shmups.system11.org

Easy - A2
chum - 41,375,180 - C - 0.1% Replay

Easy - B1
chum - 34,634,720 - C - 0.2% Replay

Easy - C1
chum - 36,343,090 - C - 0.2% Replay

Easy - C2
chum - 38,310,550 - C - 0.2% Replay

B1 and C1 are records atm

A2 run was a little harder to get that high though because dat RNG. I just never quite seemed to get good RNG...

C2 run is balls but whatever.

moving on to Normal now

Normal - A1
chum - 57,459,930 - C - 0.4% Replay

Normal - A2
chum - 68,041,970 - C - 0.3% Replay

Normal - B1
chum - 53,643,290 - C - 0.4% Replay

e:

Extra
chum - 95,123,770 - C - 0.3% Replay

Heartbreaking.

Extra
chum - 97,106,580 - C - 0.2% Replay

Heartwarming.

Normal - C2
chum - 57,884,340 - C - 0.4% Replay

Normal - B2
chum - 56,913,110 - C - 0.3%

C2: it is new record, but the contents are not entirely satisfying. it was fun to figure out so many working ideas for the category, essentially reworking and reshaping the entire run. there are many intersections and complicated problems here, and some brutal demands, if you want optimized run. difficult category and my run doesn't do it justice at all, it does not show off the potential, by often forcibly taking that worse path during the many score potential intersections. nothing that endless resets and persistence can't fix, but I'm in the phase where I quickly move on to new category, so let's leave this be for now.

B2: it was just run for practice. some big points were lost on the way. my estimate is that you can get 60m on this, which would be a huge improvement on the current record. the record has issues, and I have fresh ideas.

I want to route Hard C2 now. at a glance I am unsure if I can bring anything (particularly noticeable) new to the table. I will go for Normal B1 and B2 records later.

the long term goal is all category combined score = 1,75 billion (it means, beat a lot of records and get at least fairly close to all the ones I don't beat)

Hard - C2
chum - 76,242,570 - C - 0.3% Replay

new record again but so many problems that ill keep pushing it

-

Hard - C2
chum - 77,488,540 - C - 0.4% Replay

third time breaking the record for this category and I'm glad I didn't just accept the first or second runs, because this one was so much better and I am actually rather content with it. 77m was the target, but I didn't expect to get this much more, especially not on a run with such an average Luna Child. Sunshine Needle was a thing of beauty, luck and execution all ideal. I could realistically push the score to 78 million by either getting that luck there again (without choking! choke and the luck just goes to waste, this was in fact the first time that I got lucky and milked it all the way through) or instead getting luckier on Luna Child Wall Break spell. I can't even say that I was unlucky on it, I think I handled it wrong that time and that I could've had better freezes. It's hard to tell how to freeze it on this mode, so I don't know if I was unlucky or not. then there's the other, finer problems of the run. If someone really was to grind for the stars to align on all the important spots (of course, two spots in particular make or break the quality of the run) while also playing flawlessly throughout, 80m+ is not impossible. Then again, you can't expect that kind of dedication to a fringe category in an already fringe game.

regardless, a good run. nearly 2 weeks before getting this run I was unsure if I could bring anything new to the table, and for the first week I was in despair, because I really couldn't think of much to do, and worse, I couldn't execute the already existing ideas as well as the previous record. I'm glad I could finally put my brain to use in the end, and even happier that I exceeded the 77m target.

What a lovely game this is, eh? Perfection.

Lunatic - C2
chum -103,158,390 - C - 0.5% Replay

smashed the old record thanks mainly to my improved Moon Stillness strategy. my stage 3 stage portion patterns and my superior stage 1 freezes + extra bomb helps too. Luck was fair during Sunny Milk as well, she stayed up during the beginning of Sunshine Needle allowing me to milk it sufficiently. The run was almost too good to be true (from my point of view then; I know now however that this is the type of run I could definitely do again, and even a bit better) until the Three Fairies happened. I was nervous and still not as familiar as I'd like to be. Lunatic doesn't change a whole lot compared to Hard, and I did play Lunatic a bunch previously, just not enough recently. I think I was just playing for 2 hours practicing my patterns and doing some light experimentation before this unexpected run occurred, I just didn't get enough practice in before getting this run, and... what am I saying. I was more than enough prepared, I simply played like shit. I knew what to do. I failed to do it. I didn't just fail execution but I messed up my resource route, too. I even chickened out on the last nonspell and went with a single bomb pattern in order to leave me wiggle room in case I'd die on Fairy Overdrive. I don't blame myself for that detail, since I did want the WR, and I'd rather get a low 103 than risk losing the entire run for a high 103 at that time. It is a shame, but I feel good just having learned enough to say that 105m is a possible target. I'd be more than thrilled with merely 104, though, since even a half-assed-but-not-quite-shit Three Fairies would've been enough for that. secret technique: use 2 suicide on Sunburst. didn't get to do it this run. I figured out that pattern improvement for Normal mode, but I am dead certain that it should be applied on Lunatic as well under the right circumstances. you dont want to reach 300% PF before starting the last nonspell after all.

oh, but who am I kidding. I'm happy just to have improved on the old 101 run. if competition arrives for this I'd expect them to blow me away and make a total fool out of me. I'd expect it and I'd love it. this game deserves so much better treatment. did I say 105m is a possible target? yes, but that is speaking personally. a superplayer could get over 106m. I feel that C2 is pretty much confirmed as the second highest scoring route on Lunatic because I highly doubt A1 could go as high anymore. can't say for certain, though.

Lunatic - A2
chum -114,809,670 - C - 0.5% Replay

mega kek

Lunatic - A2
chum - 115,086,390 - C - 0.6% Replay

overall wr

Lunatic - B1
chum - 97,827,900 - C - 0.2% Replay

L B1 is an amazing category and i feel bad for putting it down before.

i produced this run in order to help a friend but i'll get deeper into it another day. I think that CLR's high score is very scary, but i still want to try to beat it. 101m is reasonable score but for me just cracking 100m is enough.

Lunatic - B1
chum - 100,248,630 - C - 0.5% Replay

Jackpot! 100m broken and WR.

Hey, it's been a while since the last time! Yeah, I did these things some time ago earlier this month:

Lunatic - A2
PurpleTheGuy - 73,577,660 - C - 0.430% - Replay - Video

Lunatic - C1
PurpleTheGuy - 37,442,180 - C - 0.550% - Replay - Video

I also did this fun little thing, where I did a No-Bomb, No-Ice 1cc of GFW A1 Easy mode. It's not at all a scoring run, but eh, I'll submit it here anyway, since this was my 1st ever clear of GFW A1 Easy mode, at any rate:

Easy - A1
PurpleTheGuy - 8,173,590 - C - 0.190% - Replay - Video

Well, that's some of the things I did this month! I guess, now for the next thing...

Easy - A1
chum - 35,412,820 - C

Easy - B2
chum - 39,231,480 - C

Easy - C2
chum - 40,198,900 - C

Normal - B1
chum - 55,881,900 - C

Normal - B2
chum - 60,305,550 - C

Normal - C1
chum - 59,624,220 - C

Hard - A1
chum - 74,349,760 - C

Hard - A2
chum - 87,425,610 - C

Hard - B1
chum - 74,603,130 - C

Hard - B2
chum - 76,905,060 - C

Hard - C1
chum - 75,333,510 - C

Lunatic - A1
chum - 103,423,130 - C

Lunatic - A2
chum - 116,199,330 - C

Lunatic - B1
chum - 102,109,460 - C

Lunatic - B2
chum - 102,935,550 - C

Lunatic - C1
chum - 102,247,660 - C

Lunatic - C2
chum - 105,483,820 - C


All the replays except Hard A1 were submitted to RF so they're all downloadable there, and none of the slowrates exceeds 1%. Most were also streamed.

In addition to accomplishing these scores, I've NMNBed every route on Lunatic, which while not nearly as impressive, might still be interesting. It was definitely fun to play for survival too. The boss fights go by so fast in survival.

(Lunatic) Managed to make a lot of progress developing the patterns on the B and C routes, but CLRs A route patterns were already well developed. His B and C runs were just missing more details here and there that I figured out how to do better. I still want to improve my A2 score by about a million, but A1 annoys me too much, and B1 being the most difficult and fun category actually does have a lot of room for improvement, so the long term goal would be a 2 mil improvement for that. Meanwhile, my C1 score is damn near optimal. I don't see much more than about a 1 or 2 million improvement being very possible without some radical pattern changes. It still has its flaws, but if there's one score I've done that I think will stand the test of time, it's that one.

Other categories have been largely rushed. Normal B2 and Easy B2 took some intense routing and a bit of grinding to bring that high, but still not too severe, and they are still worse quality than my Lunatic runs. I also only beat mizoru's Easy B2 run by like 100,000, I had really underestimated that run at first.

I am not the only person who plays this game, but the other serious players have almost all been focusing on different categories. there's still 5 records I don't have, held by 4 separate players, and at least 4 of those records would be damn hard for me to beat. As in I might never beat them.

This is still my favorite game of all time and I will keep praising it until more people notice its brilliance. I also do have the audacity to feel somewhat proud of my scores. This game isn't actually easy or irrelevant.

Gonna watch some of these replays later. Astounding achievements!

Which one would you say is the more interesting?

Well if you want to watch the A2 replay ARF did an annotated side-by-side comparison video with the previous record here. It's quite a watch.

thanks! great job with that gunbird clear too, by the way. and here i thought that the cloud girl was practically mandatory.

you can't really go wrong with any of the lunatic runs, but I think the C1 is the best one, and B2 probably the most interesting (and for me, a bit shocking) given how so many things went badly but so many things also went so well that the score still came out that high. B1 is the toughest route to score, so unsurprisingly it was a bit rough around the edges, and there's a pretty big improvement possibility in a more efficient milking strategy on one of the final spells (nightmarisly hard, unfortunately), but still quite good if you ignore the low stage 1 pre-boss score.

A1, A2 and are a little further behind in my opinion, particularly A1, which consists mostly of borrowed strategies anyway. A2 requires the most luck out of all the routes, so I have to be patient when I play that, and I don't have a run I'm really happy with yet. my current run is lackluster in several respects. C2 is like "in between".

the resource system of course makes it tough to hold runs together, but it can also add interesting diversions that have huge long-term effects on the runs. I usually try to route in such a way that I would spend all of my lives if needed, but because of the dynamic nature of the game, normally I get to die once or twice on accident (which of course have different effects depending on where it happens)

Thanks a lot for the link. I was afraid I wouldn't have understood much from the replays, but the annotations and the side by side comparison in the video helped tremendously in that regard.

And thanks chum for the in-depth response. There's definitely a lot of depth to this game that isn't immediately obvious. I can see what you mean about the game displaying an high level of dynamicity, with small error having potentially far reaching consequences all over the course of a run.

I really like that in this game grazing doesn't give points, but instead charges your freeze power; something you use to great effect all throughout your runs. The precisely timed suicides to cancel the damage from the freeze, I don't honestly know how hard they are to execute, but they blew my mind.

I actually had no idea bombs and suicides were so important to scoring. It's a very Garegga-y game in that respect. There's also a lot more dodging than I expected with all the flame bullets flying around, at least on Lunatic.

I've had some interest in the game in the past, but in the end I never took the time to explore it in depth. I'm really glad that some one did.

I found your runs very inspiring and I'd like to renew my congratulations for your astonishing results.

Well done!

EDIT: Obligatory question: What's the biggest freeze you've ever got? I think I saw a couple 92% in there, is that roughly the highest possible?

suicides timings are pretty lenient unless you need to reduce an extra pixel or two of damage dealt, haha. bombs are a bigger issue actually, often times just a few frames off on the bomb can be pretty bad. damage distribution difficulty trumps both easily, though, and optimizing your freezes, particularly in combination with the health bars/time given, is definitely the main difficulty

I always thought this game was a lot like the Raizing stuff in regard to the resource system. if you look at Garegga, you need to score to get lives, and you need the lives to get score, and in addition, you can hoard bomb fragments. in FW, you need to score to get bombs in particular, but also optimizes your lives, and you need the lives to score as well. I like the FW resource system a bit better because the bombs are so dynamic with the percentages, but the similarities are there. From my experience Batrider is even more similar because you need to meet score quotas to keep your team route in check. That is exactly like meeting perfect freeze quotas in FW.

B1 is a very dodge-heavy route. I mean, on Lunatic FW is really manic with many fast bullets and heavy RNG in almost every boss bullet pattern. so it's a more dodge heavy game than many other danmaku games imo. of course the manicness is toned down pretty significantly if you lower the difficulty setting

The highest freezes are probably between 960-975,000 (98%~) either on the Luna Child version of the final spell, or on the first spell on the C2-2 stage.

Normal - C2
chum - 58,542,010 - C

there were some losses (most notably 300,000~ dropped from a slightly mistimed bomb) but overall this is satisfying. 59,000,000 is doable, but I'm uncertain if 60,000,000 is. I don't see my strategies being enough.

Normal - A1
chum - 58,945,400 - C

Nothing too noteworthy here, but I just want to leave these. I'll probably get back to these at a later time. Started putting time in to figure out FW.

Easy - A1
Prinny - 20,912,440 - C - 0.1% - Replay

Easy - A2
Prinny - 19,218,410 - C - 0.1% - Replay

Easy - B1
Prinny - 19,313,420 - C - 0.0% - Replay

Easy - B2
Prinny - 21,206,320 - C - 0.1% - Replay

Easy - C1
Prinny - 20,338,780 - C - 0.1% - Replay

Easy - C2
Prinny - 23,211,040 - C - 0.1% - Replay

Normal - A1
Prinny - 33,958,840 - C - 0.2% - Replay

Normal - A2
Prinny - 37,210,430 - C - 0.2% - Replay

The easy runs were pretty much done with minimal research. Pretty much had to rely on whatever information I remembered from when FW was covered by STGWeekly and from when I understood when I tried normal runs earlier in the year.

I did look at Chum's normal runs for the A routes. Used his runs to get a general idea on how to freeze stuff and timing bombs. Still took me some time to figure out how I should approach the stages. I think A2 didn't take as much time as A1 did.

Started working on B1, but I don't see myself getting anything done anytime soon.

Nicely done. Cool to see a scoring approach on first 1cc's.

Easy - B1
chum - 35,685,310 - C - 0.3%

Easy - C1
chum - 36,504,750 - C - 0.3%

Normal - A2
chum - 72,022,590 - C - 0.4%

Extra
chum - 98,920,590 - C - 0.3%

Easy B1: Well optimized. I could see a low 36 being possible, particularly with a specific pattern change, amazing luck, amazing execution, maybe even a mid 36 (doubtful). I am quite certain 37 is absolutely impossible.

Easy C1: Work in progress. 37 would be lovely here, but a high 36 might be satisfactory.

Normal A2: Completed. Thanks to Gasoline's essay, I could better understand some techniques, and because I had some improvements figured out on my own, 72 million somehow came out... It was a stunning achievement and totally made my day. 73 million is possible but I don't want it.

Extra: Work in progress. 100 million would be absolutely amazing and my new favorite achievement in this game if I get there. I'm unsure about beating mizoru or DATA's scores (100.7 and 100.5m). Of course I want to compete and get the record, but the luck and execution required is something else, something a bit beyond what I've attempted in the past.

Normal - B1
chum - 56,655,040 - C - 0.2%

Hard - A1
chum - 76,931,940 - C - 0.5%

Hard - B1
chum - 76,667,300 - C - 0.4%

Lunatic - B1
chum - 103,524,990 - C - 0.5%

Normal B1: Definitely unfinished. Not as good score as the other difficulties.

Hard A1: Not one of my more remarkable runs, but I find it notable since it beat the previous record holder by exactly 320 points. Since A1 is my least favorite route, it's also the one I'm least interested in optimizing, so I'm fine with just having the record for now. mid 78 is very realistic, even 79+ is possible, and I might try to bring this up another million or more in the future.

Hard B1: Well optimized. Stage 1 is weaker, mainly due to RNG. 77 is still realistic but I'm quite proud of this score all the same.

Lunatic B1: A little less optimized than Hard mode, a suboptimal pattern on Star 2nd nonspell, and a real heartbreak with shit luck in the 3 fairy 2nd nonspell that brought down the score considerably. On the bright side, stage 1 was the highest of all time, insane luck and managed to not let that luck go to waste. Around 105 million seems realistic enough and 106, even 107 should be humanly possible as long as the luck doesn't let up. It'd be complete insanity at that point...

Hard - A2
chum - 89,236,060 - C - 0.7%

Lunatic - A2
chum - 116,786,470 - C - 0.7%

It's a happy update because I reached the goal of 1.8 billion score in all categories combined. I've been improving little by little by a half million to a million here and there, now and then. Finally the overall score has reached its final landmark since 1.9 billion is not actually possible. I think aiming for 1.81 doesn't have quite the same ring to it, but that might take another half year or so of playing non-obsessively, so I may as well extend the goal to that.



Easy mode: A1 should get to 36, C1 to 37.5 or so (maybe even 38 is possible). A2 I would like to extend that all the way to at least 43, but it's the least favorite category, so I don't play it. C2 is also a bit slight. B1 is very good, and B2 should have 40+ potential but is relatively difficult.

Normal mode: A1 should be improved by 2-3 millions, C1 by up to 2 millions. B1 and B2 I think should be improved a little as well.

Hard mode: A1 I would like to extend by a million. B2, C1, and C2 should all have significantly higher potential. I think, particularly B2, but I'm not sure. So 2+ millions per category for them.

Lunatic mode: This is the mode I played the most, so I think it makes sense that these runs should also be the hardest to beat in the long run. A1 should get to about 105 before I'd consider it done. I'm not very fond of it, so I don't want to aim higher than that. A2 to 118, maybe even 119, but like A1, it can be a bit annoying because of the "all or nothing" luck aspect. The others I would like to think I can improve by at least a million each. B1 is special for me, so I am most willing to play that one forever and never actually stop no matter how high the score gets. So let's get crazy and say 107 million target!!!

Extra: 100 million would be amazing, but let's say +2 million to beat the WR.

Taking all these hypothetical improvements into consideration, I could just about end up with 1.83b total score, so ~1.2m improvement on average per category. In fact, since I'm currently not considering playing some categories again, if I also hypothetically considered improving those in the future (which I probably will eventually), maybe even 1.84b could become a goal in the distant future.

Anyway, about the newest runs...

Hard A2: Very few flaws in this run. The biggest one was mistiming a bomb on stage 2. Luck was high, execution was mostly spot on. The stage 1 midboss should be luckier and better, and I subtly failed some optimizations here and there, particularly on stage 3, but it's overall a success. 90 million would be amazing and far more difficult than 89 million, which was already far more difficult than 87 million, so.

Lunatic A2: I went with the old resource pattern on Stage 1 since I was below 295% PF at the start of the first spell. I'm not 100% convinced one is truly better than the other, though. The resource paths through stage 2 and 3 are heavily influenced. Stage 2 is the worst in this run because the luck was not good. I got a good luck on the midboss, but Sunny Milk's first spell was not high (a million lost compared to a great luck) and her second spell was also not high and it was also 2 forced deaths and some crazy dodging. Stage 3 was high, that's become a staple in my A2 runs. However, Fairy Overdrive was a big trainwreck, so I lost around... 800k from that.

I don't really feel like pushing it more right now honestly. Might look into a different game (Crimzon Clover?) or return to one of the photo games for a bit.

Always nice to see you keep pushing this game, but it'd be nice to see you destroy another game for a while.

Which routes do you have WR for, again?

I'm lacking 3 WRs (Easy A2, Normal A1, Extra) and I've never had any real plans of getting those, unless if the Extra WR somehow came in a random attempt which would be ridiculous considering my PB is almost 2 mil behind. I've never actually been on pace for it late into the Marisa fight, but I have been on pace for a 100 mil run. It's possible I'll go for these some day but they're all amazing scores and I really don't like Easy A2 at all.

The record history section is interesting. I'm most concerned with making sure my scores are as good as possible, if I had just stopped after getting a record, I wouldn't be half as good as I am now. I'd like some more competition honestly but it doesn't really seem like any other GFW scoreplayers are active anymore except for possibly DATA. A back-and-forth can be really motivational (or crushing if you're nowhere near as good, heh...)

And by the way, I am maintaining a scoreboard (capped at 5 entries per category) mainly since so many players removed their scores from royalflare. It shows that there's been a good amount of competition over the years (although not as much as the game deserves)

It seems that when I play non-GFW/StB/DS I always get bored/impatient long before I actually reach the point where I can play at a top level. Besides, few games feels like they're even fun enough to think about it in, and the best ones, the games I actually would like to be able to destroy, typically are extremely hard and have steep learning curves (like PCB or Dragon Blaze) so it's a huge time commitment.

Just going to put these in. Sort of went back to easy since I wasn't too satisfied with what I previously had. Also dropping normal runs.

Easy - A2
Prinny - 30,285,650 - C - 0.1% - Replay

Easy - B1
Prinny - 25,557,250 - C - 0.1% - Replay

Easy - B2
Prinny - 30,221,630 - C - 0.6% - Replay

Easy - C1
Prinny - 26,486,090 - C - 0.1% - Replay

Easy - C2
Prinny - 30,027,740 - C - 0.1% - Replay

Normal - B2
Prinny - 31,545,680 - C - 0.2% - Replay

Normal - C2
Prinny - 32,067,310 - C - 0.2% - Replay

Easy runs were pretty much done kind of quickly. I haven't quite figured out how to approach some of the patterns in the game yet. The main thing I was trying to learn was where to bomb. I'll probably see if I can end a run in easy at 1000% motivation and work from there if I do get around to doing more runs on easy.

Pretty much just trying to go for a clear with the normal runs. Did try to learn B1 earlier, but I didn't get much progress. Kind of just taking small steps.

Lunatic - B1
chum - 104,755,780 - C - 0.5% - Replay

Somehow got this, definitely my best run yet. I'm pulling nearly all the stops I'm aware of in this run and the only thing I think sucks about it, the early accidental deaths, actually didn't really hurt the run at all, since
1. they didn't interfere with my patterns, causing no real immediate loss
2. there's nothing particularly great i can do with those lives anyway
3. the extra PF caused me to change my PF pattern and actually gain score out of it

So even the part of the run that looks bad was actually fine. I would've been in deep shit if I had died at all on Star Sapphire or stage 3, but since I didn't, no problem. Dying on Star Sapphire is typically an instant reset at my level regardless of how the run is going anyway. Almost everything was simply great this run. There is still room for more score, you can have better luck on Sunny Milk, better three fairies, in particular you can go even crazier on Break Canopy if you're completely out of your mind, an optimal resource route along the extra PF from luck and such, and other minor subtle details that can go better really shows that I'm far from pushing the limit in this category. There really is no limit even if you start breaking 107 million or something insane like that.

Can break 100 mil consistently without a problem. Much more than that starts requiring resets but I'm pretty pleased with how far I've managed to take this high-execution and very RNG heavy run.

Fortunately or unfortunately I'm not sure which, there just ain't no other (non-photography) games I can play seriously. I tried crimzon clover for like 2 or 3 days and just wasn't feeling it, I went back to GFW again. Maybe some day I'll go all out with a different game.

Easy - A2
chum - 42,457,640 - C - 0.1% - Replay

Didn't get garbage luck (nor awesome luck) but there's a lot of subtle suboptimal things and major mistakes like missing a item set on fairy overdrive.

Don't think I'll manage to get close to Dream Dwellers beastly 43.8m run without a lot of effort, which I doubt I'll be willing to put into this category.

Easy - A1
chum - 36,217,610 - C - 0.2% - Replay

Easy - A2
chum - 43,016,820 - C - 0.1% - Replay

Easy mode update. A1 improved by almost a whole million! I don't think there's enough room for score for 37 million to be possible in A1, but while I had an idea for a pattern difference on Star Sapphire that unfortunately seemed impossible, I'm still wondering if I'm missing something. Said improvement would likely increase the potential by about 200k. Another 600k from luck and optimizations is dubious but in the end, 37 mil may not be strictly impossible. Just not a score anyone's ever gonna reach. 36.5 mil could come out but I'm content with 36.2. As far as flaws go, I think the big ones are: 1- Frozen item set lost on 3 fairy nonspell 1. 2- Stage 3 bomb left some fairies alive. 3- Lower luck on Luna Child than wanted. 4- Frozen items lost on Star's first spell. 5- Lower luck on Stage 3 midboss. 6. Lower SCB on some cards due to easy mode ice shenanigans. Several hundred thousand points worth of flaws mentioned.

Easy A2 is a different story because the luck, and optimizations are much broader. Like, way broader, the luck particularly. I'd like to say 44 mil is obviously possible but... an incredibly steep target, to be sure.

I think I'll spend the rest of the year on easy mode. B2, C1 and C2 are gonna be fun. Still undecided if I really wanna leave the A2 score like this.

Easy - C1
chum - 38,077,290 - C - 0.2% - Replay

Easy -C2
chum - 41,589,020 - C - 0.1% - Replay

38 million on C1 was a nice milestone to reach. The score could stretch a bit more, but it's fairly optimized.

C2 should feasibly reach 42 mil at the very least, but I got 41.5 quickly and am fine with it. I had a pattern idea for this that made 42 mil more realistic but I didn't put it into action. With this pattern a 42.5 would be a really great score, I think.

Not tired of FW at all but I've put it on hold. A few days ago I really got the itch to clear some more "difficult" titles as a change of pace. FW is paused as such, since I need to put my concentration into other titles if I want to clear them smoothly. The reason I feel this way is because I have become so used to just optimizing and doing difficult scoring stuff, and have not struggled to try to clear a game in a long time, and I am more confident in my skill now. I got Strikers 99 out of the way the other day, and I'm looking to try to clear Strikers II next (keyword try) beyond that there's Ketsui Omote and some other things. What I want primarily through this is just clears, but a secondary hope is to find another game to play for score.

Extra
Shepardus - 60,482,730 - C - 0.2% - Replay

First time getting over the maximum clear bonus. Probably won't take much to improve this, but I'm submitting this anyway.

Lunatic - A1
chum - 106,648,440 - C

Lunatic - A2
chum - 119,625,730 - C

Lunatic - B1
chum - 107,163,550 - C

B1 is the best score I've ever set in anything by a considerable amount, and it honestly might stay that way. We'll see. I wrote a blog post talking about some of the possibilities I'm aware of, because the score can go beyond 108 million. It is a matter of taking some of these absurd strategies seriously, piecing together this delicate puzzle because obviously certain patterns will rely on different "branches" of patterns or, even more luck. Redoing it all from scratch, and relying on that amazing luck all over again, while doing all these insane things and somehow putting all the pieces together without fail. I believe in 109 or even 110, it is realistic if you have immense skill and dedication and if you take more of these TAS-tier patterns seriously. It doesn't end at Break Canopy 4x+ sweeps. Players in other high-skill ceiling games prove that the possibilities don't just end where you expect them to.

A1 L should have a pretty similar ceiling to B1, though the skill ceiling itself is lower. The luck involved is gentler and the strategies are easier. I figure my resource pattern is somehow sub-optimal if you squeeze in a huge number of 30%s. But, I don't know if it is humanly possible to count on. The best idea I have to try to bring in yet another bomb in a run involves the 3 fairies 3rd nonspell, where it should be possible to get more 30%s without damaging the side fairies. The luck involved would be enormous and it would be necesary idea to PB-graze Star every wave possible, so unfortunately, it may be a TAS only strategy. Then 30%s galore in the end-game. The mere idea is sadistic as hell. Either way, without resorting to this stuff, I'm certain that 108 million can be achieved.

A2 L, of course, is a matter of luck. 121 million is realistic if you get great luck in all the main places (Lily Black final non, Sunny's first two spells, and midboss Star) while following my patterns. There are multiple viable resource branches, if like me you you use an extra bomb on Sunny 2nd non, then die 2x on diamond ring, you can use yet another death and bomb on st2. Then use 2 suicide pattern on Star and it pretty much evens out the potential. More can be done than what I do, there is a slightly improvable path, and especially the 3 fairy 4th nonspell has a high score ceiling that you cannot reliably reach without resorting to a TAS-tier pattern. Ice Dissolver also has a high ceiling which relies both on luck and skill and it is difficult to estimate the possibilities there. All things considered, I believe in 122 million as a possibility on this route, but 121 million is the "realistic" estimate.

Both A1 and A2 have very tight motivation gain to get a final life in, which is also scary as hell when estimating some of the possibilities since a final suicide would be essential for some of these insane patterns to be remotely realistic.

The estimates are all assuming that the player has a very high fundamental ability because it will be impossible otherwise no matter how many 1000s of hours you put in, unfortunately. Some of my ideas may just be impossible full stop, but theorizing even potentially impossible things is important to bring the game forward...

The other 3 routes are lagging behind now, so I will try to bring them up as well. I'm especially confused by the score in B2 and a little in C2 right now... but I believe C2 is a lot like A1 except slightly ahead and less evened out with the luck. If I believe in 108 for A1, I'm pretty sure 109 could come for C2, but C2 relies on huge luck in a few key places, most importantly is stage 3 midboss spellcard. B2 just confuses me greatly in a number of ways, but I'm sure 106 should come out. C1 is surely the lowest scoring of all the routes, and my current estimate is that 104 is a good goal while 105 could be a possible goal. I have one major improvement idea for it.

Great Fairy War that never ends!

Great stuff as always. I loaded GFW up tonight after reading that - still can't clear Normal, which is infuriating in a three stage game, but I always appreciate how unique the game is. Maybe this time I'll stick with it long enough to get the clear.

Normal - A1
chum - 61,392,690 - C

I'll put this here now as well, as I don't know when I'll get back into shmupping again. It's my most optimized score in this game beyond Easy mode (the actual most optimized ones I believe would be Easy C1 or Easy A1). That doesn't mean it's perfect, though, far from it. 62 million is realistic with my knowledge but it could take a very, very long time, and I personally do not really enjoy this category enough to go through that kind of hardship. Thus, 61.3 will have to settle for the time being. I'd rather have the other routes on Normal go higher first. C1 especially is no good at all, but I think B1 and B2 as well should be played much more in the future. A2 and C2 are "acceptable", but of course I hope they'll find a bit of an improvement as well, like the mythical 73 in A2.

A certain suicidal finn beat my A1 L score. Ever since I became I guess the main authority on GFW by taking nearly all records and improving them etc, that's the first time someone beat me! Me solely producing any new records for like 2 years has been pretty sad so seeing this event made me glad. I'm not playing currently, but when I start this game again, A1 shall up to at least 108, though the real aim is 109. My prediction is that my final score will be short of 109, to be honest, but I'm going to at least try. It'll depend on a few details that I have to look into thoroughly how well they can establish in a run. I also want to put a lot of emphasis on the old neglected routes, B2, C1 and C2. The records are now 1 and a half years old, and I know for a fact that they're not as well researched as A1 is now. Likely they're all going to get improved by a few millions when I come back.

This is the thing. There's finally practice tools for GFW that allows you practice individual bosses and stages for any route / difficulty. IMO, this changes everything. So I'm going to try to be more thorough on for example B2 L and see if maybe I can incorporate some stuff.

Lunatic - A1
Square_Air - 25,755,460 - C - 0.1%

May as well drop off this survival clear I got over the summer since I've moved on to other games. I definitely need to come back to this next year to improve that score. That extra stage looks pretty cool too.