Advertisement
IhavenonameSDA

ISC scene comments

Dec 11th, 2023
99
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 25.39 KB | None | 0 0
  1. The first 71 clears were done in a single 5 and a half hour stream on a fresh file, as well as a lot of attempts at 5-2. The last 4 were done over the following week (mostly the following weekend) after another 5 hours of playtime. Comments on each scene are presented in the order they were cleared, which is also the order they appear in the video. I tend to have a lot more to say about the harder scenes, which I also tended to put off until the end, so skip to roughly Day 10 if you just want the more detailed thoughts.
  2.  
  3. 1-1 (1 attempt)
  4. Very easy once you notice the gap on the opposite side of the ring, though it's microable anyway.
  5. 1-2 (1 attempt)
  6. Pretty fair even without misdirecting/circling tactics.
  7. 1-3 (3 attempts)
  8. Cirno has a bit more edge than I had remembered - the pattern unfolds very quickly and it can be hard to tell exactly where the safe areas will end up.
  9. 1-6 (1 attempt)
  10. How is this considered impossible?
  11.  
  12. 2-1 (2 attempts)
  13. One scouting attempt with weak damage, and then one attempt to kill it in time. Going right to left really does help on this one.
  14. 2-2 (1 attempt)
  15. The only difficulty here is in knowing what to do.
  16. 2-3 (2 attempts)
  17. This one is mostly just a speedkill - it's possible to not need to manage any bottom waves, though dealing with a single one isn't actually difficult.
  18. 2-6 (6 attempts)
  19. I made a lot of poor misdirecting choices while going for this one, and while I still don't really understand how to manage everything, I understood well enough to get through without too much trouble. Day 2 without sub items was one of the things I was most worried about for the stream - it's not strictly required for 75/75, but starting from a fresh save and committing to no items meant I'd have to actually reach doll sub.
  20.  
  21. 3-1 (1 attempt)
  22. It's 1-1 again. Exactly the same trick.
  23. 3-2 (1 attempt)
  24. It's a DDC pattern, at roughly the same level of difficulty. While not free, I couldn't see this taking more than 3 or 4 tries to get past.
  25. 3-5 (1 attempt)
  26. Without the bottomhug strategy, this isn't hard. With the bottomhug strategy that renders half of the threat impotent, it's borderline free.
  27. 3-7 (2 attempts)
  28. Mokou's resurrection bursts are the only dangerous part of this survival, outside of letting it live too long. I narrowly missed a first try here, though the expectation was probably 3 tries.
  29.  
  30. 4-1 (2 attempts)
  31. Actually pretty tricky, and yet it's Yuyuko's easiest spellcard. She's not playing around in this game (and honestly, I should have left this for doll sub)
  32. 4-4 (1 attempt)
  33. Hold down and shoot. You win.
  34. 4-5 (1 attempt)
  35. Like with 2-2, the only challenge here is in the strategy and knowing what to do. The dodging is slightly more involved but still, this idea could be made a lot harder and it would still fit perfectly fine in a standard game.
  36. 4-7 (1 attempt)
  37. Learning how to misdirect Seiga to deal damage takes some effort, but with the right approach this is an easy enough clear.
  38.  
  39. 5-1 (1 attempt)
  40. You just dodge. "You can't rely on items alone" please, I haven't been relying on items at all!
  41. 5-6 (1 attempt)
  42. Shotgunning safe area number way too many.
  43. 5-7 (1 attempt)
  44. Would you believe me if I said this was the first time I tried the spawngraze safespot, and somehow got away with it? Even without that trick, I don't find this pattern particularly threatening.
  45. 5-8 (1 attempt)
  46. Okay this one is actually pretty hard. Yatsuhashi is NOT playing around in this game and I definitely got a bit lucky to pull this one first try without doll sub. Even still, this is pretty cleanly the best choice for a 4th scene this day, for reasons I'll get into during the backtracks. For now, though, 5-2 is one of the most dangerous patterns in the entire game so that one is definitely out.
  47.  
  48. 6-1 (1 attempt)
  49. Freebie. But it unlocks sub items so rushing it is the only way to go.
  50.  
  51. 1-4 (50 attempts)
  52. I probably spent about 40 tries going for a quick react dodge at the bottom of wave 6 without much success. I was getting frustrated - I'd gotten to this point in under 20 minutes and had already spent 10 minutes on just this Day 1 card - so I tried going for micros on the waves before they split. And to my surprise, this was actually mostly reliable!
  53. 1-5 (3 attempts)
  54. Difficulty to do reliably: not very high. Difficulty to spam attempts until lucking through the one dodge I went for: zero.
  55. 2-4 (4 attempts)
  56. Most of these tries were spent figuring out how to misdirect the lasers effectively. With the right approach and doll sub, easy. One attempt here was entering it instead of 1-4, oops!
  57. 2-5 (8 attempts)
  58. I don't even remember what my strategy for this was, I just remember that it felt a lot more comfortable than I ever remember it being. Not sure if that was from strategy or being better at processing information than a decade ago, but from what little I recall it's mostly the former. The latter - improved reading ability - absolutely helped in a few places, though I don't think it was significant for most of the game.
  59. 3-4 (2 attempts)
  60. Skipping 3-3 for now to keep the progress train rolling. The singular non-winning attempt was from entering this scene on the first pass and leaving to come back with doll later - it's not hard with a triangle strategy for misdirecting Mokou.
  61. 3-6 (1 attempt)
  62. Probably should have done this one first pass as doll sub doesn't really help, and it's super straightforward. I guess that means if I do this again, first pass of Day 3 is 3-1, 3-5, 3-6, 3-7? Or probably 3-2 instead of 3-5. Wait, am I actually considering doing this again?
  63. 4-2 (5 attempts)
  64. LOL this should have only taken one. Failures were: attempt without doll, attempt without doll, forgetting Seiga moves up, and the double KO shown before the clear.
  65. 4-6 (4 attempts)
  66. While I didn't practice before the stream, I was still aware of the existence of many strategies so in some cases, like this one, I already knew the gist of what to do. Back in 2014 the relative alignments of the waves was not known so this actually required reading everything - I... just spammed attempts until I got lucky. Here, I got the hang of it pretty quickly, and should have done this instead of 4-1 on the first pass.
  67.  
  68. I skipped 4-3 to keep momentum going. 4-3 is hard but I could certainly do it, I just wanted to save it for later. The other scenes I saved for later were: 5-2, 6-5, 7-4, 8-6, 9-1, because outside of Day 10 I expected these to be significant problems. Some were, some weren't, and a couple scenes I expected to not be a struggle were, just like 1-4 already. I still stuck to my intended order, though.
  69.  
  70. 5-3 (4 attempts)
  71. It's a safespot three times. The only thing to watch out for is that the safespot is lower than it might appear, because the hitbox of the notes is the body of the note, but the area that shows the background is higher than the safespot - due to viewing it through the staffs! Without doll sub, this requires 5 waves IIRC. Still not crazy hard, but 5-8 is less affected.
  72. 5-4 (1 attempt)
  73. Free with doll sub, painful without.
  74. 5-6 (1 attempt)
  75. It's a buffed DDC pattern again - and Requiem is one of the more infamous. Still, I always found this version pretty manageable, possibly due to the different colored notes and usually macroable rest wall?
  76.  
  77. 6-2 (1 attempt)
  78. Aya just requires strategy, and here the strategy is just keep away.
  79. 6-3 (3 attempts)
  80. Hatate also just requires strategy, and while I generally think this scene is free I still messed it up twice going for greed damage. Honestly this is a good lesson in hindsight: don't be greedy going for damage unless that damage is necessary to kill something a wave faster AND that wave is threatening. Foreshadowing.
  81. 6-4 (1 attempt)
  82. I got a little lucky here between the waterfall spawns and Nitori's movement. Not too bad most of the time.
  83. 6-6 (5 attempts)
  84. I don't know the strat for this even though I know I've worked it out multiple times. This kind of worked but was super scuffed in a few places and I probably should have been hit. Though Nitori was never going to be a big wall, that honor goes to Momiji. Who I skipped.
  85. 6-7 (2 attempts)
  86. I tried dodging low and got a hard wave that I made the wrong snap decision for. Then I dodged high and captured.
  87. 6-8 (2 attempts)
  88. anna_says_hi raided me here so hi anna if you're reading this. I'm kinda glad that my responses were all contained in one scene where not much else is happening. IDK, 6-8 is just kinda there. Aya is super tame in this game, especially compared to past appearances.
  89.  
  90. 7-1 (1 attempt)
  91. As long as you know where Marisa is going to aim and kill it quick enough, this is pretty free.
  92. 7-2 (2 attempts)
  93. How is this impossible, again? This wouldn't be out of place in a mainline game, there's even enough time to process what just happened and figure out what to do during each wave.
  94. 7-3 (3 attempts)
  95. Both deaths were to not spinning waves I should have gone for a spin on, but thought I saw a gap.
  96. 7-5 (3 attempts)
  97. Same comment on 7-1 basically, but without having to speedkill to avoid dodging. I was very bad at misdirecting here.
  98. 7-6 (5 attempts)
  99. I hate this card. Sakuya can move on top of a bullet, then teleport swap you, and you just die with literally nothing you could have done about it. That happened to me during these 5 tries, and since the actual escape move isn't that precise... yeah this one just kinda sucks.
  100. 7-7 (3 attempts)
  101. Poor Youmu, showing up in ISC only to do basically the same thing twice.
  102. 7-8 (1 attempt)
  103. Usually free, sometimes weird. This one was weird and I read it pretty well.
  104.  
  105. Day 7 is honestly probably the easiest day outside of its skipped scene, and even then... well, you'll see.
  106.  
  107. 8-1 (1 attempt)
  108. The last of the simple nonspells for story scenes. This is actually probably the least threatening of the set.
  109. 8-2 (2 attempts)
  110. Reimu rammed me on entry on the first one, and then I nailed the micros. Fun fact! When trying a new keyboard, this is where I go to test it, as something that I can do consistently that also requires substantial precision on both timing and keypress length. You would not believe how many keyboards enforce a 20 millisecond - or 30, or more - minimum press... let me do my 1 frames! I actually play on my laptop keyboard even though some parts have been dying for over a year (the whole SDER area, which has spread to include C) and have a second wireless keyboard for typing.
  111. 8-3 (1 attempt)
  112. Really hard if you stay at the bottom, basically 100% consistent when dodging up high.
  113. 8-4 (2 attempts)
  114. This so easily could have been a wall and I'm glad it wasn't, the difficulty of individual waves is so heavily variable and I feel like I rolled fairly kind here.
  115. 8-5 (1 attempt)
  116. I don't think I've ever done this strategy before and while it doesn't kill fast, it does kill at least!
  117. 8-7 (39 attempts)
  118. Pain. I knew this card might cause me problems, but I didn't expect quite this much trouble. Over the course of this mini-grind I went from just winging restream timings to actually settling into a cycle that should be pretty repeatable. Of course I messed up my cycle in the clear (since entering it is one of the hardest parts) but it was still enough to eventually make it through. My strategy was to make a big restream on the left side after dodging a set of blue knives, then stream to the right side and make a big diagonal move in the same sort of gap. On the return, the trick is to not double back and use that larger run to set up the second restream. This gives a pattern of dodge, restream, dodge, dodge, restream, repeat. Worked well enough and now that I know it, clearing this again should be less painful.
  119.  
  120. 9-2 (10 attempts)
  121. This isn't actually that hard!? Like it's not easy, but given everything that comes before this one really shouldn't take this many tries. I think my eyes were straining a bit from coming straight off of 8-7 (a very fast pattern) that switching to an omnidirectional one like this threw me off.
  122. 9-3 (2 attempts)
  123. Boat. Not sure how this approach worked but I'll take it.
  124. 9-4 (1 attempt)
  125. Free with doll sub, extremely hard without (though I did do it as the first upload for returning to Touhou in general!)
  126. 9-5 (17 attempts)
  127. I'm not sure how consistent this one can really be, it's quite hard to gauge where the explosions are going to end up and when they're going to, in addition to the newer spawns. The top of the screen is always safe from the explosions but then you deal no damage and aren't safe from random spawns. Even in this clear, I'm not sure I would have been safe a second later.
  128. 9-6 (1 attempt)
  129. I remember this one taking more effort than this.
  130. 9-7 (4 attempts)
  131. I've experimented in the past with this headsit/stay centered approach over the more standard dodging approach, though never quite managed to put it together. During the stream, I decided to go for it and felt quite comfortable and captured the card much faster than I had expected to. I don't think it was a fluke either, and this approach definitely allows faster speedkills. Not sure I've seen anyone else do this either?
  132. 9-8 (2 attempts)
  133. Free with doll sub, extremely extremely hard without. Somehow I died after saying it was free which was funny.
  134.  
  135. 10-1 (3 attempts)
  136. One scouting attempt, one failed headsit I almost kept to scout but then reset, clear. Static fast lasers really are just about learning.
  137. 10-2 (56 attempts)
  138. The lasers here are almost unreactable and having to survive 6 whole waves of the round bullets is rough. I had quite a few 5/6 tries here and was about to move on to 10-3 before coming back when I finally broke through. Miko is without a doubt the hardest character to deal with in ISC.
  139. 10-3 (1 attempt)
  140. I kept not dealing damage but also not dying - I feel like my reads were on point here but I probably should have died, and easily could have been walled.
  141. 10-8 (1 attempt)
  142. Like the Suika cards, this one is pretty easy with doll sub and EXTREMELY painful to attempt without it.
  143. 10-10 (3 attempts)
  144. I threw, this one is basically just an extra stage final and shouldn't take more than 2 tries.
  145. 10-9 (51 attempts)
  146. Duck, Duck, Bat is all in the approach, and I spent quite a while refining my approach and trying to work out something reliable. Some early ideas intentionally triggered an explosion but this proved too difficult to execute reliably and didn't give a good continuation after a certain point. Eventually I ended up with this strategy which I think works quite well - though I messed up the ending, I had done enough damage that going off-route no longer mattered.
  147.  
  148. And then there were 10 scenes remaining. I got here in about two and a half hours. I still had 8 hours left. The hardest scenes are just that hard.
  149.  
  150. 8-6 (18 attempts)
  151. In 2014, when I first completed all scenes no items by current community standards, I pause buffered 5 scenes. I only did this after another player mentioned doing so for 4-6 (I believe) and the response was "ooh, another tool in the toolbox", the clear being seen as legitimate, and another high level player mentioning pausing for scoring Double Spoiler 7-1 (reading Shou's lasers). With that in mind, alongside the community stance on pause buffering having changed to be considered cheating, even though I didn't cheat I still had redemption on my mind. I claimed to have done something that I both did and did not do, at this point, and I really wanted to erase that asterisk and the stain on my past reputation.
  152. So I spent a half hour failing at 10-5 before coming here and knocking it out with a pretty crazy clear in a relative handful of attempts. This was now 71/75 all time and 66/75 modern. This card demands full attention and awareness for nearly a minute, and yet it's still one of the easier cards I had left.
  153.  
  154. 4-3 (16 attempts)
  155. So I went back to 10-5 and failed it for another half hour before going to 4-3 for a micrododge break. Once I realized that the side I dodged on did actually matter and it wasn't just where a gap appeared to be, this one fell fast.
  156.  
  157. 10-5 (162 attempts)
  158. So I went back to 10-5 and failed it for yet another half hour before finally breaking through. Before 4-3, I had an attempt where I failed to deal enough damage and had to face a second instance of wave 2, which I tragically died to. Wave 1 is easy when not worrying about damage, wave 2 is basically pure luck, wave 3 is ideally 3 extremely tight micros (I had to deal with 4 here!) and wave 4 is all about quick reads and not choking. I used the yin-yang orb sub even though I've heard it does nothing - switching away from doll cost no waves and at worst made the true hitbox easier to see, and I found that I was surviving wave 2 more often after the switch. Honestly, I think this was the least fun clear of the bunch.
  159.  
  160. 6-5 (55 attempts)
  161. Oh look, the other least fun clear of the bunch because the lasers are unreactable and the variance on the bite sometimes just kills you. At least it's short. Yes, that does mean Momiji's single scene took me longer than the entirety of day 6 and 7 combined, after accounting for...
  162.  
  163. 7-4 (3 attempts)
  164. Back in the day this scene was painful, but that's because I was trying to move into the wrong side after the lasers. After two extremely rough grinds I was happy to make quick work of a scene for once. Still the hardest of its day, but now it's at least manageable and not a total crapshoot.
  165.  
  166. 9-1 (41 attempts)
  167. I went for 5-2 next for a while without success after trying 9-1 and getting absolutely nowhere. I still can't say I understand the strategy here and either got into a groove on the method or got lucky. A clear's a clear I suppose.
  168.  
  169. I then went for 5-2 a while longer and made a few tries at 10-4, and while I nearly cleared 5-2, this was now over 5 hours in, with the last 3 being high intensity. I got one last good try at 5-2 and called it a stream with 71 scenes cleared, leaving only 4 I had previously buffered. I considered streaming the rest but elected to do them on my own time offstream, so that I could play for 20-30 minutes and then take a break.
  170.  
  171. 5-2 (429 attempts)
  172. After a few of those sessions, I finally broke through on 5-2 after several near kills and chokes. I learned so much about how to misdirect the arrows for a restream throughout this process, starting by always trying to move diagonally up (good when Yatsuhashi is centered, less good when she's on a side), later realizing that a flat horizontal restream is better when in her corner, and later realizing that moving up before moving out would spread out the bursts more. By the end of it I realized the key was to be as close as reasonably possible to Yatsuhashi and move perpendicularly to her to create the biggest spread and thus the safest restream. Still an incredibly difficult spellcard and, at this point, is the single spellcard that had taken me the most attempts to capture since coming back (topping the previous winner of Hidden Breezy Cherry Blossom at ~330). Though, 5-2 always felt more in my control than HBCB, and my biggest problem was either misreading from a note, or going for a kill prematurely.
  173.  
  174. 10-6 (20 attempts)
  175. Then I recorded the laser path, overlaid it on itself to see it was a 20 second cycle, plotted what looked like a safe route from the recording, and went for it. I had to tweak the path a bit to have easier to follow visual cues while playing, but before long I had reached 73/75 and the third redeemed scene. I mostly put this off because I thought it would be a huge pain to route, learn, and execute, and while this sort of pattern is never my favorite, none of those steps were painful. A good refresher before the hardest pair.
  176.  
  177. 10-7 (272 attempts)
  178. Given that I'd already made about a dozen tries at 10-4 without making much progress at either winning the damage race or completing the spin, I elected to try 10-7 first. When faced with a super high skill check as one of the final obstacles for a completion, I tend to leave that for last as it being the final thing is itself a motivating factor for me to finish it off. When going for lunatic 1ccs, UFO was in this category until LoLK came out and both games appeared to be high skill checks, that then mentally needed to be the final one for me to go for it. In hindsight this was a mental block, but I was also pretty checked out of the series by this point.
  179. Anyway, 10-7. My first attempts actually went pretty well - I was averaging about 50% success on an individual wave and usually getting in enough damage to kill it in 8 waves, giving an estimated clear at around 256 attempts. While this is obviously a gross oversimplification and sometimes I would just die on wave 1 over and over, over time I started getting more 3 miss runs. And then I broke through to 2 miss, and started farming 2 miss runs. Then I broke through around 230 attempts in and got a 1 miss run. But at this point, I'd only survived 6 waves in a row as a top streak, and only started a single attempt with 6 - so I knew that even though I had a 1 miss, like with HBCB where I had a 1 miss after about 30 tries, something this random could take anywhere from "soon" to "many hours later". Fortunately, I got on this run before too much longer and was able to use the tactics I'd learned from those other attempts to boost my wave by wave success rate to about 60%. Staying still for the curvy laser burst if possible, picking a side based on where the cards would end up being sparsest first, orb heights second, Miko's position third (since Miko being closer means the cards don't spread as much BUT you get more damage from getting back to her faster), and so on. Then, 6 waves deep, I miss the headsit, forcing the attempt to go to a 9th wave.
  180. I survive 8 waves, and "should" capture here, but my damage is too low.
  181. And Miko gives me a free final wave and I reach 74/75.
  182.  
  183. I haven't really talked about my reactions to any clears yet here. After 10-5, I said "it's done!" in a clear tone of relief. After 8-6, it was shock and vindication and excitement and pride all in one epithet. After 1-4, it was a dry response of "okay, back to the game". Most of these clears aren't really significant enough to get that sort of reaction. Even 5-2 was mostly just "I did it!" once and then moving on with my day.
  184. After clearing 10-7, I started hyperventilating from the sheer roller coaster of emotions those final 3 waves were. Hopeless and hopeful at the same time, continuing for practice, the sinking feeling of a choke and wondering how long it would take to get back, the utter shock at how free that extra wave actually was, the nerves as I still barely killed it in time and almost choked anyway because I was borderline shaking, the vindication of clearing one of the biggest obstacles to full redemption, riding on nearly a decade of "what if"... it was a lot. It took nearly 10 minutes for me to calm down enough to continue with my day at that point.
  185.  
  186. But there was still one scene left. As I put it, only the nightmare remains.
  187.  
  188. 10-4 (344 attempts)
  189. The first 60 attempts went absolutely nowhere: I was throwing myself into a brick wall repeatedly, always losing the damage race (usually not even close) and always failing the spin (also usually not even close). But through those attempts, the spin was slowly starting to click. First the transition timings and movement timings at the top, though I was fully reliant on muscle memory. Then how to enter the right side. Then the right side to top transition. And finally, the timings on the right side and I broke through, getting a 2 segment spin (entry to top center, and top right to finish). That was proof of concept for half of the card, but what of the other half? In a complete coincidence, this was the attempt I first won the damage race, and before too much longer I had back to back attempts where I completed the spin and won the damage race. This was now a two segment Fitful Nightmare, instead of 3 or 4. But going from 2 to 1 is always brutal, and I still had no consistency at the damage. So I turned my attention to trying to squeeze out as much as possible. I hit on a 4 phase approach.
  190. Phase 1: headsit (8 options) for about 3 knife spawns, then move down. This probably kills two thirds of attempts, but the resets are fast and the minor gain is worth it.
  191. Phase 2: sit in a sweet spot where Seija's center options and all of one side hit. Stay there until the first big speed change, dodging and returning as necessary. Dodging by only moving down is best for preserving damage.
  192. Phase 3: shotgun directly below Remilia, aiming for quadpokes (hitting with all main shots). Even if this whiffs most of the time, it virtually guarantees hitting with 3 shots and any time hitting with 4 is pure benefit.
  193. Phase 4: at the penultimate speed increase, back off to the bottom of the screen and try to hit with 3 shots by being slightly off center. Hitting with the main shots is most important.
  194. Then it's off to the spin - I found that holding right was a good start, some up taps got the diagonal, about halfway up the wall switch to tapping up, then up left, hold left a bit at the top until roughly top center then quickly tap left, and slower/longer taps of down+left for the final exit. Over time I started being able to time these by sight instead of by feel, and around attempt 220 I had sufficient damage and completed the spin. Remilia forced me to back down during the second damage race - because maximizing damage has to be done twice, not just once! - and while I still had enough, I had to ride the lane of the second spin and died to the penultimate knife that could have. 90 tries later, the exact. same. thing. happened.
  195. 30 tries (about 10 minutes) later, I broke through. No big reaction, just a big feeling of satisfaction.
  196.  
  197. And in the end, redemption.
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement