Originally I had planned to write up a post when I had reached a score of 3 billion for Imperishable Night extra. However, as the plan of simply getting 3 billion have gone out of the window, I will write something up now, mostly about what will happen with the category and such.
What do I mean by saying that the plans for 3b have been thrown out of the window? Well, I don't mean that I'll quit. I couldn't quit if I wanted to knowing that I have some really shitty things in my personal best. What I mean is that I won't be shooting for just 3b, but rather to beat ASL, which means 3.1b. Now, ASL's run has been the world record for over five years - and for a good reason, I might add. It's a very strong run with few actual mistakes. Add to that that only two different people have achieved a score of 3 billion (to my knowledge) and one can maybe see why it's been a world record for so long. "But then," you might wonder. "Why are you trying to beat it? Are you crazy?" In short, yes, I am. The long version of that is that while his run is strong, it's still old. New, better strategies have been found and his grazing isn't unbeatable either. Which is funny because that's a part of why his run is so strong. With the exception of Honest Man's death, I haven't seen someone graze better in a run. It doesn't even matter which attack. He has strong grazing on all of the ones you milk. But as I practiced, and practiced and practiced some more, I noticed that his gains aren't super amazing. And that everyone else just didn't have a good strategy for grazing or just simply weren't good at it. ASL's time-gains on the big spell cards are as followed:
Ichijo: 7,400;
First nonspell: 3800;
Xu-fu: 13,600;
Honest Man's Death: 16,200;
Hourai Doll: 10800;
Imperishable Shooting (not really that big, but I'll list it anyway): 3600
Out of those gains, Xu-fu is most likely the strongest followed by Hourai Doll; and Imperishable Shooting the weakest, followed by Honest Man's Death. And to me, it's understandable why those two in particular were the weakest spells. Gaining time on Honest Man's Death is dependent on how high you are on the screen, but this naturally makes the attack also quite a bit harder, in a different way than it does for Ichijo or Xu-fu. Once you've set yourself on a height, you can't back out and go lower, while on other attacks, you can go a bit slower and less risky if you feel that it's needed at that moment. And Imperishable Shooting should be obvious to anyone who's ever had a good run near the end. He was obviously shaky, and did not want to take unnecessary risks. Even if ASL had done a 40 million Imperishable Shooting capture (which is roughly 4400 time), he would have gotten around 10 million more than he did. Not a very worthwhile risk on a run that would beat the world record by almost 100 million at the time. Some of you reading this may be thinking to themselves now: "Well, by how much can you beat his time-gains, or the score overall?" And sure, I'll touch on that now. Let's start with the specific time-gains; I'll go by my best performance in practice, which obviously means that it's unlikely that I'll match that in a run. Imperishable Shooting in particular is questionable, so I will add a number for each spell which I think is realistic for me to get in any given run.
Ichijo: 8,100 (7,600-7,700 in a run);
First Non-spell: I don't really know, actually. I never saved a replay with a good first non, but I should be able to gain around 3,500, which admittedly is less than ASL, so I have to work on it;
Xu-fu: 14,300 (13,600; I should be able to match ASL, maybe slightly worse, but I'll make it up!)
Honest Man's Death: 18,000 (Anything from 17,000 to 19,000, really. My height is slightly different every time, but virtually never so low that I get below 17,000. 18,900 is the highest I've seen, but should also be doable for me with some luck. Maybe I'm being optimistic here though. Even with a mid 17,000 it's most probably my biggest improvement over the world record still.)
Hourai Doll: ~12,300 (11,500+, Hourai Doll is a bit difficult to tell because at this point in the fight, Mokou has spawned over 1,400 familiars during the course of the fight. If I go by the formula 0 < a <= 4 familiars: 16 * a + 6 * a * a + 2 * b, I should gain something over 2,800 time during the cancel of the attack. "a" being the familars on the screen and "b" being the familiars that have been spawned during the fight overall. I am not sure this is 100% accurate though; maybe I am overestimating the cancel here)
Imperishable Shooting: 4,500 (4,000-4,200. It is not unrealistic for me to get a 40 million Imperishable Shooting, which again, would be 4,400 time. But I think once I am in a run, being shaky and all, that I would not dare go for it. 40m needs solid character control, and some more risky grazing which I am not too sure I could do being undoubtedly shaky at that point in the run, so I'll opt for something less extreme)
And do keep in mind that every bit of time gained over the world record also means more graze, as well as some more score in the end-bonus. Beating ASL by 300 time on everything would add up quite nicely. Another thing to know is that while we're pretty even on most other spells, I have better strategies on at least three of them. Those are Flying Phoenix, Woo and Fujiyama Volcano. The time-gains aren't very big (100, 200 and ~300 respectively in that order), but still worth nothing, especially since I also kill Flying Phoenix and Fujiyama Vulcano faster, thus losing less spell card value. There is a bit more to it for Fujiyama Vulcano, too. My bullet cancel is larger as well by a few millions. To me it seems silly to say this myself, but at this level I plan to play at, a few millions is significant enough to go for it. There are other attacks where I have the potential to gain time or score on ASL, such as Flaw of Forgiving Shrine or Phoenix Tail. With some luck, you can cancel with more familiars, but it's not worth it waiting for it, since you would lose too much spell card value. It's also risky on Phoenix Tail, since it's easily possible to get hit. So I'll leave those up to luck.
Now for the things I am worse at and that I urgently need to improve. I already established my weakness on Mokou's first non-spell, I'm also really bad at the last non-spell. As it stands, I bomb it at least once every time, which is over 10 million points every time I face it. I can not allow that. The other thing I am much, much inferior with is the stage portion. It may seem obvious that if my boss fight is much stronger that my stage has to be worse, but it's just that I am a lot worse than I really should be. Out of the top five runs, my stage portion is the worst by a bit, and by far if we leave out AM's run. It is not only bad generally, but I'm very inconsistent in getting even acceptable cancels. And my stage cancels are also the reason why at first I wrote off the chance of beating ASL and settled for just 3 billion. But people rooting for me set ablaze a new flame and I want to at least try to do my best and not be lazy with the stage portion. First, I have to see what I am doing wrong fundamentally. It's not a matter of execution right now. It's a matter of understanding how to reach the numbers I currently need and only then does it become a matter of executing it. How consistent I will be at the end of the day, I can't tell. I don't even really know if I'll actually find out how anytime soon.
I talked about what I am better at (Mokou's fight) and what I am worse at (the stage portion, 2 of the non-spells), but assuming I break even on the stage, by how much can I beat ASL? I didn't do the hard math, and doing it would actually take quite some time I am not willing to invest for something as silly as that. I can give an educated guess though, but I'm not sure how accurate it will be. Parts of it might end up being nonsense, I will admit, but here goes: Imperishable Shooting alone has room for improvement by 10 million, and assuming that I got 1000 time more on Honest Man's Death, that is roughly another 5 million. If we say 1000 more time on Hourai Doll as well, that would again add about 5 million. Those three improvements alone would gain roughly 2800 time which would be added in the end-game bonus (which is time x 1000), so again, another 2,8 million. Overall, I believe I can potentially save up to 35 million, if I broke even entering Mokou (and the stage portion can be actually improved by ~1500 extra time, which is about 2 million just from Mokou's items + end-game bonus, albeit this implies an absolutely, astonishingly good stage portion and Keine fight). In a game like Imperishable Night, that isn't a lot, but if I improved it by only half of that, it would me a tremendous success to me, so I'll definitely try my best.