Monday, September 14, 2015

Double Spoiler adventures

When Shoot the Bullet grinding yielded less and less results because improvements are hard to make, moving on to Double Spoiler seemed natural. At first it was a slow start last year, and with pretty much no decent scores at all, I put the game on hold. Around 7 weeks ago now I restarted the challenge, fueled by a new passion for the game and for conquering the unseen and unknown. I've learned a lot about the scenes and about general playing since then and although I am nowhere near pushing DS as much as I've pushed StB, I'm still content with the work I've poured in for now, and It's time to take yet another break.

Some of the thoughts I have regarding transitioning from StB to DS:

- The most immediate thing to get used to and that makes switching between the games somewhat awkward is the major difference in lens behaviour, manipulation, and the fact that the frame angles towards the boss (so no more oval "corner aiming" like in StB)

-The Two Shot requirement is much tougher in DS. In StB you can have a sizeable portion of your sprite cut off, in DS, you can't even cut off your feet (and if you take it from certain side angles, you can even have your whole sprite fully visible and some excess space, and still not get the two shot).

-The hitboxes are much bigger in DS, and they are circular. Forget everything you knew about hitboxes in earlier games, they're all bigger in DS, sometimes considerably so, some are even bigger than the sprites. This takes much getting used to and it's sad that people are unaware of it.

-DS has Completely random movement. The bosses follow you in StB so the only random part is the vertical movement. This was already a pain when scoring so imagine how completely random movement sucks. It can really turn the game into a nightmare of endless waiting and timeouts. I'm pretty used to it by now, but I still think it's horrible.

-DS is unfair. StB was by and large a pretty fair game where you don't really get a whole lot of stupidity or forced deaths, it was more a matter of grinding for RNG or very very tight execution requirements. DS goes way further because it still has the grinding for RNG, it has even harder execution, and it is way, way more unfair. Tons of scenes have random unavoidable insta-deaths or other more or less obscure problems that may or may not end any run should they occur, it makes it less fun to work on every scene, when I played StB I wanted good scores everywhere, in DS this can quickly drain your motivation to continue.

-DS has a lot more depth to lens manipulation and specific problem solving for optimizing your bonuses in addition to base points. Despite having so much experience with StB, having to learn all these new things really did feel like I was learning an entirely new game, and not just new levels. I like that the sequel took the original and not only changed completely how you play it, but also add tons of depth that the original didn't have! 

-DS is much harder. Although a lot of DS scoring aspects feels annoying in comparison because of the unfairness, by and large increased challenge is a good thing and that's what I like about DS. Don't get me wrong, StB had some incredibly tough scenes to score like 9-1, EX-8 etc, but DS has harder scenes and tons of them, the WRs in many scenes are nowhere near "optimal" even when not accounting for RNG, while in StB that's never the case in any scenes, Almost all the score that's missing from WRs is just due to RNG. This is partly because StB has had a bit higher level competition but mainly because DS is just that much harder.

-Hatate is cute.


Which game do I prefer now? Honestly, sometimes when I play DS, I feel like it really might be the better game, because many scenes really benefit from having the gameplay that it does, and the increased challenge and depth gives more reason to play it longer. Having two characters to switch between is also fun. But in the end, I still have a lot of gripes with DS, because so many scenes just feel so stupid to grind. In the end, I love both games dearly, but they're both flawed games. It really just comes down to accepting those flaws and loving these games for what they are. It feels strange and awkward to play StB now, but until I've put at least half the time into DS that I have into StB, I really can't bring myself to make this choice. Both games are among my absolute favorite games of all time without a doubt.




But I didn't just want to blog about starting (and now, temporarily quitting) DS, I wanted to talk more specifically about what I've accomplished, future goals, and about some of the scenes.

In my 300 (and one) hours of playtime, I've somehow already managed 91.3 million with Aya, and 94 million with Hatate. This exceeds what I expected. I think it is safe to say that my first ~100 hours were pretty much a warm up, and I got better and better the more I played. Replays are always submitted to royalflare if they are first or second place, otherwise, I only submit if I found that the score is decent enough from my current perspective, or even if I have a score that feels decent, if it needs just a tiny push to reach first or second place, I have held back on submitting. I don't expect that anyone besides myself and maybe a few Japanese players really cares about replays for DS anymore, but just in case, my apologies for keeping so many replays to myself.

The number of first place scores (stars) is 27, while the number of second place scores (diamonds) is 46. With 207 categories (reverse spoiler excluded) that means I have yet to reach top 2 in half, nonetheless it feels very satisfying. Of course I consider that non-submitted scores beat mine at times. CLRs most notably, but I made up my mind to only use his replays for reference, rather than considering them seriously. Mizoru deleted their Iku records, t1100 only seems to submit first place scores with very few exceptions (perhaps when he suspects the top score is TAS?) and some people submit TASed replays, for instance I am 100% certain that 9-7 top score is TAS. I think it is safe to assume that t1100, and perhaps some other players, have plenty of my scores beaten with their non-submitted replays. Until I actually know of the existence of these replays it is pointless to think about it too much, so due to that, I have no issue with saying I have the number of first place and second place scores that I do.

In the goal of the future, there are many scenes I want the record in, and for overall scores, 100 million with Hatate seems actually reachable. It may be a stretch, but with enough work, I think I might be able to pull that off.

The following scores of mine I feel are the most accomplished:

2-1 (Aya), 3-5 (Aya), 3-6 (Both), 4-4 (Both), 4-6 (Hatate), 5-7 (Hatate), 6-8 (Both), 7-1 (Aya), 7-6 (Both), 8-3 (Aya), 8-5 (Both), 8-6 (Both), 9-1 (Both), 9-6 (Both), 9-8 (Aya), 10-3 (Aya), 10-7 (Both), 11-8 (Both) and 12-8 (Both). Depending on my mood I might exclude some and include some others but let's not make the list excessively long, I'll just talk briefly about these scenes.

2-1 with Aya: 287,397
Sum of best  60,9xx + 69,193 + 87,447 + 101,641 = 319181. I found a new strategy for the first photo which allowed for a higher score, which was great. Memory is a little hazy but, from what I recall, the timing for dashes, and for stopping was incredibly strict, and you needed the right movement RNG after the dash to avoid losing score. In addition, to safely make the dashes without dying, you need a great reading ability, so dying was the usual on this one. A very hard scene and not particularly "fun", but because I did possess the skill to actually play this one, I do want to aim for 300,000 in the future.

3-5 with Aya: 375,539
My sum of best is incomplete on this, but it is likely only higher by about 10,000 or so. This score was dear to me because I even beat CLRs reference replay somehow, it was very unexpected. One picture (I think the third) was mistimed by a few frames (high 1.4 nice shot) which is terrible, as it removes also boss and angle. The gist of this scene from how I played it was to get very lucky movements within the time limit, and then to luckily have the bullets miss you if Kogasa isn't quite low enough that the bullets won't reach you, and then try to take the photo on the right frame. My run was very lucky, the bullets kept missing me which gave a small risk bonus addition, and the base point luck was high on the last photo. Besides, you need a high luck just to get the right movements within the time limit. Most attempts are timeouts. This score is most likely final.

3-6 Both:
Not a whole lot to say overall, but it's more fun than 2-1 or 3-5. I borrowed the strategy from t1100 except with a misdirection difference. The only difference when using Aya compared to Hatate is to tap focus once to align the frame. The score value fluctuates wildly based on luck so my sum of best is much higher than my high score. In the future I want to improve both scores.

4-4 Both:
You need a high luck with the bullets to avoid dying, and Momiji has to move the right way or the photo gets completely bent, so chances are very low that you can finish this with a high score. Timing is adjusted later and later for each photo, on the last photo, you should aim for a low 1.2 nice shot so 2~3 frames before you miss the bonus entirely is ideal. You can aim for risk shots in this, particularly it can be aimed for easily on the first photo by positioning so that you die the frame after the photo is taken for max risk

5-7 Hatate: 537,907
The best photo is 136,705, multiplied by four you get 546,820. My score is pretty close to that and iirc, was very light on subtle flaws. Perhaps the macro could be a bit higher, but most of the score gains on this scene comes in the form of base value RNG, and mine was high, I remember that much.
This is a pretty tough scene because you need a very high precision to survive it, and luck so you don't get walled by pink bullets inside the lasers. I memorized the initial pixel to go to by using a dot at the bottom of the screen (outside the play area) as a visual marker. Surviving is very strict. after following the laser lane all the way up to  ichirin, the hardest part is timing your tap and its length correctly so you make it into the next laser safespot. This is where I die the most often. If you successfully survived it, aim down and aim for the first frame of the nice shot. Repeat 4x.
It's surprisingly fun but I'm not a huge fan of precision stuff, but I did spend the time grinding out the RNG until it was well above average base value luck. Somehow I had more trouble finishing with Aya, so no RNG grinding as her.

6-8 Both:
This was annoying, but I found an interesting way of collecting the 9 blue balls more consistently with Aya. Anyway, I've learned a lot about this.

7-1:
Both the Aya and Hatate scores are work in progress, but I felt the Aya one was a bit closer to reaching a high level of gameplay, even if the score is still lower than I want. This scene is somewhat hard and very luck based, let's just leave it there, but I want 1 million with Aya if possible, and top score with Hatate would be nice!

7-6 Both:
Excellent luck and smart compositions with both characters yielded me some great scores! I love this scene, it is complex and once again I'd like to keep some details to myself, besides, there's way too much to talk about in this scene to go over right now.

8-3 Aya: 2,987,288
For sure I wanted 3 mil and as a bonus it would be nice to beat my 8-3 rival Andrew but excluding that stuff I think this is one of my better scores. Actually, if it wasn't for Andrew pushing his score in 8-3, my own score would be significantly lower, probably. That really sparked something in me and made me want to understand this one better. I feel that my understanding is still somewhat limited. The simple down-up-around-Orin's right shoulder composition, I understand how to do that and roughly how to adjust depending on fairy/Orin distance, but what I don't understand is how you control where the fairies end up after the defensive shot. I also am not sure how to compose when Orin is further left. Even with a good understanding of 8-3, the execution shouldn't be taken lightly. Timing is key. 1.5 nice shots are very important, even high 1.4s ruin the pictures completely. Surviving when spinning around to herd the fairies to the bottom right can be troublesome. Cool scene, I want higher score in the future, and also with Hatate.

8-5 Both:
Tedious endless waiting while taking pictures and scouting for fairy positions combined with Orin's position. I enjoy this surprisingly much. The luck involved is out of this world. Want higher scores.

8-6 Both:
Hands down one of the most fun scenes in any photo games, but it is very simple. I adopted ckt's extra sun strategy that increases the difficulty tenfold but only gives a small amount of average higher score. The overall score you get is all about luck. The difficulty of using ckt's strategy on the final photo is so insane that I would only do it if I felt I was lagging behind. Kinda the same with the 4th photo, I did the easy strat if I had excellent luck thus far. The difference in score is minimal anyway.

9-1 Both:
I love this scene. This is the height of DS memo-scoring, up until the final phase which has nothing to do with memo anymore. So much depth and an insane difficulty for optimization. I would like to get the record with Aya in the future and a higher Hatate score would also be nice.

9-6 Both:
The troublesome third phase is super luck based and is just endless waiting. This is a very tedious scene with a high degree of difficulty. To make matters worse, you may die on the laser phase with a bad luck because I think it is beyond human ability to read openings in some waves. The third phase is the most troublesome because of luck, but the second phase is actually hard as well, and significantly different depending on character. With hatate on the third phase, I was forced to charge instead of focusing all the time when waiting for the picture, which makes dodging a lot harder. It is necessary to not drop the two shot. The 8th shot is very troublesome with hatate because the bullets have to dodge you rather than the other way around or you will drop boss shot because you cant go close while charge. Mainly though, I have no issues with surviving the barrage, the hard part is intuitively taking the pictures the right way at the right times. Want improvements for the future.

9-8 Aya:
Confusing with regards to the base value, I don't know to what degree is the luck involved, but on the last photo the bullets can guide you into better angles depending on how they form, I think this was key for me to get my unexpected 222k photo in my record. As for the other photos, I memorized the proper timings as an attempt to get the highest base values. There's no nice shot, but I think you get small, visually unnoticeable bullet spawn bonuses. Probably not gonna improve this score.

10-3 Aya:
I wanted to beat CLRs reference replay but the luck didn't lead me once... Anyway, it's a good score for a pretty shitty scene based on endless waiting for the right movement luck.


10-7 Both:
I got the record with Aya, but not with Hatate. Actually, I understood Aya composition very well but Hatate is a little mysterious. Either I don't understand Hatate as well, or the top 2 Hatate scores on Royalflare are just a lot better than their Aya scores. Anyway, it's an interesting scene with a high degree of luck with base value, but you still have to play specifically to raise the score to the optimum. It's a very unfair scene with the stray bullets walling you and with stealth bullets appearing out of the laser when you want to take the photo. I'd say my Aya score is final, but because I don't have the record as Hatate, I want to get that as well.

11-8 Both:
Positionings and timings are super strict. Hard scene. Did not beat CLRs reference replays.

12-8 Both:
Originally I didn't like this at all. It grew on me a bit but I still find it a little annoying. t1100s Hatate score is a complete mystery to me, so for the future my goal is to understand what the heck they're even doing.

Well, that did end up being a lot of text, but hey. I tried to keep the details to a minimum.

There's  high scoring scenes left where my scores are utter shit (not to mention lower scoring ones as well) and there's tons of stuff I want to improve either way, so more DS in the future is a given! I just want to do something else now.

Below are pictures of my current overall score, and of 9-4 since it's probably the main scene I should focus on next if i want a higher overall.




Saturday, August 8, 2015

Double Spoiler full highscore list

Double Spoiler scores as of 2019/01/31

Scene + Character - Score - Best Shot - Shots Taken - Date (incomplete)

1-1 Aya:     262,945 - 74,838 - 4004 [2017/04/19]
1-1 Hatate: 308,330 - 87,164 - 2266 [2017/04/21]
1-2 Aya:     241,792 - 88,243 - 498 [2015/08/20]
1-2 Hatate: 281,573 - 100,433 - 614 [2015/08/19]
1-3 Aya:     157,678 - 86,580 - 636 [2017/10/07]
1-3 Hatate: 176,839 - 90,139 - 2512 [2017/10/06]
1-4 Aya:     185,748 - 63,636 - 1751 [2017/10/09]
1-4 Hatate: 217,324 - 73,440 - 833 [2017/10/09]
1-5 Aya:     190,648 - 70,899 - 675 [2016/06/12]
1-5 Hatate: 202,614 - 76,469 - 476 [2015/07/30]
1-6 Aya:     223,004 - 72,883 - 50
1-6 Hatate: 234,368 - 64,805 - 105

2-1 Aya:     287,397 - 101,641 - 1278 [2015/07/24]
2-1 Hatate: 404,217 - 124,823 - 2270 [2016/03/23]
2-2 Aya:     211,547 - 78,214 - 709
2-2 Hatate: 234,621 - 85,602 - 227
2-3 Aya:     166,464 - 48,860 - 12989 [2018/03/27]
2-3 Hatate: 190,928 - 54,097 - 1928 [2016/04/23]
2-4 Aya:     162,806 - 60,350 - 1417 [2018/03/15]
2-4 Hatate: 182,697 - 63,045 - 935 [2018/09/20]
2-5 Aya:     354,952 - 95,515 - 4399 [2019/01/25]
2-5 Hatate: 418,875 - 112,630 - 2342 [2019/01/20]
2-6 Aya:     210,155 - 58,733 - 618 [2015/08/22]
2-6 Hatate: 243,833 - 69,135 - 606 [2015/08/22]

3-1 Aya:     392,164 - 116,087 - 456 [2017/02/23]
3-1 Hatate: 415,291 - 132,362 - 453
3-2 Aya:     178,003 - 46,??? - 137 [2015/07/31]
3-2 Hatate: 229,003 - 59,486 - 296 [2016/03/27]
3-3 Aya:     306,448 - 84,695 - 1806 [2017/10/19]
3-3 Hatate: 356,812 - 98,134 - 1707 [2017/10/20]
3-4 Aya:     459,959 - 102,446 - 1834 [2015/01/09]
3-4 Hatate: 577,073 - 128,894 - 2326 [2016/09/08]
3-5 Aya:     375,539 - 80,016 - 903 [2015/08/07]
3-5 Hatate: 466,862 - 99,486 - 1946 [2016/06/11]
3-6 Aya:     230,260 - 90,088 - 558 [2015/08/06]
3-6 Hatate: 288,771 - 111,450 - 514 [2015/08/06]
3-7 Aya:     320,761 - 109,255 - 1159 [2016/03/11]
3-7 Hatate: 393,247 - 133,254 - 1215 [2016/03/06]
3-8 Aya:     300,212 - 90,755 - 5804 [2016/06/03]
3-8 Hatate: 320,404 - 93,345 - 959 [2016/06/04]

4-1 Aya:     505,120 - 128,588 - 1985 [2019/01/18]
4-1 Hatate: 553,017 - 142,172 - 609 [2016/09/03]
4-2 Aya:     348,640 - 88,179 - 745 [2015/08/27]
4-2 Hatate: 270,926 - 90,633 - 687 [2015/08/27]
4-3 Aya:     191,926 - 61,874 - 1970 [2016/11/09]
4-3 Hatate: 204,550 - 63,510 - 1965 [2016/06/21]
4-4 Aya:     279,723 - 71,751 - 2819 [2017/02/24]
4-4 Hatate: 333,143 - 84,010 - 4711 [2017/02/03]
4-5 Aya:     596,564 - 153,310 - 1360 [2016/04/29]
4-5 Hatate: 698,138 - 181,855 - 1276 [2016/04/27]
4-6 Aya:     272,978 - 58,413 - 326 [2016/06/13]
4-6 Hatate: 315,347 - 68,946 - 1337 [2016/06/14]
4-7 Aya:     359,516 - 115,298 - 3828 [2018/08/21]
4-7 Hatate: 344,960 - 98,460 - 1032 [2018/08/22]

5-1 Aya:     318,539 - 76,874 - 451 [2016/11/22]
5-1 Hatate: 367,358 - 88,051 - 532 [2016/11/22]
5-2 Aya:     400,126 - 82,743 - 4900 [2017/08/27]
5-2 Hatate: 513,066 - 105,918 - 2815 [2017/08/24]
5-3 Aya:     700,093 - 182,691 - 4616 [2019/01/14]
5-3 Hatate: 689,287 - 182,738 - 416 [2017/04/12]
5-4 Aya:     700,041 - 159,421 - 3226 [2017/03/30]
5-4 Hatate: 873,318 - 193,755 - 2368 [2017/04/07]
5-5 Aya:     420,282 - 84,412 - 4702 [2017/04/17]
5-5 Hatate: 549,799 - 108,815 -2805 [2017/04/30]
5-6 Aya:     487,608 - 121,343 - 1802 [2016/02/16]
5-6 Hatate: 573,679 - 148,909 - 1731 [2017/04/07]
5-7 Aya:     444,312 - 112,243 - 1149 [2017/04/11]
5-7 Hatate: 543,697 - 137,042 - 456 [2017/04/11]
5-8 Aya:     575,732 - 143,933 - 161 [2016/04/14]
5-8 Hatate: 715,230 - 180,689 - 404 [2016/04/14]

6-1 Aya:     603,014 - 149,730 - 231 [2016/09/11]
6-1 Hatate: 626,292 - 153,935 - 753 [2016/09/11]
6-2 Aya:     403,818 - 95,352 - 686 [2016/11/21]
6-2 Hatate: 509,430 - 118,146 - 1243 [2016/11/18]
6-3 Aya:     299,446 - 64,148 - 3389 [2017/10/27]
6-3 Hatate: 389,579 - 84,884 - 2554 [2017/11/01]
6-4 Aya:     898,789 - 324,474 - 5246 [2018/03/11]
6-4 Hatate: 960,526 - 360,493 - 10563 [2016/03/12]
6-5 Aya:     659,973 - 170,233 - 1239 [2016/09/10]
6-5 Hatate: 720,120 - 180,361 - 751 [2018/02/27]
6-6 Aya:     473,143 - 110,665 - 739
6-6 Hatate: 637,458 - 155,087 - 478 [2015/08/31]
6-7 Aya:     726,190 - 150,319 - 2263 [2018/12/23]
6-7 Hatate: 931,872 - 194,806 - 6382 [2018/12/28]
6-8 Aya:     746,201 - 191,599 - 738 [2015/09/02]
6-8 Hatate: 742,915 - 200,212 - 2248 [2015/09/01]

7-1 Aya:     1,010,186 - 188,216 - 1342 [2016/03/01]
7-1 Hatate: 1,224,103 - 230,693 - 2453 [2016/02/29]
7-2 Aya:     475,249 - 83,122 - 866 [2017/01/13]
7-2 Hatate: 521,579 - 92,388 - 241 [2018/05/24]
7-3 Aya:     945,905 - 196,362 - 3093 [2019/01/06]
7-3 Hatate: 1,205,687 - 247,613 - 786 [2019/01/20]
7-4 Aya:     293,819 - 82,494 - 656 [2016/03/01]
7-4 Hatate: 328,628 - 95,038 - 6361 [2018/06/03]
7-5 Aya:     725,405 - 145,846 - 2679 [2019/01/31]
7-5 Hatate: 723,949 - 147,034 - 5521 [2019/01/29]
7-6 Aya:     1,832,273 - 388,845 - 3622 [2017/10/16]
7-6 Hatate: 2,002,607 - 410,623 - 5041 [2017/10/14]
7-7 Aya:     1,028,104 - 208,039 - 632 [2015/09/10]
7-7 Hatate: 1,189,200 - 239,229 - 448 [2015/09/10]

8-1 Aya:     860,894 - 156,171 - 1421 [2019/01/07]
8-1 Hatate: 935,587 - 172,420 - 90 [2015/08/27]
8-2 Aya:     914,050 - 225,814 - 1588 [2016/02/23]
8-2 Hatate: 952,963 - 245,442 - 1629 [2017/07/16]
8-3 Aya:     3,051,703 - 634,695 - 2940 [2017/02/07]
8-3 Hatate: 3,828,363 - 805,247 - 1308 [2015/08/17]
8-4 Aya:     300,661 - 90,611 - 396 [2015/09/12]
8-4 Hatate: 404,189 - 114,338 - 614 [2015/09/12]
8-5 Aya:     2,284,161 - 389,613 - 9971 [2019/01/08]
8-5 Hatate: 2,119,955 - 366,126 - 4712 [2017/02/22]
8-6 Aya:     858,992 - 201,781 - 3094 [2018/12/31]
8-6 Hatate: 1,089,635 - 255,333 - 7112 [2018/12/31]
8-7 Aya:     661,871 - 127,180 - 3281 [2017/09/20]
8-7 Hatate: 768,382 - 149,052 - 3714 [2017/09/08]
8-8 Aya:     652,333 - 135,556 - 664 [2016/11/19]
8-8 Hatate: 801,704 - 168,238 - 1083 [2016/11/19]

9-1 Aya:     1,055,436 - 271,148 - 4610 [2016/02/28]
9-1 Hatate: 902,548 - 205,026 - 2858 [2016/02/27]
9-2 Aya:     1,400,492 - 306,968 - 3996 [2018/02/14]
9-2 Hatate: 1,770,386 - 394,933 - 5990 [2018/02/21]
9-3 Aya:     3,258,460 - 669,771 - 969 [2017/02/06]
9-3 Hatate: 3,766,834 - 786,656 - 2257 [2017/02/06]
9-4 Aya:     1,041,887 - 157,538 - 628 [2016/04/17]
9-4 Hatate: 1,386,033 - 244,725 - 1025 [2016/04/17]
9-5 Aya:     640,267 - 198,821 - 213
9-5 Hatate: 684,293 - 231,572 - 180
9-6 Aya:     2,148,047 - 418,490 - 2359 [2015/08/04]
9-6 Hatate: 2,342,238 - 461,516 - 2321 [2015/08/04]
9-7 Aya:     671,761 - 258,732 - 7674 [2019/01/22]
9-7 Hatate: 844,197 - 317,684 - 4619 [2017/08/16]
9-8 Aya:     690,649 - 222,269 - 1242 [2015/08/09]
9-8 Hatate: 738,470 - 204,192 -  649 [2015/08/10]

10-1 Aya:     1,003,120 - 252,964 - 13771 [2018/09/21]
10-1 Hatate: 1,081,945 - 276,783 - 20970 [2018/10/06]
10-2 Aya:     1,548,738 - 362,480 - 3998 [2018/02/06]
10-2 Hatate: 1,290,535 - 351,561 - 5285 [2018/02/03]
10-3 Aya:     908,622 - 145,941 - 534 [2017/01/17]
10-3 Hatate: 1,011,354 - 172,851 - 2819 [2017/01/18]
10-4 Aya:     655,413 - 148,858 - 288 [2015/02/07]
10-4 Hatate: 628,575 - 183,827 - 511 [2015/02/07]
10-5 Aya:     5,152,386 - 1,068,516 - 10832 [2018/12/18]
10-5 Hatate: 5,110,838 - 1,035,770 - 5171 [2018/12/22]
10-6 Aya:     959,834 - 183,194 - 5754 [2018/01/31]
10-6 Hatate: 1,068,792 - 210,655 - 3249 [2018/01/19]
10-7 Aya:     3,580,900 - 552,532 - 5577 [2018/10/13]
10-7 Hatate: 4,116,249 - 652,915 - 10266 [2018/10/08]
10-8 Aya:     1,298,391 - 278,111 - 1647 [2017/01/22]
10-8 Hatate: 1,613,411 - 341,599 - 1121 [2017/01/22]

11-1 Aya:     1,514,918 - 304,942 - 5985 [2017/05/17]
11-1 Hatate: 1,740,186 - 352,574 - 3738 [2017/05/08]
11-2 Aya:     1,167,845 - 225,311 - 992 [2017/02/22]
11-2 Hatate: 1,136,759 - 210,805 - 1371 [2017/02/22]
11-3 Aya:     190,465 - 58,433 - 470 [2016/03/02]
11-3 Hatate: 229,150 - 68,538 - 410 [2016/03/01]
11-4 Aya:     352,289 - 91,361 - 1517 [2016/04/14]
11-4 Hatate: 380,685 - 100,777 - 916 [2016/04/20]
11-5 Aya:     807,035 - 139,807 - 892 [2016/04/16]
11-5 Hatate: 1,058,111 - 184,273 -1703 [2016/04/16]
11-6 Aya:     1,352,193 - 311,141 - 778
11-6 Hatate: 1,278,249 - 376,939 - 275
11-7 Aya:     414,219 - 88,592 - 1866 [2018/04/24]
11-7 Hatate: 486,786 - 103,990 - 3132 [2018/05/21]
11-8 Aya:     1,938,360 - 511,734 - 2049 [2019/01/07]
11-8 Hatate: 2,048,452 - 563,521 - 2275 [2019/01/11]

12-1 Aya:     2,469,222 - 678,662 - 440 [2016/03/12]
12-1 Hatate: 2,749,564 - 777,353 - 624 [2016/03/12]
12-2 Aya:     848,896 - 236,175 - 3174 [2017/09/27]
12-2 Hatate: 969,694 - 277,035 - 1648 [2017/09/28]
12-3 Aya:     1,776,617 - 390,374 - 3294 [2018/09/19]
12-3 Hatate: 1,827,790 - 414,355 - 2607 [2018/09/19]
12-4 Aya:     1,412,926 - 310,897 - 6580 [2018/04/08]
12-4 Hatate: 1,398,372 - 280,842 - 14190 [2018/04/08]
12-5 Aya:     3,118,859 - 626,638 - 402 [2016/03/14]
12-5 Hatate: 3,659,433 - 602,274 - 1367 [2016/04/07]
12-6 Aya:     1,551,898 - 415,572 - 582 [2016/04/19]
12-6 Hatate: 1,974,466 - 535,760 - 3690 [2016/04/25]
12-7 Aya:     1,812,676 - 328,034 - 3050 [2018/09/11]
12-7 Hatate: 1,987,172 - 364,238 - 4521 [2018/09/17]
12-8 Aya:     3,019,404 - 399,900 - 11877 [2017/07/08]
12-8 Hatate: 3,457,116 - 468,656 - 14843 [2017/07/07]

EX-1 Aya:     509,179 - 139,037 - 1008 [2017/02/12]
EX-1 Hatate: 623,015 - 140,822 - 1014 [2017/02/16]
EX-2 Aya:     896,097 - 282,522 - 504 [2018/04/18]
EX-2 Hatate: 931,772 - 285,245 - 908 [2018/04/20]
EX-3 Aya:     717,656 - 217,999 - 1568 [2018/03/22]
EX-3 Hatate: 825,277 - 242,566 - 2120 [2018/04/18]
EX-4 Aya:     1,643,936 - 482,697 - 173 [2015/02/05]
EX-4 Hatate: 1,828,776 - 525,046 - 141 [2015/02/05]
EX-5 Aya:     932,806 - 210,435 - 749 [2017/02/12]
EX-5 Hatate: 1,242,550- 276,834 - 1860 [2017/02/12]
EX-6 Aya:     1,239,180 - 408,300 - 13472 [2017/11/29]
EX-6 Hatate: 1,405,537 - 468,231 - 17920 [2017/11/23]
EX-7 Aya:     814,623 - 168,272 - 787 [2016/04/23]
EX-7 Hatate: 898,330 - 185,968 - 1136 [2016/04/22]
EX-8 Aya:     1,273,373 - 427,205 - 511 [2017/10/10]
EX-8 Hatate: 1,297,690 - 405,162 - 559 [2017/10/10]
EX-9 Aya:     4,145,294 - 917,422 - 2524 [2017/05/21]
EX-9 Hatate: 4,969,680 - 1,097,540 - 325 [2017/05/21]


SP-1: 152,164 - 152,164 - 56 [2015/08/26]
SP-2: 669,649 - 173,272 - 414
SP-3: 521,095 - 186,199 - 90
SP-4: 932,695 - 173,677 - 596 [2016/05/18]
SP-5: 189,944 - 189,944 - 78 [2015/08/26]
SP-6: 383,853 - 109,721 - 95
SP-7: 429,311 - 292,001 - 130
SP-8: 519,506 - 112,274 - 81
SP-9: 1,568,906 - 277,744 - 4337 [2016/11/17]

Aya score: 99,827,354
Hatate score: 106,065,230
Playtime: 1450:29.29






https://www.mediafire.com/file/9yimy8ird8z301s/2021_may.rar/file

https://www.mediafire.com/file/75r3g2ffcmq44xv/2021_may.zip/file

Saturday, July 18, 2015

StB return and FW things

Last month I made some attempts to reach 100 million in FW Extra. I gave up pretty fast, and went to StB, which I've been playing a good chunk the last 30 days... and I got some pretty satisfying scores, as well as finally reaching 67 million total score. 68 million in 2016? We'll see!

Anyways, the new StB scores are as follows, in chronological order:

6-5: 373,350
10-1: 2,129,740
EX-8: 1,811,200
8-1: 822,140 
8-8: 3,112,980
2-2: 115,440
2-4: 141,130
6-2: 232,760
6-8: 523,890

Thoughts on the runs/future prospects of the scenes:

6-5: I am very happy how this turned out. The luck was fantastic on the last two photos which allowed this score, but I really like how you can choose to bide your time in this scene. The time is obviously not unlimited (in fact, you really aren't given much time to stall) but there's enough time left over that you don't need to feel forced to take the pictures if you can tell that there's not enough pellets for a good photo. Despite that it's still a matter of luck in the long run because the timer is limited and, at least for me, it's not possible to know the outcome of a photo before I take one. With a lot of experience in this scene you can learn to use your intuition and see roughly if a photo will be good or bad.

There is definitely room for more luck, but this should still prove to be incredibly hard to beat. I have no intentions of improving it, but it's a fun scene, so you never know.

Lastly, surviving in this scene is definitely far from trivial, but I suppose I'm pretty good at it. I felt pretty comfortable dodging the barrages although the deaths were somewhat frequent, certainly not unbearably so.

10-1: In my opinion this is the best score I have in this game. My 6-5 and 9-6 scores are comparable, but this is the one that seems hardest to improve right now. For the first 6 photos, you want frame perfect photos. 1 frame early or 1 frame late costs a lot of score. In addition, I figured out great composition on the first 3 photos, but I only used the best composition on the 3rd photo in order to save time since it takes a long time to set up for. Photo 4-6 can be optimized further by getting the optimal movement patterns, but that's not something I want to stall for. The last two photos are RNG, the base value of the photos vary wildly, but it's also very tough to survive and get 2.0/1.9 boss shots and self shots.

In my run I got great luck on the last two photos, as well as 2 or 3 (I can't remember) of frames total mistakes, which is around 20,000 loss or so. My sum of best for this scene is roughly 2,188,000.

I will never try to improve this score.


EX-8: Sloppy run and with a compromise photo as the finish because the last two photos are almost impossible to take optimally... The compromise still aims to take a self shot and a nice shot but drops majorly from the boss shot multiplier.

I want minimum of 2 million out of this scene in the future, but it is so insanely hard... 

8-1: I don't enjoy this scene because of the stalling in the first three photos, but this is a pretty bad score so I'll improve it eventually.

8-8: Around half the time I simply die right after a photo because of the leftover butterfly bullets. Since there's 4 photos, that means that there's 3 such times per run, which means that around 7 out of 8 runs just die. That's pretty brutal, and in addition, the runs take a long time, the scores are all over the place, and it's just weird, subtle optimization to try to set up for good photos. I started grasping it more and more eventually, but it's still a bit hazy. Nonetheless, this was a pretty satisfying score, although I would like a 3.2 some day! 

2-2: It's one of those easy luck-grind scenes. I feel pretty satisfied with the great luck I got, 116+ is just too tight for me, so probably not playing this one again. 

2-4: This is another luck-grind scene but I actually like this one. There's a cool approach in Fisty's run that I started sort-of copying (but with my own, simpler techniques). In the end I got a great run which is only 30 points behind Fisty. I kept playing another day after that but couldn't get close to that luck again, so I moved on. Maybe in the future I will go for the record, but it feels kind of silly to aim for 40 points improvement... and I don't feel strongly about a personal score, my ego just wants the record lol.

6-2: This is an unique scene in how the photos are taken, and in the end I settled on the unfocus - dash - aim up-right technique. I use my mouse cursor to tell where to stop, and both a part of the screen-border background and the mouse cursor to tell which horizontal position to start from. I try to find the right vertical position during the little time to get, then I count to 3 when Youmu slashes, dash up at 2, and aim up-right at 3 and let go. The composition is fairly consistent (Youmu sometimes ends up outside the picture, but that's ok, since you can just retry with no penalty except time. I never really get Youmu too far right/down in the picture, which is important.) but the timing is hard, so missing the frames leads to a lot of restarts. In my run, the first photo was bad, but the last one was really high!

6-8: This scene is a luck-grind but it's more based on skill than 6-4 and 6-6, and I think I just now got to the point where I actually need to be lucky to make improvements. I think this is my favorite Youmu scene. I don't think I can take the record, but perhaps if I ever get a superb final photo we'll see about it... I think that around 540-550k is pretty realistic for me if I work hard at it.

Right now I want to make improvement son 6-4 and 6-6 in addition to 6-8, to give Youmu some love. 6-4 and 6-6 have been stuck for a long time, though. They are major luckgrinds, particularly 6-6 seems nearly impossible to improve at this point, and I'm actually pretty fine with the score staying where it is, but I've been pretty close to the record on 6-6 since summer last year, and I definitely have a chance...

As for other scenes, I have been playing a good chunk of 2-3 as well, another luckgrind but honestly quite tough to get the photos right in at the end. Still, getting the luck is super important, particularly on the second photo where you get a blue shot like 1 in 10 runs, and without that blue shot, it's almost hopeless. A few other scenes have been part of my rotation but they aren't particularly on my radar right now. Maybe next month?

Picture of my current overall score.
 

-

In addition, played a good bit of FW, mainly Extra and some Easy B1.

I improved the Extra score to 98.9 million, and Easy B1 to 35.6 million. Easy C1 was also improved to 36.5 million, although that one is a work in progress. Extra is a work in progress as well but because the competition is at such a high level, it's hard to make improvements... I think that realistically, 102 million might come out, but I'll leave that to someone else... 100 million is enough for me and, possibly the record, but that might never happen...

Easy B1 is good as it is.

Total score should be 1.791b now, so another 9 millions worth of improvements to make. That should be doable. 

I also got 360~ million in PoFV with Reisen, but I think I can get higher than that... 

Wednesday, June 10, 2015

Recent efforts

Updates for May and June.

Highscore list is updated

Normal A1: 57,459,930 > 58,945,400 [+1,485,470]

Normal A2: 68,041,970 > 69.7 > 70.5 > 70.57 > 71.0 > 71.1 > 72,022,590 [+3,980,620]

Normal C2: 57,884,340 > 58,542,010 [+657,670]

Total score: 1,782,005,240 > 1,788,129,000 [+6,123,760]


I will go over the contents in the Normal A2 replay, it being the more notable accomplishment this time.

The C2 replay is also quite good, and A1 is a work in progress, so I will try to raise it. How much I don't know. All the same, Gasoline's A1 replay is absolutely amazing, so I may not try to score higher than, say, 60 million.

On the Normal mode, I am also very interested in improving B1, and want to raise my current score (55,881,900) much higher as well. Aside from this, I am probably going to tackle the Extra stage next, when I feel ready.

Anyways, going over the A2 replay:

Stage 1 before Lily: poor. I get the first 30% area, but afterwards, freeze is messed up and kills extra fairy, removing points from the next freeze.

Lily White: Nonspell 1 is passable, with 3 out of 4 30% area. however, 4th freeze killed lily, so the 5th (small freeze) is excluded.

Nonspell 2 is OK, but I don't understand how to get highest score on the last freeze without losing frozen items.

Second half of stage 1 is fair.

Luna Nonspell 1 is terrible, with the first 6 (smaller) freezes being far below average, and the next 3 (bigger) freezes all below 30% area. However, the second nonspell is OK,  which is pretty important for the run. Two 30% area freezes on Luna Cyclone, this is important to make stage 2 better, but otherwise Luna Cyclone gave me some terrible movement luck which removes score. Overall a bad stage 1 score, but enough PF for stage 2.

End stage 1 with 7,590,570 score, 253%PF, 567% motivation. Keep in mind I do not use PF on Luna Nonspell 2, instead, I use it on stage 2.

Stage 2 pre-Lily Black:  Fine. The PF nets me 309,600 from the freeze, and an additional 133,480 (items, damage) totaling at +443,080. Worthwhile, in my opinion.

Lily Black:  Nonspell 3 luck is what is truly important here, but Nonspell 2 is also rather easy to mess up.  Nonspell 2 was good, but Nonspell 3 luck was just OK.

Stage 2 post-Lily:  The Knife-Ring Fairy 3rd freeze was wrong, so I lost a 30% area. However, I got all the frozen items from the three duos at the end. Besides, I didn't end up needing the extra PF from a better knife ring freeze, so it's not that bad.

Start Sunny with 13,368,600 score, 230%PF, 780%motivation. A decent start.

Nonspell 1: Close to an early defeat, but this is ordinary. Final PF position was weak.

Prism Flash: Good. I actually want to defeat it on the last wave, but doing this with ice without losing the SCB is hard. So this works.

Nonspell 2: It is important to start this with high enough PF to avoid having to use 3 suicides. Thankfully my PF was easily high enough for this. I do a little bit of milking at the end to ensure I have enough leftover PF for a smooth stage 3. PF positions were good.

Diamond Ring: This is the bane of all runs. In order to get awesome scores, you want to get good luck here. At the very least, avoid bad luck. The difference between terrible and awesome luck is something crazy like 2.5 million. Not counting the last freeze which is always lower, I average around 550,000 per freeze. This is good luck.

Blue Deflection: The point here is farming enough PF  and then using PF at the end. Sometimes I would hardly get enough time, so a lot of runs almost died here, or actually did die here. I prefer to start stage 3 with at least  136%PF. I need 200% by the time I use two bombs, and I can easily get 64% on the way. However, in critical situations I can make do with 135 or even 134%PF, but this is very risky. The PF you gain on stage 3 before the 2nd bomb goes as follows: 20%+PF>20%+PF>20%+PF,2-3%+PF>2-3%PF. 2-3 depends on those very tiny freezes you do. If I know I have to get 3%, I will make sure I get that, but it may delay the timing of the bomb, kill a fairy, or some other stupid thing like that, in a worst case scenario.

Anyway, stage 3 is started with 27,756,100 score, 144%PF, 671% motivation. This is a very good start.

Right at the beginning, I make a little mistake with the 2nd freeze, which is smaller. However, other than that, no big complaints before midboss, but notably the very last freeze was really high.

Star Sapphire midboss. In my opinion, this is hardest part of the run. The nonspell is pretty easy generally, but may be hard to optimize. Her spell, on the other hand, ended many runs.

Going over the nonspell, this is a major source of points, and luck is also a big component. The strategy is 4 suicide freeze, 3 bomb freeze (not necessarily in this order) and hopefully enough health left over for a final, smaller (but still 30% area freeze). If Star moves too close to the side, you cant go through two rings like you are supposed to, losing you points. However, there is an upside to a movement pattern like this, which is that if you bomb freeze, the regular freeze will not hurt Star, and so you are guaranteed to have enough leftover health for the final freeze. My movement pattern went something like this. Star generally hung close enough to the middle (but not always in great spots) that I could go through two rings, but at one point she went too close to the side, losing me points. However, because she had taken so much damage, this ended up being OK. I didn't gain back the points I lost compared to good luck, but I did get 20% extra PF, so it's not a terrible trade. So, the luck was good enough, but not amazing.

Moving on to her spell, it was extremely scary. Before second freeze, my point-blanking was not good enough, and I get only about 60% ice power. A mistake like this will often spell doom, but somehow I recovered. I quickly noticed my charge was too low, go back under side for a split second to shoot her, and then end up with around 76% ice power instead, which makes it more feasible to recover. There were still many bullets in the way to try to pointblank for the 3rd freeze, but I made it. Even so, several of the dashes to pointblank Star were close calls. I'm lucky I didn't die on this.

Stage 3 second half: Clean.

3 Fairies (Star leader) start: 41,234,110 score, 118%PF, 605% motivation. This is the highest score I've had at this point. I've had multiple high 40s runs before, but never started with over 41m prior to this. Both PF and Motivation is good for executing my strategies.

Nonspell 1, Spell 1, and Nonspell 2, all go well according to plan.

Ice Dissolver was a little scary, but it went well. The plan is to milk it for a higher PF, so I can execute my strategies. Although some points are dropped from SCB, overall this is worth the extra points I can gain from using an extra PF and suicide.

Nonspell 3 is immediatley a mistake. I pointblank Star and get only 60% ice power. I do a suicide freeze at the start, and without higher ice power, I lose score. Around 100,000 dropped.

Next, strategy is to alternate between these 4 things, not necessarily in order.
Very big Freeze without damaging Sunny. Use this when you respawn from a suicide. Calculate how to freeze depending on all the fairy positions as well as your own, for maximized score without damaging Sunny. Sunny is the major source points. You can get 300,000+ with good movement luck for a freeze like this.
Smaller freeze without damaging Sunny. If you can, pointblank Star Sapphire and you may be able to use a bigger freeze. Otherwise, You can try to nudge Star+Luna with your shot to up your ice power further and you may be able to even reach 30% area freezes.
PF. With my strategies I can use 3PF here, with only 2 suicide. Very important to make sure 2nd PF does not kill Sunny, it will kill her if you damage her with your regular freezes.
Suicide Freeze. Do at least 2 of these, but possibly 3.

Unfortunately I only got to do 1 suicide freeze, because I lost the second life on accident instead of suiciding. This cost some score as well, but I didn't need to spend a 3rd life because my PF% ended up high enough for 3 bombs. Luck was also not so good, so many points were dropped here compared to a great score.

Orion Belt: I had some trouble adjusting to the RNG, but I did manage to keep Luna alive for a long time. Unfortunately, this attack was defeated one wave early. Dropped a little score, but not a terribly huge amount.

Nonspell 4: First freeze is a nice trick to gain 20% PF without actually damaging Star. It segues right into a bomb. I can't complain too much, but this really could have gone way better. On the second to last freeze, the 3rd ring is completely unfrozen! This is horrible, and I tried to adjust, but then the last freeze defeated Star without me wanting. She must have had just barely not enough health to survive that freeze. Lame! But, good enough to not kill the run.

Fairy Overdrive:  Fair. Resource farming was as good as necessary, and I got the sidefairy frozen items. SCB is pretty much always going to be low on this, and at least I did cap it in time. I didn't get Star's items, so that's 60,000 dropped.

Final Spell: This was scarier than it has any right to be, but nothing majorly wrong.



(Ignore #1, it was ultra patch)

A very good run in my opinion, but I can't stress  how important Gasoline's 71.3 million run is. The research Gasoline has done to make this score a reality is amazing, and I simply followed most of their strategies with only a few changes and improvements. FW, with its lack of practice modes and the like, is not an easy game to research, and especially not because of its depth and organic nature. A2 in particular is very very much based on luck. Patience and consistency is huge to have that luck in a well executed run.

I will abstain from writing commentary on the C2 run bit by bit, but I will say that it was also quite well played and is worth watching for those interested in FW. I feel it is a little worse, because I have not spent the proper time to truly polish all the strategies to that level of optimization. plus the luck could be much better. But, it may be hard to beat all the same.


On a sidenote, with a total score at 1.788b, 1.8b is getting closer and closer. Perhaps I can reach it before 2016.


Other games: I did play some Batrider and managed to reach letter score (10 mil) on Special Course with Solo Carpet. However, I did not clear the game... Which is a shame, since a clear would have been something like 10.6 or 10.7 mil. I believe 12 million is possible if you are willing to grind out incredible luck over the course of the run.

Monday, May 4, 2015

April overview

I played Lunatic only in FW. The following scores were updated:

B1: 101,082,530 -> 102,109,460 (+1,026,930)
B2: 100,524,840 -> 100,981,660 -> 101,401,960 -> 102,935,550 (+2,410,710)
C1: 100,164,660 -> 100,950,740 -> 101,185,780 -> 102,247,660 (+2,083,000)

+5,520,640 to the total score.

I think the C1 run is the best score I have gotten so far.

Stage 1: C1 is definitely the easiest stage 1 on Lunatic for optimization, mainly because of the simplistic boss fight. The luck elements just aren't severe in the stage portion or boss fight, and the challenge level is generally low.

That's not to say that it's trivial, the first 30 seconds of the stage in particular is difficult to get past with my patterns.

Pre-midboss, the score variances when things are going "relatively" right aren't too severe, but I still demand above average luck on the very first portion.

The opening spam is done with two freezes, and I restart if I don't have 330,000 after this, I estimate around 40-45% chance. In addition, you may die at times, so maybe 2 out of 5 runs pass the spam. Note that with super luck you can get over 400,000 from the spam, so in eventual future attempts I may up the bar a bit.

The next section with the 4 corner fairies is much tougher. My pattern is to be near the bottom left when the 2nd spawning fairy (in the top left) starts shooting the 4th set of aimed bullets. Exactly as it starts doing this, you sprint through the mayhem towards the bottom right and freeze there. Do this correctly for around 140,000 and without killing the 3rd (top right) fairy.

My pattern is still somewhat lenient, because I keep going if the 3rd fairy dies. It only removes a small portion of your score, so no big deal.

The rest is routine including the midboss. However the largest mistake in this stage 1 was at the end of the midboss, the last freeze was complete shit. 100-150,000~ loss.

Post-midboss has two sections, the first section is easy, but the second is tricky. Let's call this the knife ring section.

My pattern can get just about as much out of it as I can even think of being possible, but it is tough to optimize despite 0 RNG. Your positioning and timing needs to be spot on, before and after every freeze, but FW isn't so simple that that would be all. You have to adjust your positionings and timings according to your ice power and, if any freeze should be slightly different from the ideal pattern, the rest of the pattern also must be altered.

In this run it went well enough except that one freeze didn't follow the exact pattern, resulting in some freaky readjusting strats for the next freeze, quite difficult.

Star Sapphire is by far the simplest stage 1 boss, and I'd say my fight was okay. The nons really aren't worth mentioning, but the first spell is quite tough to optimize since the luck element is relatively high and the danmaku is difficult. My strategy includes a planned bomb for this which is better than capping it if you can get some good milking done, besides, there's not really much you can do with that bomb.

The last spell was bad, but it's actually not a big deal at all, so I can move on safely. A death anywhere on this stage is an instant restart, but thankfully most of the deaths happens at the start, and not on Star.

Stage 2: C1-2 is pretty difficult and the luck element is fairly severe. The pattern is half-assed along the way to the midboss, missing two 30% areas. Lily White is so luck based and difficult to optimize but scoring  wise I have little complaints, decent RNG on the first nonspell, 2 suicide 1 PF is used on the 2nd nonspell, the last freeze on the 2nd non was a bit low. The main issue here is the accidental death at the end of the 3rd nonspell. My suicide patterns are on the strict side so accidental deaths in the wrong place at the wrong time can completely kill a run. In this case the death shoots me up to 300% PF during the chaos section and I've lost a potential suicide later.

Important to keep in mind that with Sunny Milk as the stage boss, PF usage on Lily and the stage portion is less worthwhile. Due to this, spend only 1 PF on the second half of the stage. I reach 2 out of 3 30% area at the end, It's ok, because I still start Sunny Milk with 279% PF and I don't need any more than that. It is not the best idea to bomb the very first wave of the first non, better to wait at least until the 3rd wave and hope Sunny moved back to the middle (and downward) for that wave. If she doesn't, you may want to sit at 300% PF for a short while while waiting, or just fire away the bomb. This is totally just preference. I prefer just firing the bomb before I reach 300%. You need all the PF you can get.

Fatal Flash wasn't the best luck and I botched the final freeze I was supposed to do. The second nonspell was a little wrong, main issue is that the second suicide was timed too late, eating up bullets from the wave after. Cross Diffusion was completely wrong. I was trying to follow a new pattern I had just developed but the very first freeze ended up being messed up so I had to do some crazy dodging as a result. Yellow Deflection is quickly killed off for high SCB and enough PF to bomb the stage 3 ring fairies without any more suicides. In the end, the scoring on Sunny Milk was really just passable, but I still consider the fight good because there were no accidental deaths at all and that is a difficult feat in a scoring run, and I can follow a solid resource route on stage 3 without spending any needless lives.

Stage 3: Opening section is low and the very first bomb is mistimed. I counted 1 extra ring for some reason (You are supposed to bomb on the 6th shot, I bombed on the 7th). At least I adjusted on the second bomb. I didn't hear the first ring, so I counted to 6 but bombed on 7... Stupid.

Luna Child's nonspell was great, it can go higher but it is extremely difficult to optimize. It is just one of those things where the base strategy is pretty simple, but getting the most out of it is just ridiculous. I followed the pattern on Silent Flower successfully, and the luck was good. However, my execution was lacking. The first bomb can be better and one of the last freezes was low, neither of these were due to bad luck. But, still, I consider the midboss a success overall.

The second half of the stage goes well, with a freeze hitting exactly 420,000. I was not high during this run. Anyway, the time based fairies went well, and thankfully because this is C1 and not A2, I don't have to worry about an optimized bomb at the end which can be such a tough bomb to get right. NB patterns are smoother.

I start the Three Fairies with 589 motivation, 167 PF, and 39.58m. This is pretty much exactly enough to execute the kind of boss fight that I want, so overall, if things go well, one accidental death is all that is really allowed over the course of the run (aside from some odd instances in the final boss, where you can maybe "accidentally" die on the 4th non for example)

First nonspell: Goal is to quickly get the frozen items (very difficult) and then get 30% areas from Luna (also difficult). Each second you spend trying to get the items is penalizing. If you take too long you can kiss the run goodbye. Even if you do it quickly, if you can't milk Luna well, kiss the run goodbye. So this part of the run is very dramatic. I missed a frozen item set, and scored not even a single 30% area, so overall it was a poor nonspell and I was very very close to losing the run due to that.

Full Moon Light: Goal here is to be able to use 4 PF, with 1 suicide. No problem if your PF% is high enough at the start, but since my PF% was low, it was a big problem and I was half-expecting the run die right here. You must score the maximum number of regular freezes without killing the side fairies. Then you must get 30% area from the side fairy suicide bullets. Then you must get 30% area on the suicidefreeze and then you can farm up the final bomb with tiny freezes. I had 101% PF when I used my bomb, so it was very close.

Second nonspell went according to plan. Ice Dissolver was excellent. I used to have a 2PF pattern for it but I've since switched to 3PF. Third nonspell was good execution but poor RNG, regrettably. You can't get everything you wish for even in a good run like this. On Triple Light, I do not do a speedkill, but that's not just because speedkilling is harder, but I actually need PF now to score 1 bomb for the 4th non. I milked the 4th non very aggressively without dying, it was crazy, and a single death couldve easily ended the whole run too. I start Fairy Overdrive with less PF than what I would prefer to have (114, I want at least 120). If I had had 1 less life, I would've wanted 145. FO is milked pretty well so I have 219 PF (244 after the suicide) for the final spell, which is really cutting it close. I'd say 220 (245) is the absolute least you should ever have and anything less than that is scary. My motivation was also at 890 (so 790 after the suicide) which is the minimum you should have without losing the maxed clear bonus.

The run held together throughout the boss fight somehow but it was so, so close to not making it, even a single minor freeze difference anywhere in the fight and the run would've ended, and a single accidental death would've ended it too. 102 million on C1 is very satisfying and I never thought I could reach it when I played last year. With some skill improvements, some pattern changes here and there and a new mindset, I somehow managed this. C1 can still go a bit higher, at least 103 million is fairly possible and 104 is certainly not impossible. But, this is the most "finalized" of my Lunatic scores and I will not beat this score anytime soon.


-


Next, let's talk about B1 and B2 briefly.

B1 first.

Stage 1 stage portion was cheap, this is probably the worst aspect of the whole run, but Sunny Milk was a success.

Star Sapphire was very shaky, with a poor first nonspell, and 2 accidental deaths in the fight. Still, the score was actually not so bad overall. A regular freeze on Comet Stream somehow scored 500,000, which is hands down one of the most confusing things I have ever managed to do in this game.

Stage 3 stage portion was excellent, and the boss fight was good overall, but there is room for improvement. In particular, there is a better pattern you can use on Break Canopy, but I consider it so hellish that I didn't want to include it for 102m attempts. If I ever try to improve this run, I will probably start doing the difficult Break Canopy pattern, but I don't expect to ever finish with that pattern...

As for B2, I collected the most major losses (compared to an estimated near-ideal)

Sunny Milk nonspell 1: big freeze -150,000, bomb -200,000

Lily nonspell 3 bomb: -400,000 (probably less with my RNG) Knife ring fairies: -200,000 (brain freeze)

Star Storm: -500,000 (dead) Great Triangle: -400,000 (dead & subpar freeze pattern) Nonspell 4: -400,000

Fairy Overdrive: -250,000 (frozen item loss, low SCB) Big losses totals: 2,500,000

No extra PF could be enabled by avoiding accidental deaths, nor did I miss any meaningful suicides.

The run was pretty uneven but the amazing luck on Sunny Milk and solid milking on Ice Dissolver among other successes really makes it a good run overall even with how horrible some of the mistakes were.
-


I spent a lot of time playing Etrian Odyssey 3 this month, but sort of lost my motivation in the 5th stratum. I want to beat the game so hopefully I'll get back to it. My party is Hoplite/Ninja, Princess/Ninja, Arbalist/Gladiator, Zodiac/Arbalist, Ninja/Zodiac.

I still play Money Puzzle Exchanger and I've finally started to crack a million points sometimes. My best score now is 1.4 million.

I've played some Batrider and upped my Normal Course score to C.9 (12.9) million. In the advanced course I've done mostly Carpet solo runs and also a few Garegga/Carpet runs, however, nothing but Game Overs so far. If I can finish a good Carpet run I should be getting at least a H or I (17-18 million). Teams are iffy for me, I prefer playing Carpet until I have a better run at least.


-


For future FW, I want to improve the Lunatic A2 to at least 117 million. After that, I am unsure. Either I will battle with the Extra stage, or I will start working on Normal, I think.

Tuesday, March 31, 2015

March overview

The score posts for Shoot the Bullet (link) and Fairy Wars (link) have been updated.

The following scores were produced for Fairy Wars:

Normal B1 - 55,881,900
Hard A2 - 87,425,610
Hard B2 - 76,905,060
Hard B1 - 74,603,130
Hard C1 - 75,333,510
Normal C1 - 59,624,220
Lunatic A2 - 116,199,330
Lunatic A1 - 103,423,130
Hard A1 - 74,349,760
Lunatic C2 - 105,483,820
Lunatic B1 - 101,082,530

The Lunatic scores are the most satisfying.

Sum of all categories (all routes, all difficulties, and the extra stage) is currently 1,776,484,600. Slowly inching closer to 1,800,000.

Abstain from blogging about any particular play because there were just too many. Although I am still playing Lunatic B1, so I may dedicate a post to that sometime. Like if I reach the 102 million score.

There's still many categories that feel obscure, however. My understanding of Extra and many others is limited. I want to do the absolute best I can in every category. Every single one. It may not seem like it, since I play mainly Lunatic, but that's how it is. The journey won't end until my motivation runs out (ha ha)

While I am working on the Fairy Wars for the most part (at a modest pace! don't want to burn out on my favorite game) I occasionally play other games, and below are some other runs I recall having done.



In Shoot the Bullet:

9-4 - 723,580
6-5 - 361,840
9-7 - 1,054,770

Reasonable score with bonkers luck feels like around 730k for 9-4 and 375k for 6-5. 9-4 crazy score is deliberately put as a little lower due to a higher difficulty of finishing. Both are highly reliant on luck after certain thresholds (around 715+ and 350+)

I still want to push the 6-5 score higher, until I feel that the run is so lucky that it is almost impossible to beat. Play occasionally and hope for a miracle.

9-7 is not well understood at all, as such, my score isn't very high. I may try to understand it better in the future. I don't think there's a luck factor in 9-7.



In Ten Desires:

I achieved 2MNBNT on 3 separate occasions.

All 3 runs died in different spots.

Because the game is mostly very boring, I feel it is a waste of time and effort to play much more NBNT. Originally I wanted LNNN, but I am already sick to death of TD, so probably no more attempts.

There is still the will to achieve LNN(N) in a mainseries game. EoSD is my favorite mainseries game for LNB, but achieving LNN in EoSD feels like a pipe dream at the moment, with how terrible my consistency is for the books. And I really can't feel satisfied with a 1MNB or something. As far as NB is concerned, I can only be satisfied with NMNB at this point. Because of this I don't play EoSD LNB at all since the PB sits comfortably at 3MNB and improving it to 2M or 1M wouldn't be satisfying at all.



In Metal Black:

Score of 582,160. It is low because I was playing and learning very little, and haven't even properly memorized the stage layouts.

Although impressed with the aesthetic and soundtrack, there is no will to improve the score because scoring in Metal Black is rather boring.

Maybe want to try another Taito Hori game for score in the future, such as Darius Gaiden. We'll see.

Overall: my most fruitful month so far for Fairy Wars, it seems I am still improving and understanding the game more and more. I suspect there's many more FW updates to come!

Monday, March 2, 2015

#2: B1 Normal, 6/6 LNN

B1 Normal score was pushed to 55,8 million points.



It was a bit rushed and isn't one of my better plays, but I'll still go through the run  to give a rough idea of the contents.

I don't plan on improving the score for now, but I want to improve it significantly in the future.

Stage 1: Pre-Daiyousei is routine, but the entire stage portion, including Dai's third nonspell, was very good. I got every 30% area, and start Sunny Milk with 289% PF.

Sunny Milk: Execution a bit shaky on the second nonspell, but the luck was good during the fight. Overall a great stage 1.

Stage 2: Start with 7,31m, 363 motivation, 165 PF. I somehow start stage 2 with a higher score than the B2 run, despite having 2 less suicides and 1 less bomb! Unbelievable, it was truly the best Normal B1-1 I've had in my life.

Stage 2 is much worse. The entire first half is full of freezes that are below my aim.
Just to get this out of the way first, Lily White is the most difficult midboss in terms of balancing damage distribution with score gain. Greed is of course important to reach the highest scores, and greed has caused the 1st and the 2nd nonspells to have been defeated early so many times. On the 1st nonspell, there is a strong luck element with damage, whereas in the second nonspell, you're forced to get lower freezes the lower on the screen Lily goes.

Lily White nonetheless went quite well. 2 suicides and 2 bombs are distributed, and my PF is far above the minimum.

Star Sapphire: I find that Star is the make or break it part of this run, particularly the second nonspell. Let's fast forward to that. Star's second nonspell is intensely luck based and difficult to optimize, some of my freezes being only ever so slightly off resulted in very high score loss. However, I did get fair RNG.

Stage 3: Start with 24,94m, 556 motivation, 123 PF.

I think clock started with 25,6m and Star's second nonspell is the cause for this disparity. I do use less suicide on it, since reducing the damage dealt isn't necessary, so I have more motivation.

There is little to say about most of the stage. It went well enough, up until the Knife Ring section, which I botched pretty hard. While it could have been possible for me to save a life here and still reach 100% PF for the end of stage spam, when I completely missed a 30% area I went with the suicide plan instead, which was more than fine considering my motivation was so high.

3 Fairies: Start with 30,94m, 710 motivation, 11 PF.

With these resources, I can distribute a non-essential suicide, that is nonetheless an useful suicide for reaching the maximum number of bombs. This suicide is non-essential because reducing the damage dealt on any of the attacks in the first of half of the fight is unnecessary, but I need to spend the life because otherwise I'll get to 1000% motivation.

First Nonspell: Smooth, I get the items quickly and spend most of it gathering that delicious Luna milk. Life is spent here.

Bright Night: I milk this card. It is highly based on luck, since the longer Star and Sunny are kept alive, the better the milk. At the end of the milk, I spend a bomb.

Ice Dissolver: life is lost on accident here, rendering the previous suicide quite useless.

3rd nonspell: I use my "keep Sunny alive" strategy but I don't manage to get the big freeze at the end so she is kept alive for longer than planned . That is not a run ender, and I gather her items from Break Canopy instead.

4th nonspell: I want to milk this efficiently but I am completely unable to. This is where the run "fell apart". I kept dying on accident, but somehow the route was still intact!

Last mistake is on Fairy Overdrive, where I only get 1 frozen item set.



-



LNN was achieved on A1, B1, B2 and C2.



Because A2 and C1 were already achieved, that means all routes were completed. For non-weebs reading, LNN means clearing without dying or bombing. I am not the first to achieve this feat, EOH had already done all the routes, but my replays (being rather different from his) are still worth the watch for anyone interested in FW survival play.

Here is how I would rank the difficulty of the routes for LNN: B1 > C1 >>> C2 = B2 >> A1 > A2

It was an enjoyable challenge. However, despite being enjoyable, it lacks contents in the challenge. LNN can be achieved fairly effortlessly if you already have a high score, and how can you "improve" an LNN, other than going for a higher score (redundant) or add more challenges (arbitrary) ?

It is still satisfying accomplishment.



86,3 million was achieved in A2 Hard.



This was a highly shaky run, where I do almost everything as well as my knowledge allows me to. There are very few "conscious" mistakes. The score loss is almost entirely lack of optimization knowledge. The score is not a world record, but the aim is to at least overtake the WR, and maybe push a bit further. 88 million may be reasonable, but it does sound a bit high.



Lastly, 868,000 was achieved in Money Idol Exchanger's solo mode. I play this pretty regularly, for little chunks at a time. The goal is a million.