense to leave it out.

3. I don't know why it is an IGT category, but I like that is. It's the same thing I mentioned in the Editing thread, being able to just hit record and play through the game is quite nice. That being said, for "107%" or "HAPTKEYM", it probably wouldn't matter either way.

For "Diplomat" I would really prefer IGT, mostly because if not it would eventually become a question of who can mash the fastest through dialogues. Plus the added benefit of being able to take a short breather is nice. Especially in categories like this where you might want to stop to consider if you've gotten everything up to that point.

4. I already said this, but yeah, I'd probably end up doing at least a couple of runs in each category if they ever were introduced.

5. In all honesty, the current 107% is less hassle than the proposed "completionist", and so it is more fun to run. From a speedrunning point of view though, I would much more prefer a category that makes sense, and as I said, excluding "Diplomat" doesn't.

My two cents on this topic:

1. I agree with Busker on this one. Adding complicated rules will only deter people from wanting to speed run this game, and it might even discourage old runners from coming back. One of the big reasons I started speedrunning this game was that it was so simple to do, just hit record and play through the game. I also don't see any easy (or complicated for that matter) way of doing this reliably.

2. I can't think of any way to detect minor changes, but I also think changes like that wouldn't actually improve someone's time that much. Now there might be a grey zone where the change is small enough to not be noticeable but big enough to give an unfair advantage, but I doubt it would make a huge difference even then. Any large amount of time gained from cheating will probably be fairly easy to spot.

Also, most of the current active runners (including the mods) are very familiar with the game at this point. We would probably be able to tell if someone is pulling off a trick we know to be impossible without modification. If there was ever any doubt I guess we could ask the runner in question for further explanations or proof to verify that what they did is actually possible without modifying the game in any way.

3. I am a bit on the fence one this last one. I see how allowing some modifications might be problematic and lead to grey areas about what can and can't be done, and it would be difficult to draw clear lines. And maybe in some cases purely graphical changes might affect the game in ways we aren't aware of. I do understand how a strict no modification rule is the best thing to go with.

On the other hand, we could also use this to even out the playing field a bit, as Rezoons suggested with the auto-saving, which can cause some trouble. Also, I'm not sure if it does create frame-rate issues for some people, but if you ever go "OOB" the game will start throwing a bunch of errors because it's trying to load the name of a room that doesn't have one defined. We could fix this now. I'm not necessarily saying we should. I'm just putting the idea of bug-fixes on the table. I would love to hear other people's opinion on this particular matter.

As I said, I'm still unsure. A no modification rule is probably the best option, at least for the time being. If we ever did allow some bug fixes, or changes to the auto-saving, I think we would have to put something together as a community. Figure out what part of the game creates issues for players, and in what ways we could potentially fix it. A sort of less-lag-pack that people could "install" if they wanted to is essentially what I'm thinking of here.

This is a bit off topic, but a response to the the last part of Busker's post regarding the hassle of "installing" and "un-installing" custom maps. There is a way to non-intrusively add custom maps, namely to put an entrance room somewhere out of bounds that you can get to from the main game. (My solution was to put it above "Lava Pits", since it's early on and not to hard to get to)

While messing around with custom levels I found a new glitch that allows you to warp an entire room to the left or right. And for once it's actually useful in runs as well. Here is a video of me showing it off:

(I'm aware it looks edited, but it really isn't)

The set-up for this is actually fairly simple, and here's a picture that kinda shows what's going on.

(The blocks you see are from the corner of "Lost Mountaintop", and the other three rooms are "OOB")

The white lines show where the edges of the rooms are and the four tiles that are essential for this trick are highlighted in red. You need one of the bottom of the four to be solid, while all the other three need to be empty. And then all you have to do is walk into the corner of the solid block. You do this by walking of the block and then turning around pretty much immediately.

It seems it is somewhat random whether or not you get warped to the right or the left. I have at least not found any consistent way of influencing it, but there might be something that I haven't thought of. It's still a very new discovery.

This is my best theory as to why it happens. Whenever you try to transition to a room and end up in a solid block the game pushes you back to the previous room. You can see this a lot of places "out of bounds". When you do it between two rooms the game kinda freaks out, but it doesn't end up teleporting you anywhere.

We are essentially doing the same thing here, but since it's on the corner you are rapidly transitioning between four rooms rather than two. So my guess is that at some point the game get's confused about which direction you are entering the room from. In the first part of the video, the game knows you are entering "Lost Mountaintop", but it thinks you are doing so from the left, which then puts you up on the ledge. In the second part the same happens, but with the room to the right of "Lost Mountaintop".

As it stands, if you perform it quickly I'm guessing this might save somewhere between 3-5 seconds (crossing an entire room takes about 6 seconds), so it might be worth doing in runs. Also I can't think of anywhere else this might work, but I probably haven't looked everywhere, so there might still be some place(s).

I went into the page source and found the background image. Here you go: