Posts for AnS

1 2
26 27 28
Posted: 6/6/2006 4:35 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Haha, look what I've found. http://shedevr.org.ru/temp/sound/RK-trip.fcm Keep watching the scroll. If you wait several minutes, screen will come down to quite recognizable levels from Channels 4&5. The game takes bytes from ROM and creates level map on the fly. I doubt this glitch can be used in speedrun, but it shows that the game has some real bugs to study (another one can be seen in my Rockin'_Kats-basic.fcm - in Channel 1 boss fight: collision detection fails because of huge lag).
Posted: 6/6/2006 3:21 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Good example of 'WTFmovie'. I was impressed by scrolling speed (haven't seen this even in Sonic games). Voting Yes.
Posted: 5/30/2006 10:00 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Baxter wrote:
a great part of the mm1 and mm2 movie are based on glitches... which would be a lot harder to implement in this sort of movie, and wouldn't be any more special than what you can already see in the normal version of mm1 and mm2.
But don't you think that some people would like to see good mm1 speedrun without glitches? While there's one good glitchrun on nesvideos, there's no need to show those glitches again, everyone know them already. Also, not using those glitches slows down mm1-2 progress and makes their length similar to 3-6.
Posted: 5/30/2006 8:32 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Nothing was perfect. :-/ I noticed that you still refuse to fight bosses with memory watcher, that's bad, because this game lags as hell - it's critical in boss battles, especially in this one you recorded. In fact, the game lags even during levels - every time you release punch glove you lose from 1 to 5 frames. Well, tomorrow I'll try to continue current recording (watch my file) and finish miniboss earlier than 3100 frame. Even though I'm not sure I made perfect work before approaching the boss - really evil game, that is.
Posted: 5/30/2006 7:51 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
AngerFist wrote:
AnS wrote:
And the fact of finishing whole set of games is more exciting than the fact of finishing just 4/6.
Keyword from that quote is: "just" :)
Hehe, I'm already tuned into that mood of perfection, when the world is not enough.
AngerFist wrote:
Dont worry, of course our aim is to make it entertaining as possible, see my first post in this thread why we cannot include Mega Man and Mega Man 2.
I've read it, but I mentioned those length differences as an aestetic issue - it's not impossible to add short games, but it badly affects performance (cuts some entartainment ways) - so you need to choose between perfomance and significance of the fact. I'd personally choose fact (follow main goal to the end).
Posted: 5/30/2006 7:28 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Great movie! And I'd really like to see Lost Vikings 2 (SNES) played that way...
Posted: 5/30/2006 7:23 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
It would be so cool that the fact itself becomes more exciting than watching process. IMHO, the movie would be more like evidence, not an entertainment. Because of that I think adding Magaman1&2 fits the aim of the movie more than some aestetic issues do. You may forget other aims (like "fastest time" or "no damage") but this goal ("show the fact") is the most important for such type of runs. And the fact of finishing whole set of games is more exciting than the fact of finishing just 4/6.
Posted: 5/10/2006 10:30 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Phil wrote:
Ans: No, Monopoly is the shortest movie.
Yes, but I meant genres that are more common to video-game definition. To my mind, clear nature of TASes can be felt only while watching platformer/action speedrun. This opinion comes from the word 'ninja', used frequently here. It's little strange to call someone ninja for precise dice-throwing or perfect memory value manipuletion. I highly respect all the work behind Monopoly (brute force after reverse engineering - that's what I call perfect solution), but this just doesn't suit my expectations of tool-assisted speedrun. From speedrun I expect a lot of fast-paced tricks allowing character to fly (literally and sometimes actually), when screen scrolls forward without slowing for a frame. That's what I wish to see having shortest TAS here.
Posted: 5/10/2006 9:17 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
The game is one of my favorite NES platformers, although Samuri Pizza Cats series were officially translated to Russian only recently, that's why back in childhood I was mostly watching Japanese KNT version. And these were lovely days! Uhh... And now Ferret Warlord made my day. This speedrun has plenty of interesting tricks, several moments were totally unexpected even for a fan who played this game... like for 30 times already. So, I hope this TAS will be published. Dunno why, but I have strange feeling that the game is little underrated.
Posted: 5/10/2006 7:19 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Damn it, this game is evil. I'd sue Atlus programmers not only for making almost bug-free game, but for sadistic password system. ]-\ Well, here's the story. Recently I was trying to improve Baxter's run in 3 ways: 1. Find a password that could be entered faster. 2. Run levels better. 3. Finish boss faster. Current statistics: 1 - 1 frame saved 2 - 17+11 frames saved 3 - working on it, 20-25 frames can be saved Summing up, not more than 1 second can be saved using ways 2 and 3. But 1st way had a potential of saving 2-3 seconds... if it worked. I studied the game password system (using FCEU debugger), and found out unfortunate news - there are two types of passwords: generated and predefined. For ordinary Channels they are generated (taking into account items / weapons / money) but for Channel X (MAPHARDMAP) it's taken from ROM. Many games use predefined passwords for cheat codes, but not for levels. If you are interested in it, (A426-A462) - table of predefined passwords. At least I'm satisfied with founding FIGHTBOSS[1-5] password (which isn't mentioned anywhere). ;[ Aaand I still fail to improve levels 2,3,4 - seems they are 1-frame perfect. Baxter, you did really good job here! But I want this game to have absolutely perfect TAS(es).
Posted: 5/8/2006 9:05 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Baxter wrote:
It is true that I am not hitting the boss every 33 frames the entire battle. The battle is (when playing normally) very laggy, and I avoided it as much as possible... but it seemed not possible to get it entirely without a few frames of lag. Also note that 04B3 doesn't notice if there is any lag, so you should watch out for this.
Yes, 4B3 isn't the main address I was speaking about. I'm watching 4E2 and manipulate lag (lag manipulation is way harder than luck manipulation 8-)) to be able to hit boss every 33 frames. Now THIS is tedious...
Posted: 5/4/2006 4:51 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
To think it, Bastet is totally different from what I made in this hack. Bastet ruines essential feature of Tetris gameplay - ability to create logic structures. Creativity is the main component of Tetris fun. Whole game you construct various structures, turning abstract tetrominoes into something useful. You build your own patterns and enjoy the process more than result. Bastet's aim is to prevent you from creating something recognizable. It simply verifies every possible structure you can build with one of 7 figures and then chooses least humane structure. That's why you can't be addicted by Bastet (only excited by challenge). And S- and Z- stuff is of another matter. Every bit of Tetris' fun is preserved, but you have to deal only with 2 types of details. This feature only makes the game more repetitive than usual, but you still can have the same fun as while playing normal Tetris.
Posted: 5/4/2006 4:38 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
CtrlAltDestroy wrote:
I found a strategy which was able to be repeated for a long time, but it couldn't go on forever because the two sides kept going out of balance.
Whoa, you managed to triple my result. That's great! But I wonder if it's possible to get 999.999 points? Judging by their phrase "This game must end (in a loss) before 70,000 pieces have been played" it's really possible somehow. One note. You stopped movies too early, it was possible to survive about dozen of figures more. But still you receive "best ending" with current score!
Ramzi wrote:
I read but could not find...
Here is Windows version of Tetris with Bastet mode.
Ramzi wrote:
Does Bastet give all of the pieces, or only continually the Z and S pieces? It says, "Bastet uses a special algorithm designed to choose the worst brick possible." Doesn't this just mean always Z and S pieces?
No, its algorithm is based on points counter. Before choosing next figure, programm calculates every result of placing this figure (there are 7 of those tetrominoes) into every possible place. Assuming that you are perfect player (that you'll choose place with max points result) it finds min(max(tetromino,place)) and gives you worst tetromino (with minimum points result). S- and Z- are not always the worst figures judging from point of view of score counter. They just force you to leave gaps (that's what cause gameover), but they give good score (as you may see in our movies). Bastet can give any figure - S&Z make you lose game and other figures make you lose points. Bastet algorithm has some negative side-effects - it constantly gives you series of similar figures, that makes the process look more artificial then S- and Z-only generator.
Posted: 5/3/2006 4:41 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Hmm... Bastet algorithm is pretty simple, but implementing it on NES would be rather difficult task. That seems more difficult than writing PC Tetris with needed tools involved (savestates+frame advance/no gravitation).
Posted: 5/3/2006 4:23 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Swedishmartin wrote:
How did you change the number generator? Notepad? :P
Why notepad? I used hexeditor+FCEU debugger to modify programm code (rewrite one subroutine). I'm romhacker after all.
Swedishmartin wrote:
About the FCM: as long as you get into a "scoring pattern" as the one on frame ~8000 you should probably be able to reach the 9999... ponit mark. (From now I will call these scores "niney" scores.)
Yeah, there is much room for improvement, especially at the beginning.
KDR_11k wrote:
I'd rather see a TAS of Bastet, trying to manipulate the algorithm into giving you blocks that allow you to survive...
Hah, didn't knew this! Maan, this is frustrating, I can't fill even 7 lines with this. Need to practice...
Posted: 5/3/2006 1:05 PM
Post subject: Surviving Tetris
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Did you guys knew that Tetris is THE Game? Of course you knew. :) But did you knew that Tetris is the game you can't win? If you don't stop after some score mark, you are doomed to lose sooner or later. Well, in theory. Probably you also knew this fact (from wikipedia or other source) - about S- and Z-shaped figures. Maybe you even played Java version on this site (I hope there are diehard fans of tetris on this forum). In this version you have worst luck in a Galaxy, one chance on million - getting S- and Z- in a row for a long time. I tried it and found it pretty interesting to play/watch playing. Although it's normal tetris with classic rules, the feeling (emotional experience) is very different from playing Tetris with natural luck. Usually, while playing Tetris you relax, because if you make a mistake you can fix it later with other figures. But when you are playing against the luck, you can only hope to prolong playing time and collect as much points as possible. This type of gameplay is more frantic - it's a real challenge for those who think they play Tetris well. Someone can be called Tetris guru (or whatever) only if he can show his best even when situation is inhuman. I find that Java version too uncomfortable to test your skills. Mostly because of bad controls. That's why I decided to play NES version, which provides plenty of tools, you know what I'm talking about. Using luck manipulation I could put this intention into effect. But then I realised how tedious would it be to make and how uneasy to watch (constant flashes because of pausing). It's better to make little hack that changes random generator procedure - so I did, now it operates only with S and Z. Nothing else was changed. This hack is merely tool for providing necessary luck without manipulating by key input. You can download this patch for your Tetris_(U)_[!].nes from my page. (ips+exe versions) Now about the movie. Eh, it seems I'm not that big fan of Tetris as I thought, my record isn't that high. Still it worth watching, I doubt you've ever seen such gameplay. First minute has nothing special, but soon after frame 5000 the fun begins. Here is FCM file for this record. I encourage everyone who loves Tetris to play this type of the game and show their records. Now just imagine you are playing Tetris during the break and suddenly you get worst luck in a world. How long can you stand it? ;)
Posted: 4/30/2006 5:22 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Yeah, this game is destined for TASes. IMO, game with such unique gameplay deserves two runs on this site. This little version seems like preview for bigger one. Also it would be the littlest movie (with AVI size less than 2Mb) anyone can download in a wink. (especially good for first-time viewers, because it shows real meaning of TAS) Although it can be improved a little bit... I'll post details in Rockin' Kats topic.
Posted: 4/30/2006 5:06 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Baxter wrote:
You mean you have a movie completing the first 4 Channels?
Yes, but I doubt it can provide valuable sequences for final run. I was testing various tricks, combining weapons, finding some ways for entertainment and so on. Anyway it's supposed to be rewritten almost completely. Here is it. Maybe you'll notice some interesting tricks or strategies (e.g. walking down sewers at Channel 5 is faster than Diman's previous decision). As seen in this WIP, there IS some room for entertainment tricks (without losing time). Boss fights aren't optimized at all, and what's important here - they should be recorded using memory viewer, because it's not always obvious that boss' invulnerability gone. Here are useful memory entries I found today. 4B3 - boss current status 01 - intro 02 - normal (waiting) 03 - attacking 04 - moving 80 - suffering 05 - invulnerable FF - defeated 07 - outro When 4B3=80 or 05, you can't damage him, but you should do it first frame when this turns 02. It's even possible to avoid 02 at all - as seen on Baxter's Channel X run (still there are moments when 02 appears, even for 2 frames). When 4B3=05, it countdowns before changing to 02 4E2 - timer 4B9 - boss life (when becomes zero - 4B3=FF) 4C5 - invulnerability data (you can't affect this directly) 4C8 - for storing status data while in 80/05 state. After 4E2 becomes zero, 4B3:=4C8. That's why boss can continue his attack after you hit him. But usually it's just 02. That's it, now you are bound to make boss battles 1-frame perfect! ;)
JXQ wrote:
I thought levels had different enemy layouts depending on the order they're played, meaning hex-editing isn't really an option. Just keep recording the run.
Yes, they have, but if once decided to proceed reverse order (4-3-2-1), they are TASed same as if they were predefined. Pieces are savestate-based, starting points for them can be picked from existing reverse-order run (currently published or my). So I don't see a problem here.
Posted: 4/30/2006 3:58 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Diman wrote:
Well, i really appreciate that you guys out here give me support and tips! Yeah, I have to take my time to make sure the run isn't sloppy.
That's not enough already, it's time to make perfect RK run or not to make anything.
Diman wrote:
Why does the playback stop at the first level for me?; (
Because I stopped recording right at frame 1285, it's same situation as frame 1365 on your movie example. 1 frame was saved at menu and 79 - at level itself.
Diman wrote:
I don't know how to use a hexeditor (I have forgot how to hack), but I can use the frame advance function more careful.
Right now it's not necessary for you to worry about hexediting. First we should have all levels' records, and only at the end somebody will assemble them (well, I can do it, but only when I'll be sure that current versions are perfect). That's only way to make perfect movie without much tedium (because you can record levels in any order, rewrite levels quickly and easily, share specific levels' tactics and so on). You can do the following. Choose a level (each Channel contains several levels), start recording at the moment when screen shows from black (when first BG layer appears, but not sprites/stats) and then finish it somewhere at the end of the level (for example when screen goes black). Key Input from this savestate-based FCM will be used later (or it'll be obsoleted before linking). With this method you can improve each level separately, and others can contribute their versions of some levels (but nobody would contribute whole game run "example"). Well, that's just an advice, you decide it.
Posted: 4/27/2006 11:19 AM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
OK, I'll try to explain. Main glitch - shooting wihout usual stop. In this game character can be two states - when it's turning around and when it's just swimming. When you are turning back, you still can move (change coordinates), but controls are reversed (you may see this looking at Input Display) and they affect not only coordinates but turning stats also. You may even end up facing same direction after complete 180degree turn. And of course you can't attack while in this state. Attack can be released only in normal state, attack initiation subroutine sets a bit that will be cleared only after 16 frames. That's it, you lose at least 16 frames for every attack (also speed meter drops to zero => you are to regain speed, so in sum you lose 25-30 frames per attack). As you may see now, developers made everything like this: you can't shoot while turning and you can't turn while shooting. But there is one flaw in their code. In terms of one frame turning subroutine executes BEFORE shooting one. So if we manage to start turning and shooting at the same time, first we'll initiate turning (but for this current frame character is still in normal state) and then we'll release an attack. Of course this can be done only using Frame Advance. This trick helps to save those 16 frames, but speed will be dropped, so you still lose 9-10 frames per attack (sometimes less if you don't need to move for a long distance - like in boss battles). While this can be debated whether it's glitch or not, it surely wasn't foreseen by developers of the game. Something neat - temporary invincibility in the boss battle. Boss battle officially starts only if you appear in boss room in normal state. If you manage to enter boss room while being in turning state, battle music won't start playing, and the game will think you still play on a level - up to the moment you leave turning mode. But you can stay in this mode (I mean state) you would be able to move (slightly change coordinates) in boss room before battle actually starts. And after the battle starts the game will not check collisions for a couple of seconds. Too bad you can't move this period. Now about luck manipulations. Mostly I manipulated enemy drops in boss battles. Usually they drop 50% of good stuff and 50% of trash (which can't be used to damage them). It depends on frame you enter boss romm, on your actions in boss room and even on your actions during a level. Methods of luck manipulation in this game: * because it's a free space game, there always are several routes with same time of reaching some point (boss room). I just compare them and choose one with best luck at the end * also it's very important how do you dance in boss room (if you look closely, you'll see some strange input several times) * you can control boss simply by timing your hits, but it's not always good for fastest completion
Posted: 4/27/2006 1:08 AM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Fabianx wrote:
Are there any disassemblers for other consoles than NES. Like SNES?
You mean debuggers? There's good Snes9x mod: http://geigercount.net/crypt/
Fabianx wrote:
And also: When I know that it changes variable 7EE7CD for example, how can I find this address in the ROM. I mean, how is the address format?
It depends on whether it's HiROM or LoROM. And don't forget rom header size. SNES memory adressing is quite a puzzle. :) Read docs for details. http://www.google.ru/search?q=SNES+Memory+Mapping
Posted: 4/27/2006 12:39 AM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
Aitamen wrote:
I'm doing an official speedrun on the NES, and I'm going to do three versions... one no hit (howl damage), one no damage, and one full speed... all three, first hacked controller and then my JS controller ^_^
Does anybody know how it's progressing now?
Tafatt wrote:
This game can be improved maybe 20 seconds.
Yeah, maybe even by whole 25 seconds! :) I'm working an inmrovement for about a week already. Several additional notes: * it's better not to take first "red W" at the beginning, because it pauses the game for cutscene and also throws you back. Quite a difficult thing to realise actually. This action is so unnormal for everyone who played the game, that I could find it only because of listening to ManForm music (I tried to keep it playing as much longer as possible, because for the whole game we'll be listening to WerewolfForm music) * I already tested that collecting 5 balls and killing Werewolf in 10 hits is faster, so now it's only necessary to find 4 best places of those balls without big delay. 5th place is last GiantHead of course * actually HOWL does a little damage on 2nd Faryan, it's better to save 2 howls for him * there is one heart in level 4 after GiantHead battle, so you can use 2 howls for him and 4 howls for Monster * thera are several additional glitches (still not as cool as L+D+R) to abuse. Charge can be saved by pressing A at correct frame. It even can be saved before fully charged (so you can charge partially) Right now I have no intention to post current WIPs, because I'm trying to figure best strategy. Lots of planning stuff, you know. AND FINAL NOTE. Does everybody here think the game is difficult?? I used to walkthrough it on real console, and don't think it's so hard - it just requires some time to master, after that it's easier than Mario. :)
Posted: 4/27/2006 12:21 AM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
No, no, that's not it, Diman! You still play like you don't have enough time and tools for your assistance. I can understand losing some frames during the level, but why did you wastefully jump at the end of this level? Normal walking from this point ends level 16 frames earlier. You could try this as one of several possible ways to reach point B (all sprites are gone - frame 1365 in your FCM). To achieve fastest completion you should try many different ways (and maybe not-so-different - even 1 frame difference sequence should be tested), comparing them to eachother EVERY SECOND (or every 100 frames, or some other checkpoint - you name it). If you just walkthrough a level several times and then compare those versions, you won't allocate strong and weak points of each way. You should analyse it precisely. Here I recorded good example of this level strategy. It's good because it contains a couple of interesting moments. Look near frame 1165 and then near 280. Notice that sometimes it's better to postpone rolling (for letting screen scroll) and that bumping a barrel (or whatever it is) not only moves you fast (still rolling is faster) but also scrolls screen faster than usual (that's what important here!) Frankly, I had plans on making this run improvement myself, but I got bored at Channel 5, so you'd better follow your signature! First, try to beat this level example (even 1284 would be good) - just be creative and eventually you'll find a way. Well, if nothing can be done here, you may insert my sequence using hexeditor. This game is hexedit-friendly (you can easily cut and paste sequences whithout fearing that it would affect next levels randomness), so maybe good tactic is to record each level separately, then choose best versions and assemble them into one (I've done this for CS-Jetman movie, and LettleMermaid is completely unfriendly to such method - it changes everything if any frame were modofied). Uh, and what about taking damage, have you decided? As for me, no-damage run is more valuable, taking damage saves only few frames in several places, it's nothing, you can run the game in less than 20 minutes without it. Well, good luck!
Posted: 4/25/2006 11:14 AM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
spockybiemmichab wrote:
I recommend to enable the "Input Display".
Hey, actually that was kind of easter egg! :) I had no plans on mentioning it as a feature. Oh, well.
Posted: 4/24/2006 3:03 PM
Quote
AnS
Emulator Coder, Experienced Forum User, Published Author, Experienced player (724)
Joined: 2/23/2006
Posts: 682
In this game it's not always so important to release an attack as fast as possible. It's more important to manipulate enemy drops and try not to slow down boss progress (every hit pauses boss motions and postpones it's actions - and by actions I mean enemy drops for my future attacks). I'll explain it in particular. In 1st boss battle only last hit really matters. First three hits don't affect time, in any case you must wait for second shark appearance. So I just make 4th hit in the first frame possible. In 2nd battle those little crabs can't be used before they start walking (you can't see this because I freese them 1 frame they try to walk), That's why main goal here is to manipulate their "sleep time". It seems current manipulation is best case of luck, but still some frames were lost by waiting. In 3rd battle it's actually better to release all 4 hits as late as possible, or boss stops dropping shells => you will lose several seconds. That's why in first part I just played shells, letting him leave quicker. 4th battle is nothing but perfect in terms of time. If I'd release any attack 1 frame earlier than I did, it would make boss to shoot stars (which can't be used as projectiles) and also it would postpone his shot by 1 second. 5th battle isn't that perfect, but it's definitely not slow. Main problem here: fishes are invincible 10+ frames after they appear (spawn from whiteball). At start of the battle it may seem I just fool around. If I didn't, some whiteballs would spawn stars (which can't be used), so again it's luck manipulation. 6th boss is just mindless battle, nothing special in strategy.
sargon wrote:
But when I watched it looked like the movie needed some end input.
Read the submission text to find this second version of FCM.
1 2
26 27 28