Hey, this is my own page! I now have one to myself!
I consider myself pretty decent at programming. A few emulator lua scripts to look at:
- Subtitles script - General purpose subtitles display.
- MtEdit (lsnes) - Rather dated, but it will use your configuration in lsnes and it has a nifty display.
- TASEditor XOR input helper (FCEUX) - All it does is read up your controller inputs and toggles the button for the current frame. It's sure convenient for me, since I'd like to use the keyboard to manipulate input, not the mouse.
- MtEdit (FCEUX) - Obsoleted by TASEditor. No plans to work on it.
I have also written a few FAQs at GameFAQs, if you feel like looking through those. Hopefully, my more recent FAQs show themselves as better written than earlier ones.
Outside of writing and programming, I have done a few TASes as well...
- NES Solar Jetman
- GBA Alien Hominid
- NES Rockin' Kats - The DTC4 winning movie. My team will say it was mostly me.
- NES Magician
- SNES E.V.O. - The Search for Eden
- NES Dungeon Magic
- NES Dungeon Magic - Cancelled due to improvements (which are published, as above)
- NES Guardian Legend, The - Rejected. But I never expected it to be accepted. A short 3 second long movie file that beats the game... If you leave the A button on auto-hold, and it also uses a password to access otherwise impossible values. Admins and judges, feel free to cite this as an example on what NOT to do!
- SNES Yoshi's Cookie "V.S. Mode" - Rejected. Another April Fool's submission. This was, however, among my first attempts into using lsnes. I figured I would at least try to have a 2P playaround. Not as though I expected success, but I wanted to provide some support to lsnes. With the Vault Tier now existing, perhaps I should try the full 1P Action mode...
Unfinished projects Stuff I started, but have yet to see completion
- SNES Soul Blazer (.lsmv (USA) | .lsmv (JPN)) - An attempt at improving the published run. If I let this rest for a bit, maybe I'll have the solid determination like I had for Solar Jetman.
- SNES Brandish (.lsmv) - It's in my head. I'm pretty close, so very close. Not many floors left.
- The Guardian Legend - Honestly, I'm not sure if I'm ever going to complete this with my habits.
- NES Salamander (JPN) (.fm2) - Better known as Life Force in USA. I hope to someday beat adelikat's Gradius run in entertainment!
- GB Final Fantasy Legend - A joint project that got real far. So close to finally finishing, with all the digging we did...
- SNES HyperZone (.smv) - Fun little rail shooter. Rather than fastest time, I aim for maximum score. I'm... Getting there. Someday.
- SNES SimCity (.smv) - Someone came by. I'm still hoping we can work together. Doesn't seem likely now, but feel free to surprise me.
- DS Etrian Odyssey - A particularly long game. As a bonus, there's a map we can draw on at TAS speeds for some entertainment. Sadly, DeSmuME updated, and my old computer couldn't run the newest versions.
... And now I realize just how many projects I leave laying around, unfinished. I'm always willing to provide help relating to my old projects if anyone wants to TAS those games.
Submission Text preview (Zook Man ZX4)
Oh, I also use this page as online storage for some upcoming submission I plan to do. Submission texts take a while to write, and it is helpful to write while the stuff that happened in the game is still fresh in my memory. Putting stuff here serves two purposes: One, I have an online backup that will almost certainly exist no matter what happens to my computer; and two, people will have a chance to correct my mistakes before I submit. As of time of this writing, I have the TAS up to, but not including, the final stage. The writing itself is WIP, so don't expect everything perfect right now.
- Aims for fastest time (standard any%)
About the game
Oh, a game produced by Vast Fame, the series is basically a Mega Man knock-off. I don't know much about it myself until some contest, so it's not as though I can say much. Basically go through stages and beat up bosses.
About the run
This game was selected for Dream Team Contest #6. Many teams entered, and surprisingly many had complete runs when it was over. I was part of the winning team, but that didn't mean we knew all the techniques.
After the contest, a few teams pooled their knowledge and turns out there were a few things that no team did, but the collective knowledge and analysis revealed these tricks. Miniboss Skip was known by at least one team, it was later discovered to also work on bosses, lock break on doors work, and we can move left to spawn enemies. Bam, we got a few boss skips in the end stages.
Some of this was looking for which team took on the next group of segments the best, copying that in, and stitching in the best team for the next segments. Author count is very high for this reason. Some of it was applying the new tricks, and a few bits are just straight optimization. Teams 4 & 3 were both very optimal, a single trick could have decided who won, and were well ahead of everyone else, so a lot of times I picked one of those two teams on a given segment. The resources of the other teams still had a bit of extra information to work with.
The Dream Team Contest #6 teams:
- Team 4 (nameless) - FatRatKnight, Pirate_Sephiroth, Exonym, and barely PersonMan.
Team I worked with, and first place in DTC6. This team also had some rather sophisticated display scripts (my fault).
- Team 3 Snail Eaters - lapogne36, Grincevent, Arcree2, and keylie.
The second place team. They made some smart choices I was impressed by. They had discovered fairly similar tricks as Team 4, with a few differences, and made different smart choices that got them within a few seconds of first place.
- Team 6 ¯\_(ツ)_/¯ - Aglar, negative seven, thelegendarymudkip, and Truncated.
Third place team, but there's a large gap between them and second place, over a minute.
- Team 7 (nameless) - Tompa, MUGG, and Alyosha (there's a fourth, but Ready Steady Yeti couldn't find time).
These guys found the Miniboss Skip, which the leading teams could not claim. After DTC6, it was analyzed and we got another 6 skips they didn't get. Seriously, after winning against the bosses the first time, no one wants to see them again.
- Team 5 We'll Hex the Name in Later - Invariel, Lord Tom, Samsara, fsvgm777.
Fifth place. They are well away from the leading teams, but at least they had a trick in Stage 2 they can claim made it into this TAS, that no other team did.
- Team 11 ThunderBears - NxCy, WarHippy, McBobX, DwainiumB.
A team who was randomly put together by the host (they did not define who they want to work with).
- Team 2 MMMM - Masterjun, MediaMagnet, Micro500, and Mothrayas.
These are cool people. They're still back here, though. Not sure what happened, but I can understand if they just didn't want to put in the time into this game.
- Team 1 We Serve Fire Trucks - . These guys analyzed the game really hard, then TASed it. Unfortunately, their analysis basically went down the rabbit hole of just how interesting those mechanics are, and not on the tricks to speed up the run. To be fair, if they had discovered how to break auto-scrollers, they wouldn't be in last place, being generally more optimal than Team 2 otherwise.
- Secret Team The Undecideds - ThunderAxe31, BoboTheKing, DrD2k9, Dooty, and Dacicus. Secret Team because they didn't sign up, there are five members, and one of them is the host of this contest.
Obviously, I pick the fast inputs, but there are times where I can do things that won't really affect the run. I don't mean play-arounds, but helpful tiny bits of input that have no effect at all on the game, but will serve as a TASing aid for improvements.
Whenever we lose control of Zook, I press Start just before regaining control. This is a sync check, letting us immediately know if we wait too many frames. If a change in emulation happened, and the stage intro changed, the pause will immediately let us know if the loading took fewer frames for some reason.
I also use the Select button. It has no purpose in game. It is used to mark "important" frames. Around bosses, it denotes when I inserted intentional delay, the moment camera begins scrolling, lag that happens before boss dialog, the first character in the dialog that showed up (frame rule check), and when the boss particles have all finally vanished. There's a few other spots I might use Select, like a specific position or segment trigger, but I can't remember too clearly about those.
We start with a basic gun, the ability to walk 2 pixels/frame, and the ability to jump. We can also wall jump. Also, we can take damage and die, but we never utilize the dying ability in this run.
Deals one damage. It's also the only weapon we can charge from the start of the game, dealing two damage. With the arm upgrade, it does as much as three damage, equal to an uncharged shot of some other weapons. The basic shots are the only weapon projectile that can break blocks at range.
You can switch weapons while an uncharged shot is present, and this will mutate the shot in strange ways, giving it different hitboxes. This is used at times to hit enemies in strange spots with the mutated hitboxes. It'll also get piercing property if we switch to laser. It's still 1 damage, but it's a rather versatile 1 damage.
Deals 3 damage, with a main projectile and two smaller projectiles. Its charged shot sends us flying at 3 pixels/frame for around 60-ish frames. The big reason to pick this thing up is for the flying 3 pixels/frame, which for some reason lets us do a mid-air jump after it finishes. No frame perfect trick here, just release the charge in mid-air, and after it's done, we can jump once any time. Critical for breaking the auto-scrollers of Stages 2 & 4. The fact the charged flame attack also deals 5 damage is convenient for destroying enemies on the way for items, as no normal enemy has more than 5 HP.
Deals 4 damage with a shot that falls to the ground, then creates a column of damage. Charged shot is two of these, that shortly move as well. This weapon wasn't really helpful, and we only pick it up after getting all the good stuff. If anything, having it will cost us time in Stage 5, so it has the dubious status of being a weapon with negative value.
Deals 4 damage with a spike that runs along the ground, or completely stationary if shot in the air. Its charged shot drops 10-damage meteors that are apparently made of pure lag, and I haven't seen times where it didn't lag the game too much. We only pick it up after getting the good stuff. Nothing against a 4-damage uncharged projectile that keeps lag low, but just having it gets in the way of weapon select in Stage 5, so it's value is negative.
Deals 3 damage with two really slow shots. The charged shot gives a 6 HP barrier that, so long as it is visible, prevents us from shooting, switching weapons, or pausing. This would be a crippling disadvantage, but Charge Swap lets us get rid of the sprite and keep the barrier, letting us fight back with impunity. That makes it a cool "weapon" we pick up right after getting all the cooler weapons.
Deals 4 damage. The missile isn't fast. Its charged shot deals 6 damage, and basically acts the same. Its one advantage is that the charged shot is basically the only one that goes away when it hits something, mitigating lag. Shame its usefulness isn't up to par with all the good stuff.
Deals 3 damage with stars, and it shoots 3 in a spread pattern. Its charged shot deals 6 damage with a pair of pretty fast moving stars in some pattern I find hard to describe. These stars can exist pretty far off-screen and is critical for the Boss Skip, but since there aren't any enemies we can use outside of the end stages, this weapon is left until after we pick up all the good stuff. Would be useful for two Miniboss Skips, but it's impossible to do Stage 6 before Stage 6, and Stage 1 has a rather vital weapon we need rather early.
Deals 4 damage, with an annoyingly long bounce around. As fun as it is to make it work, we can't do much other attacking while it's out, and we've got flame charges to do. Its charge shot fires a pair of 8-damage giant spikeballs that are excellent for taking down bosses, but this by itself isn't really a sufficiently good reason for going through Stage 7 as early as we did. The Leg Upgrade is primarily why we end up with this weapon early, but it also saves a small amount of time over just laser versus bosses.
Deals 4 damage. Not really exciting, but it's still more than basic shots. The charged shot is this mega laser that deals 10 damage and often hits minibosses twice. Lasts just slightly too short to double-hit bosses. The properties of the mega laser also frees our movement in cases where it really shouldn't be freed, which among other small time savers, also allows the Boss Skip to happen at all.
An insideous 0-damage hit that exists for 5 frames after shooting any weapon, and takes priority over any shot. Its hitbox depends on what weapon you have selected, with it non-existent if the basic weapon is selected. It gains a hitbox with any other weapon selected.
If the intent is to deal damage, this thing gets in the way. However, it can break blocks, and if the intent was just to get an enemy into hit-stun, the muzzle flash works as well as any weapon. Since it can break blocks, it is the only way to do so, attacking them point blank, when using weapons other than the basic shot.
Adds an option in the pause screen to allow you to leave a stage. We don't collect this upgrade.
You can pause at the same time as landing on the capsule, and despite the pause graphics not updating the icon, it is still selectable and will let you leave before the capsule animation. It can also be used to leave a miniboss fight, and the miniboss state will be preserved when entering the next stage, allowing for segment skips, although the set-ups are more costly than just doing the stages.
No effect. Did you seriously expect some Chinese knock-off to be fully functional? We don't collect this upgrade.
Many contenders of Dream Team Contest #6 have looked at this thing and couldn't find what it does. Description from something indicates defense, but all damage sources deal one damage to our hero, so there isn't really any damage for this thing to reduce. Well, the one thing that doesn't do one damage would be the camera edge, but I don't think boosting defense would have any effect on instant death.
Gives a ground dash. Double tap left or right to move 3 pixels per frame, 50% faster than walking normally. If dash ends due to its 40 frame timer running out or the player jumps, one frame of zero horizontal speed is applied. If the player dashes off the edge of a platform, this does not stop the player for a frame.
Shame the ground dash does not extend its momentum when jumping. Still, any flat terrain we traverse can make this upgrade useful. Because we need to double tap the direction, we're spending a few frames holding still, so it takes a few frames of dashing to make up for that wait.
Basic shooter gains a secondary, more powerful charge in 140 frames. Also, all other weapons can be charged in 60 frames, instead of can't charge at all.
Basically required for a few tricks we end up doing. Like skipping auto-scrollers, bosses, and doing much of anything stylish we've been doing.
The Game Resources page is generally supposed to be terse, to the point, and lists out what can be done. Here, as submission text, none of those expectations apply, and so I feel free to talk all day long on these things. As such, expect a worse signal-to-noise ratio.
Floors don't work. There are two versions: Ledge Clip, and Ghost Fall. The effects are similar, fall down places you shouldn't, but the excecutions are different.
The Ledge Clip is done by falling toward a ledge while facing toward it, then once past it, face away from it. The game is looking at exactly one pixel to see if you land, and it's toward your back foot. This depends on the local terrain, but if there aren't any floor tiles below the ledge, you're going down.
Ghost Fall happens when your feet are below the camera edge. For one reason or another, this disables your vertical collision with terrain. This would be fatal, as terrain is the big thing stopping you from falling to the center of the world, but there's two ways to escape without the die ability: Land on an object platform, or use Flame Charge and jump out. As it happens, every place we Ghost Fall, we have such escapes.
Ceilings don't work. Wall jumps insist on moving you upward in part of their ascending, even after hitting a ceiling. All we have to do is wall jump into ceilings, and provided our skull is thicker than the ceiling, we'll get through. Apparently our skulls are 3 tiles deep, since it is impossible to jump through 4 solid tile blocks.
In case the knowledge is important, the solid tile, the walljump tile, and the spikes tile all count as ceilings and will try to stop our ascent. The floor tiles, on the other side of the ceiling we're clipping through, do not count as ceilings, so all those platforms that look 4 tiles wide don't have the floors on the topside count, so we can go through those platforms.
Uncharging don't work. When the player loses control (commonly by taking damage), releasing B won't fire the weapon. When we press B on the frame we regain control, we fire a charged shot without losing our charge.
Note that we only keep the charge if we can fire normally in the situation -- Must have enough ammo for the uncharged shot, must not be in shot cooldown, and must not have existing projectiles that would stop us from firing. Nothing stops us from releasing the stored charge, even stacking multiple charged projectiles is fine, but storing a chain of these shots means we can't do said stacking. Also note that the weapon cooldown part prevents us from chaining a Flame Charge off another Flame Charge.
Weapons don't work. We prove they don't work by charging a weapon, pausing the game on the same frame we release the charge, and selecting a new weapon before leaving the pause screen. This mixes the properties of the original projectile with the animation of the new weapon, the results being... Unusual.
In this run, it's only used for glitching the bubble barrier. Normally, the bubble barrier shouldn't hit anything, but that didn't work, as we've selected a new weapon and gave the barrier a hitbox. Once it hits something, it goes away. Some weapons have it go away on its own. It further proves its not the barrier itself doing the work anyway, as we continue to have 6 HP of ignoring enemy attacks, and we can shoot stuff as well, because having the barrier makes switching and shooting weapons not work.
Minibosses and bosses don't work.
We need to lure an enemy that doesn't work and shoot it at the same moment the miniboss loads. The miniboss will realize it doesn't work and instantly explode, not because it took damage, but because some random enemy died. Only a miniboss should exist in the room, so obviously we destroyed one. Love the logic.
For bosses, we need an enemy to spawn by a door. This only happens during the end stages. We prove the door doesn't work by using a charged laser and freeing our control. After spawning an enemy by the door, they'll continue to exist as the boss prepares to talk. Shoot the random enemy, and the boss will just explode. Not because the random enemy died, but because the boss took damage while at 0 HP. Since the boss is the only one that should exist, obviously the boss took damage when we hit something. Logic is lovely.
The intro boss was smart enough not to keep a door that doesn't work. We can lure an enemy exactly like the usual Miniboss Skip.
Well, the intro has to be done first, and the last stage has to be done last. Beyond that, the remaining eight stages in between can be done in any order, leaving 40320 possible permutaions, not counting leaving a stage halway through.
All stages have a boss at the end which gives a weapon, and some stages have an upgrade as well. There are useful weapons, but before we can get any of them, we need to beat a boss. Which means we deplete 55 HP of a boss on 2 damage per hit, which just takes a long time.
Most weapons aren't too helpful without their charge, as they might save a tiny amount of time on getting past enemies. Body upgrade is useless. Head upgrade is more costly than it's worth with currently known techniques, so that's out too. That leaves Stage 7 and Stage 8 as our potential first choices.
Well, here's what 7 to 8 gets us:
- Spikeball and Leg Upgrade first
- Ground dash for most of Stage 8
- Spikeball to attack next miniboss with 4 damage hits instead of 2
- Charged spikeball for Stage 8 boss
Here's 8 to 7:
- Laser and Arm Upgrade first
- 3 damage hits, instead of 2, against our first boss
- Small tricks: Break capsule lock, Extended Invincibility, Break door lock
- Charged laser for Stage 7 boss
Bosses have a long invincibility time (96 frames per hit), so going from 28 hits to 19 is a huge amount of time by itself. The reason laser upgrade isn't an advantage against Stage 7 miniboss is because Miniboss Skip works there, nullifying the 10 damage advantage of charged lasers. Overall, the benefits of 8 first has proven larger than 7 first through testing these routes. The differences are also visible by summing up the best teams that took 7 to 8 and 8 to 7.
Stage 1 has the fire weapon, which when charged, gives an aerial dash that lets you jump in mid-air after it's done. A speedy 3 pixels/frame in air will save time, and also let us break the autoscrollers in Stages 2 & 4. The aerial mobility makes Stage 1 the clear choice to pick early, right after 7 and 8.
Finally, Stage 4 gives us the bubble weapon, which generates a barrier when we fire its charge. Normally, this barrier prevents us from attacking, but one Charge Swap later, and that drawback is gone, giving it much better use in the remaining stages. Thus our fourth stage is Stage 4, allowing us to pass through many enemies we have a hard time hitting.
The remaining stages don't really offer anything of value, and can be taken in just about any order we please. Actually, in one case, getting extra weapons is bad, because then we'd have to get past those weapons to select one we need. Stage 5 starts off needing an immediate Bubble Barrier, then immediately needing Flame Charge with zero break in between. Having weapons from Stages 2 or 3 (whirlwind or dirt) will get in the way, so Stage 5 next was chosen for that reason alone. 2 frames, but otherwise, the value of these weapons aren't even worth that much.
Finally, each stage has a table. Each stage is broken into segments, and a simple script can tell when these segments load. These numbers are gotten by using this script and comparing every team in DTC6 to what I've got.
Stage 0 (intro)
|Seg 1 (mB)||783||783||783||783||783||783||783||783||783||783|
|Seg 4||2215||1574||1578||1590||1586||1599||1592||1643||1605||1585||Setting up the Boss Skip|
|Seg 5||165||1663||1661||1667||1666||1667||1665||1662||1664||1664||Beauty of Boss Skip|
We sort of have to do the intro first. If we're able to escape the stage somehow, then we'd get to stage select immediately. Unless someone finds a magic way to spawn items with erroneous ID's, and summon up the head upgrade out of thin air, this probably won't happen. I suppose the next best thing is to manipulate a bunch of Heart Tanks, Ammo Tanks, and lives. Not exactly common drops, but the worst of them are 1/32, and six of those fall in our hands without any delay. Then again, we aren't really in need of them.
Team 4's frame war begins here. They did a pretty good job, just a few frames here and there, but then after DTC6 we found we can skip the intro boss and save a bunch of seconds.
s0 Segment 0
Right segment. T4 base. Improvements here are razor thin.
All teams likely have spent a fairly significant amount of time on that first jump. I know one team member who kept wanting to believe it's possible to make the gap without a wall jump. It seems mathematically tailored to be out of reach, but only just. I can explain to you this entire paragraph is devoted on the subject of this one jump, the results are disappointing. We need position x376 to land there, yet jumping as late as we can and turning around to move the important pixel further, the best we've got is x374. We are exactly one frame of movement away from meeting it. Again, we are denied this one jump with mathematical precision. If you think you have better ideas than the collective from the Dream Team Contest #6, then you are free to try. A grand total of 6 frames are lost from this wall jump, 5 from the wall itself and 1 from the landing lag.
As for the rest of the segment, as our movement isn't complicated yet, we can simply count all the frames we do not travel 2 pixels to the right. Including the first wall jump, we have 57 frames worth of not moving right. 10 frames are lost to landing lag, 5 from the first wall jump of disappointment, 16 frames at first ladder, 7 frames at second ladder, 8 frames at wall jump after second ladder, and 11 from the final wall jump.
What's the harm in collecting a bunch of items in the meantime? May as well pile up on items we'll never use like lives, heart tanks, and ammo tanks. Well, okay. We do use a few ammo tanks in the run, but since we can't really do much else with these enemies, may as well show our luck without spending a single frame.
s0 Segment 1
Miniboss segment. T4 was probably used as a base. Hardly matters.
Minibosses have 25 HP, 60 frames of invincibility per hit, and annoys us with its time sink. 13 hits is the best we can get, it is trivial to slam the miniboss with charged shot the instant it loads, and charged shots need 61 frames to charge, 1 with B released and 60 frames of actual charging. The strategy is extremely straight-forward, every team without exception cleared it in 783 frames.
Miniboss Skip is possible. Problem is, while the skip itself is fast, the set-up takes far more time than the skip saves. Yes, the 13 hits take a while, but getting the skip ready takes more time than that.
The usual method is that we lure an enemy to the miniboss and shoot that. Problem is the only one who can fly all the way there is far enough away to take more than 720 frames, meaning the skip is possible, but takes longer to do than just blasting the miniboss.
The sneaky method feels clever, but ultimately is far too slow. By the time it's done, we could have already done Intro. Suicide the three starting spare lives. For your fourth, die to the miniboss. Start again, destroy the first enemy, and get to a specific position. The Segment Skip moves us to the miniboss segment, preserving our position. It's a special spot where the trigger for loading the next segment is, normally out of reach in the miniboss segment. So we instantly get to the segment after. Shame it involves dieing no fewer than four times and getting a Game Over. .bk2 of the whole thing.
s0 Segment 2
Right segment. T4 base, small improvements.
This is the short trek between the miniboss and the spikes we crawl over. We've taken damage at the miniboss as late as we can, so that we can traverse the spikes damage free. Just enough invincibility time to make the trek, too. In fact, the timing is so sensitive we need the 2-frame rule to line up, this one affecting our invincibility time after taking damage. There are enemies below us as well, but we can avoid them, so no need to take damage on the spikes.
s0 Segment 3
Down segment. T4 base, somehow we fell a frame faster?
This is what Ghost Fall looks like. Because of our rightward position relative to the camera, we're able to fall to the bottom of the camera before triggering this segment. Since our feet are below the camera, terrain no longer stops us, which is good because we need to get down to the bottom quickly.
As luck would have it, while terrain can't stop us, and we don't have Flame Charge to save ourselves, there are object platforms we can stand on. Since they are not terrain, we can collide with these platforms, and actually stand on something. The alternative without this platform is we fall and die, so it's a good thing there's one to land on here.
We also travel to the left, because this is a time we can adjust the camera, moving it further right, and we can't move further right than the platform we need below. The camera is what determines when segments load, and ultimately we need it to reach the boss room. Or rather, with the Boss Skip, get an important enemy to load and going in the right direction sooner. This is just one example out of many where the camera is more important than the player's position.
After landing on the platform, the camera hasn't scrolled far enough down. We need to jump high enough so that our fall will then get enough scrolling done. Not too high, because then it takes too long to start the scrolling, and not too short, because then we wouldn't scroll enough. Just so happens there's a "Goldilocks" zone where things are just right for jump height. We leave the segment in screen position X 80.
s0 Segment 4
Right segment. T4 base, with a different strategy in the second half.
Ah, this one doesn't have a clear and obvious route through the big spikes pit. I recall going through a few iterations before Exonym or Pirate_Sephiroth figured out the platform above delays us less than traversing the spikes straight through. Landing lag is mitigated in a few ways ahead, one of which includes almost falling past an object platform, and snapping to its top. That "almost fall" isn't just entertainment -- The camera scrolling there meant it wouldn't scroll when we landed on the terrain up there, thereby saving us from the landing lag.
Eventually, we come to a stop in this segment. Today's excuse here is that we are waiting for an enemy, which is why we're slower than other teams. Blame the boss's big mouth and thick skull. At least doing things this way lets us avoid his speech and chipping through 11 hits at 96 frames invincibility per hit. In the meantime, uh... We're lazy so far. I hope this changes.
s0 Segment 5
Boss segment. It is unique from any team.
We just sort of jump up a wall, shoot something off-screen, then the boss is so impressed he just exploded. I accept. I thought the optimal hit pattern was 5-5-1 between invincibility attacks, when it was just plain ol' 1 the entire time.
This is the Boss Skip. Actually, a simpler version because there's no door. This particular skip worked because boss rooms do not clear enemies from the screen, normally that's the door's job. Without the door, we can lure an enemy to the boss. Game does not handle this well.
Reaching the boss segment sets the game into Boss Mode. Striking an enemy hitbox in Boss Mode will do damage to a special boss HP stat. Boss hitbox is disabled during dialog, so we can't sneak a shot at him. Normal enemies don't disable their own hitbox, so we can hit them. Oh, we're in Boss Mode, and we hit something? Deplete boss HP. The sneaky bit here is that boss HP is not loaded until after dialog, so I just damaged his 0 HP, which terminates the boss.
Would have said the boss was so impressed he literally exploded, but most people would assume I'm using a rather mutated meaning of 'literally' these days. I mean the firm, definitely not figuratively in any way, shape, or form, 'literally' that people knew it as it was back in the day, when 'literally' was real 'literally', not the fake 'literally' today. Why did I rant all this just now?
Notice: A few teams did not take this stage first (marked with !!). They went to Stage 7 first and got the Leg Upgrade and spikeball, which lets them clear this stage faster. Their times will not be highlighted on any instance they are faster. The Secret Team (ST) went to Stage 1 first, but at least they don't have the Leg Upgrade, so their times are more easily compared.
|Me||#||T4||T3||T7||T5||ST||T6 !!||T11 !!||T2 !!||T1 !!||Excuses|
|Seg 2||803||803||814||834||836||802||631||657||687||647||Secret Team wields fire|
|Seg 4||133||121||121||133||131||133||103||96||91||93||Took ladder (-15 Seg3, +12 Seg4)|
|Seg 5 (mB)||790||790||790||790||790||550||430||430||455||551|
|Seg 7||655||661||671||674||975||687||665||655||668||897||Arm Upgrade here|
|Seg 8||131||131||133||132||141||92||117||116||137||105||ST is now on fire|
|Seg 10||389||386||394||386||404||336||345||323||365||314||Triggered segment early|
|Seg 11||250||246||246||248||268||290||239||271||256||271||Waiting frame rule here|
We get laser and Arm Upgrade here. Important this is first because we slay our first boss much more quickly with the Arm Upgrade immediately. We don't have much of anything to attack the stage with, though.
Team 4 has frame warred this stage really well. We've saved 16 frames over them, except we lose 4 of them to a frame rule. There wasn't a lot we could save over them.
s8 Segment 0
Right segment. T4's run.
Without complicated dash stuff, there isn't really much to do. There aren't even a lot of enemies in the way, so there isn't much to do with handling them either. Team 4 beats Team 3 because they found a spot to avoid landing lag, making sure to land at the end of the platform.
s8 Segment 1
Up segment. T4's run.
The first of our ceiling clips. This one is trivial because it is only one tile thick. Well, two, but floor tiles don't count as ceiling. We'll be bashing our heads through more ceilings, and many of them are much thicker, don't worry.
Note that wall jumps move us upward at 6 pixels/frame, and normal jumps or ladder jumps move us at 1 pixel the first frame, 3 for the next several, and 4 for the remainder. As much as I'd like to wall jump more at the ladders, the game has solid tiles by ladders, rather than walljump tiles, and so my only course of action is to use the slower ladder to ascend.
Team 4 carries on the extreme standard in that I simply can't improve on their early stages in many places. Then again, at least it isn't just Team 4 who got the fastest known route, two other teams managed to copy the time. I'm merely copying the time once more.
s8 Segment 2
Right segment. T4's run.
Oh, this is a fun one. No landing lag on our first jump around the first enemy here, camera won't scroll with us that low. Jumping up on top of the ladder ahead is actually just a bit out of reach of a full jump, but we press Down to climb on the ladder. This lets us grab on to the ladder from a much higher position than Up, as the game is looking at our feet instead of our head for a ladder tile.
Jumping around the green guy is another tricky part. We have to damage him, as part of his ouch animation disables his hitbox. The precise timing enables our hitbox to clip just past his hitbox corner, and we're past his head without taking damage. He's got 5 HP, and it ain't easy knocking that down swiftly, but our last bullet is enough and we can end our jump sooner, falling past where he once stood and landing early enough to avoid slowing down.
That was the precise part. The rest is just jumping across the remaining platforms. Not much we can about the remaining enemies, they sort of also have 5 HP each. Note that we can't kill 5 HP in less than 3 hits for the moment, and between hits #1 and #3, we're moving over 80 pixels. We do destroy one more enemy, but that one was in a decidedly optimal position to be attacked, it still had 5 HP to start.
s8 Segment 3
Down segment. T7 inspired this improvement.
Normally, we would take a Z path, going to the right and climbing down past some obstacles. Turns out, there's a ledge we can clip past instead. We travel far enough right to trigger this segment, turn around, and fall past the spikes. This is the Ledge Clip, something you'll see more of when floors become inconvenient like now. Essentially, the important landing pixel moves 8 pixels depending on where you're facing, and the game doesn't extend the spikes under the floor, so after falling a bit, we turn around and sneak the important pixel under the floor.
Team 4 took another Ledge Clip to the right, but was slowed down due to taking damage. Team 7 took the ladder, but did not get a head bonk against a solid tile by the ladder. We take the ladder, get the headbonk, and overall is just slightly faster than the floor clip over to the right. We have to headbonk on a solid tile down a ways, as the ladder tiles do not extend to the left wall.
Of note, we need to be far enough left to trigger Segment 4. If we Ledge Clip too far to the right, we're still in Segment 3 and unable to advance, with no sign of the Miniboss. The Ledge Clip has us going to the left anyway to trigger this, which is part of the reason why the ladder route wins by a few frames.
s8 Segment 4
Save segment. Continuing T7's inspiration.
Well, we trigger this segment much earlier than Team 4, which is why the frame count is skewed a little higher for this segment. We're still 3 frames faster going this way, however.
If you're expecting more for a segment that's rather empty, I suggest you prepare for disappointment.
s8 Segment 5
Miniboss segment. Hardly matters whose team.
Just like the intro miniboss, we're stuck with basic shots and no way to really speed things up. We just plink away 25 HP on 2-damage shots with 60 frames of wait per hit across 13 shots. There's just no complexity here.
While the Miniboss Skip does work on some other stages, before we get damaging weapons, the advantage of a 4-damage weapon here does not overpower the advantage of fighting the first boss with the Arm Upgrade. Even Stage 7 has all the advantages it can get: Miniboss Skip, Leg Upgrade, a 4-damage weapon, and even a good charged shot when we get the Arm Upgrade for the boss, and even those are not enough to be worth the difference of 9 hits on the boss (28 hits down to 19) with the Arm Upgrade. Boss invincibility time is vicious.
s8 Segment 6
Right segment. We, uh... Walk the whole time.
I don't rightly see how anyone can walk less optimally. It is difficult to screw up "hold right" when there are no obstacles of any sort. I suspect a realtime run can hold right for justice and still get 122 frames without any practice. The only reason other teams got through faster is because they got Leg Upgrade to dash with, but I feel like a broken record talking about the stage route at this point.
s8 Segment 7
Down segment. T4 was improved on.
Arm Upgrade is here. At last, another thing Team 4 wasn't perfect on! The headache-inducing noisy enemy wasn't handled perfectly, causing Team 4 to wait a few frames for it. I figured out how to fall better after DTC6, so just managed to get there in time to shoot it again.
Ladder headbonk again, thanks to solid tiles. This segment would have us travel in another Z path, but in a shocking twist, there's another Ledge Clip below the ladder! The spikes and floor arrangement is very similar to before, so we just do that and down we go. Point of note, it is possible to time the Ledge Clip to stop ourselves from moving left with floor tiles, but spike tiles do not restrict horizontal movement, so there's nothing we can do to stop from moving 2 pixels left.
s8 Segment 8
Right segment. T4's run.
There's a bit of an animation frame rule going on here. We're on the good side of that frame rule in this run, so we can avoid damage from the bullet that enemy shoots. I mean, when we jump back up the spikes, that bullet lingers for a while!
s8 Segment 9
Up segment. T4 as base, it is improved on.
For some reason, the first left side ledge does not have walljump tiles, which is why we climbed up the wall as high as we did there. All other ledges have walljump tiles, so we can just use those and travel sooner to the ledge, with something to catch us there.
There are supposed to be enemies nearby the left walls. Carefully timed jumps move the spawn zones just right so that the spawn points move through the corner spawn zone without us moving upward through that corner zone. If you don't understand, we make a magic jump at a magic spot. Team 4 knew about this for the lower enemy, but didn't figure it out for the upper enemy. It wasn't me who manipulated the spawn zones to prevent the upper enemy from ever existing. Note we can't do it on the right-side enemies, as the spawning zones on the right side are extremely thick, and is still easily within that region when we finish even our largest jump.
The Secret Team brought a flamethrower. Now that they have the Arm Upgrade, they can use it with the flame weapon to position themselves nicely for skipping through the sloped platforms. It is not possible to ceiling clip through them normally, the Flame Charge gives a better position to skip through, so I can't replicate what they do in the route I chose. The solid tiles are 4 blocks thick (not counting floor tiles, which aren't ceilings), our skulls are not that thick.
s8 Segment 10
Right segment. T4 as base.
We jump. This scrolls the camera up, loading this segment earlier. Since we're in a new segment, the camera refuses to scroll back down to the slope we're returning to. Our walk up the rest of the slope does not move the camera upward, and this is enough to prevent a spawn.
Yes, that jump as we fire our charged shot was necessary. When we land, we're a few pixels further down than walking there, so our shot could hit the enemy earlier. Because we were moving to a slope, our landing does not delay us a frame, slope tiles do not restrict horizontal movement.
s8 Segment 11
Save segment. T4 as base.
There are spikes down there. We jump on platform. ... I have a feeling I didn't need to say that.
We lose four frames here thanks to a frame rule. Although finding 4 frames in this stage would let us beat it, this will also be true if we can find 4 frames in the intro, as the intro is no longer bound by the dialog frame rule. Unfortunately, until we find these four frames, we're stuck waiting out dialog for the boss, and I'd rather keep sync by making the delay at the gate here. It help that this stage is completely independent of item drops, so there should be no sync troubles.
s8 Segment 12
Boss segment. T4's run.
This is the most interesting battle. Most of them are just "do enough overflow damage, then keep up with invincibility timer." For anyone who didn't route for Arm Upgrade first, it's "shoot boss 28 times with 60 frames charging against 96 frames invincibility," as trivial as it sounds. But with Arm Upgrade and only the basic buster, the stategy then becomes "use doppler effect and Charge Storage to optimize 140 frames charging against 96 frames invincibility."
Bosses have 11 main HP and 5 sub-HP per main HP. In total, that means I need to blast away 55 HP. Since there's no chance at overflow damage with 3 damage maximum, it's just counting how many 3-damage shots it will take to remove 55 HP. The answer is 19, with 2 damage to spare. 140 frames of charging is a long time, longer than the 96 frames invincibility by a full 44 frames, plus one more frame because one of them B is released to shoot. 60 frames of charging (okay, 61 because B released to shoot) is well under, and since I have 2 damage to spare, I can replace two 3-damage shots with a pair of faster 2-damage ones, so I can shoot 17 full blasts and 2 decent blasts.
The movement necessary keeps me very busy, at least. Charge Storage requires I take damage, but enables me to shoot a 3-damage charge while keeping fully charged. This is done by releasing B in the moment you lose control, and pressing it the instant control is regained. We lose control thanks to dialog, so we get a free Charge Storage at the very start. Since we haven't been damaged yet, we walk forward to take a hit and repeat the glitch. After that, we generally walk back so we can fire our shot as early as possible, giving it some travel time before hitting the boss, and letting us charge our next shot sooner. The "doppler effect" I was talking about earlier.
Although Charge Storage lets us keep the 140 frames charge while shooting such lovely 3-damage shots, it takes 96 frames for our own invincibility to wear off, on top of the 16 frames of ouch animation to wait out, for 112 frames between shots. Sure, this is faster than 140 frames, but the doppler effect makes up for enough of the 140 frames that it still saves time, and lets us take a hit for Charge Storage without the 112 frames wait. Besides, too much damage, and we'd need a Heart Tank, costing 5 frames and is so much less stylish.
The 60 frames of 2-damage charged shots are used at the moments where it is most awkward to take damage for Charge Storage. Yes, there's even a bit of strategy to time the two we have available.
A number of teams took this stage first (marked with !!), while we took it second. Without the laser or Arm Upgrade, times made by the teams that went here first are difficult to compare.
|Me||#||T4||T3||T7||T5||ST||T6 !!||T11 !!||T2 !!||T1 !!||Excuses|
|Seg 0||954||954||965||982||986||837||1011||994||1011||996||ST still on fire|
|Seg 1||816||814||859||1035||1095||1174||1076||1047||1097||1099||Going right early|
|Seg 2||573||501||509||565||554||461||554||551||561||551||Waiting for flying foe|
|Seg 3||157||106||106||200||111||92||107||110||129||101||All for Miniboss Skip|
|Seg 4 (mB)||69||236||253||69||275||244||830||793||867||550||Miniboss Skip!|
|Seg 7||623||624||687||693||699||571||711||693||717||685||ST is burnin'|
|Seg 10||355||354||367||363||352||372||362||361||355||370||1 frame delayed for frame rule|
|Seg 11||1414||1411||1421||1569||1565||1663||3276||3276||3285||2360||Dialog frame rule|
Big reason to come here is the leg upgrade. Stage 1 has a lot of flat ground to cover, so although Flame Charge can probably save time here in Stage 7, the ground dashing likely helps Stage 1 more. As much as 3 pixels/frame is an upgrade to 2, the Arm Upgrade's benefit against the first boss encountered, as well as the side benefits in Stage 7, overwhelms the advantage that dashing gives. The Secret Team is the only one who took Stage 1 without Leg Upgrade, and Stage 7 with both 1 & 8 already done, so their times made it easier to compare routes.
Team 4's frame warring continues through this stage, but they did miss a useful skip, so this run has an advantage over them.
s7 Segment 0
Right segment. T4's run.
This segment has a few acrobatics, which makes it a bit more interesting than just run right for justice. Many enemies have 5 HP, so it helps to have a 4-damage laser shot to defeat them. There are a few hidden spike enemies along the way, if you're curious about our large jumps and awkward ledge landings.
The chamber of spikes is navigated by jumping through the middle platform. This works because the game is only looking at one pixel for horizontal collision, at the player's feet. Well, there is another pixel higher up that's used for walljump tiles, and it would also block us, but we jump late enough that we miss the walljump tiles. We also had to jump early enough so the back of our head didn't interact with solid or walljump tiles in order to jump as high as we did. And of course, don't jump too high, or else our feet gets embedded into solid tiles and we stop.
s7 Segment 1
Up segment. T4 used as base, end adjusted from T7's inspiration.
This is where the Leg Upgrade is. For some reason, we've assumed the long-term gains of fetching it outweighs the short-term gains of ignoring it. Note we used a charged laser on the capsule. Basically there's a special address (15F0) that gets written to whenever the game wants to lock the player control. Part of the laser's effects are to lock the player in place, and so does the capsule. Since these things don't really account for each other, once the laser object finishes, it unlocks us. We're free to move as we please. As a result, we steal the capsule glow, taking it with us. We have to switch weapons eventually, and the glow ends when we do, so we switch as late as we can.
We ascend. Mostly just copied Team 4's run straight over. They did frame war the snap out of this stage. Among their tricks included wall jumping from the "wrong side" of the walls, using the fact they can jump high enough into a number of the ceilings. The center platform is one example where a "failed" ceiling clip was done so they can reach the wall on the right, just high enough to grab on the non-locking portion of the ladder, saving from having to make an extra jump or waiting out the ladder climb.
The end of this segment is changed up from Team 4. Their route is to move the camera more to fight the miniboss. This route we're taking is to get the player to the right sooner, with no concern for the camera. Our Miniboss Skip set-up begins at this corner.
s7 Segment 2
Right segment. Inspired by T7.
Normally, we'd be further left relative to camera, as for most of the run, we're very concerned with getting the camera forward quickly, as it's the thing that triggers segments. This is a rare instance where the player's position is more important, because it means an important enemy moves in the right direction sooner. Until then, we manipulate more laser ammo. Fighting 5 HP enemies with only one laser shot left doesn't help, so of course we manipulate lots more ammo in a hurry.
Once we pass the flying enemy, our haste to the right no longer applies, as we're now waiting for something to drift into position. We have all this time, so charge a laser, drop into spikes, and fire it. It may look funny that we get bored and leave while the laser blast is still going, but doing this means we get Extended Invincibility. That address 15F0 is to blame again, with the damage freeing us first this time. Since we moved the laser off-screen, it didn't get a chance to finish and remove the invincibility it gave us before, so we enjoy the remainder of the stage with
invincibility cheats this glitch protecting us.
s7 Segment 3
Save segment. Inspired by Team 7.
The process of setting up Extended Invincibility was still under the time it took to wait out the flying enemy, but I must note we didn't have a lot of frames left to spare. We still get there in time to skip the miniboss on the first frame possible.
s7 Segment 4
Miniboss segment. Inspired by Team 7.
Good job. All you get to see is me shooting a laser to the left, and explosions from out of nowhere. The game didn't get a chance to unload the enemy before we shot and destroyed it, and this apparently happened after the game set miniboss mode. And since we destroyed something in miniboss mode, it had to be the miniboss, so on we go. This miniboss was big, and green, had this dragon head, and turrets that didn't look right when viewed with my peripheral vision. I should know these things!
s7 Segment 5
Right segment. T4 used as base.
Well, we're invincible. We already start its abuse by going straight through enemies whose deaths cause lag. Might as well use the inexhaustable resource you've got to the fullest extent, by passing straight through enemies right away.
s7 Segment 6
Up segment. T4 used as base.
Team 4 sets up Extended Invincibility here, but that's because they had to fight the miniboss. Extended Invincibility ends the next time you use a charged laser, as then it would finally have a chance to remove the invincibility flag it set earlier. We skip the miniboss, so that's why we set it up earlier than they did, as we no longer need a charged laser at the miniboss.
There is one wall jump here that looks bad. When we jump on a middle platform to reach a ledge above, the highest jump delays us from making the wall jump, but hold that A button for just one fewer frame, and the mid jump doesn't even latch us onto the wall. It's just perfectly positioned to make us look bad, I tell you!
The final jumps require a particular pixel positioning to make sure we don't jump too high above that final ladder. Dashing early helps to save more time.
s7 Segment 7
Right segment. T4 used as base.
The big reason for Extended Invincibility. The floating spike balls here are decidedly in the way and would slow everyone down massively. Our
invincibility cheats Extended Invincibility means we just pass right through unharmed.
Early in the segment, I manipulate the +5 Ammo like a number of previous attempts. It did not go well, the ammo went into the wrong spot by exactly one pixel. Come on, man! My laser isn't full! At least I'm able to refill just fine from later enemies, so it's all good. Not like I need every drop, so just show off an "unlucky" drop and continue on.
Flame Charge would definitely benefit this segment, as it is mostly open space with short pillars to jump on. Very little ground for dashing on. Whenever it would save a pixel, we do turnarounds on these pillars. I've set up my position to be pixel perfect so that the segment goes smoothly.
s7 Segment 8
Down segment. T4's run.
You know those Ledge Clips in Stage 8? You get to see more of those here. It may look kind of like Ghost Fall, with how we never land on anything, but we're not at the bottom of the camera. Perhaps a good thing, as there's no object platform down there, and we don't have Flame Charge yet.
We have to shoot a few things on the way. This is to stop the enemies from lagging up the place with more sprites. We still charge in time for an easy refill, at least.
We enter the next segment as far left as we go. It may not be Screen X position 80, but the spot we're in (91) still ain't bad.
s7 Segment 9
Right segment. T4 used as base.
Have I mentioned how nice Extended Invincibility is? It's still going, because there's no timer to shut it off. We also have an opportunity to destroy an enemy with our 3-damage charged shot and pick up an item drop. With all the timing changes in improvements, this particular enemy has always been compliant in giving +5 Ammo.
For the most part, this segment is just navigating the terrain optimally. We're invincible, so it's not like we care about enemies. Land on the save platform ahead for more dashing, which is still enough to make up for the 3 frames of lag.
s7 Segment 10
Save segment. T4 used as base.
Look at those spikes. They are pointy. We jump past them. Not that it matters, but Extended Invincibility gives no protection against terrain spikes. Otherwise, the Extended Invincibility wouldn't have worked to begin with, as taking damage is necessary to escape the laser while it's still going. Yeah, sorry about this idle chatter.
Our laser is so powerful, it breaks our control lock on the door! Like the capsule, we have multiple things messing with address 15F0, though this time it saves only a few frames by getting the first attack against the boss sooner. Not as though we have anything better to do with our laser ammo. Note that my Extended Invincibility ends here, because I just fired a charged laser, but it isn't needed for the boss.
s7 Segment 11
Boss segment. T4's run.
Bosses have 11 main HP and each main HP has 5 sub-HP. Yes, this means 55 HP total. However, while the charged laser does 10 damage per hit, you can't remove more than one main HP on any hit. Still, the game keeps track of total damage dealt, and although the overflow damage can't actually remove more than one HP at a time, later hits will make use of the overflow damage and a weak attack or even the 0-damage Muzzle Flash will be able to remove one main HP on the next attack. Basically, the strategy boils down to doing enough damage up-front so that later attacks just need to make the hits rather than damage. You can't defeat bosses faster than 11 hits, no matter the damage.
For this particular battle, the only weapon we've picked up so far is the laser. I have observed getting two hits with the laser, but it is so difficult and unwieldy to do, I simply gave up the attempts and went with single hits. Even if it weren't for the lock break at the door, 11 ammo will not suffice for laser only, so I spend an Ammo Tank during the battle. Other than that, I made sure to get in hits whenever the invincibility timer runs out and all that stuff. Things the leading teams did, you know? We end with negative laser ammo thanks to Charge Storage.
|Seg 1||584||584||583||604||588||620||625||664||634||614||Fall cycle|
|Seg 3 (mB)||210||210||210||210||242||270||302||515||303||782|
|Seg 4||760||760||758||794||818||812||860||943||789||948||Worth checking|
Our third stage, we finally pick up the flame weapon. We need it to break auto-scrollers, and a number of sensible moving of right through the air. There's a lot of flat ground to dash on here, which the Leg Upgrade from Stage 7 helps us here more than Flame Charge helps in Stage 7. Apparently, most enemies have 2 or 3 HP here. Rather low compared to the 5 we frequently met in the first two stages we went to.
Note that this stage is where Team 4's serious frame war comes to an end. I could only pull out one and a half frame rules from under them, but then they lose their optimization edge once Flame Charge complicates routes a lot more.
s1 Segment 0
Right segment. T4 used as base.
Right from the start, repeated dashes are the way to go. Even with the jumps, there are a few optimizations here to do.
Dashing requires a double tap to, well, dash. When we jump out of a dash, we are stopped for one frame. The double tap has a 20 frame timer between taps, so we can tap once, do a really fast jump, then finish with our second tap. For a tap to work, we have to release left or right for a frame, which also delays us. By making a tap on the frame we jump out of a dash, we overlap some of the zero speed frames, so that it wastes less time. Then we only need to release for one more frame when we land, as opposed to needing the two taps then. This is done for short ledges, but higher ledges are too high and our 20 frame timer expires before we land.
The higher ones aren't perfectly straight-forward either. Sure, we have to jump higher, and we have to make the double tap to dash when we land. However, the important pixel the game uses when landing depends on our facing, always toward the back foot. The process of facing left requires us to move one frame in that direction, so although we lose 2 pixels, we can dash 3 frames earlier, for a net gain of 1 pixel.
At the spot where we take damage from a fire, there's another fire we managed to get past. Initially, it is faster to just shoot that short platform and keep dashing, making only one jump, but we took the path we did because otherwise our invincibility runs out a few frames too early and we take damage at the second fire. In fact, the timing is so tight our invincibillity wears off and we're relying on the fact our hitbox is disabled when we're on the invisible frames of the invincibility blinks, for just those last few frames.
s1 Segment 1
Down segment. T4's run.
We enter this segment by dashing straight off the ledge. Yes, I'm well aware of using jumps to give ourselves vertical momentum by the time we reach the edge, and is a common trick in many other games. Yes, I do those jumps on some ledges in this game, in fact. The reason we don't here is because the dash speed makes up for it in fall time, and jumping does not carry dash speed through it. If we were going to a left-side ledge, then I would have jumped, as ending a dash checks the right-side pixel for whether we're over air to begin our fall. We earlier shot a glitched basic bullet to hit an enemy, but destoying it meant lag, and all we needed was to delay the enemy anyway, so it lives.
There is a plaform with a ladder in the way. We must take the ladder. We also tap left as we land, so to get a dash ready, and do a Ladder Dash after climbing down. It's a rather mundane use of our first Ladder Dash, as a convenient way to drop out of the ladder earlier, or without the upward bump from jumping off. Keep in mind the ladder locks us in climbing for part of the way.
Aside from that, it's just a straight fall down. There's a spot that needs a little precision and a turnaround to avoid hitting another platform, but otherwise, nothing stops gravity from doing its work. There really isn't anything I can do to speed up gravity.
s1 Segment 2
Save segment. T4's run.
There's a little finesse needed in picking a spot to enter this segment. For one thing, there's an enemy in the way that is best dealt with by the 3-damage charge we've been holding for a while. For another, the upcoming miniboss wouldn't want us too close, or our Muzzle Flash gets in the first hit, and then we're left frowning at the 0 damage it did as opposed to the 10 the charged laser could have done.
s1 Segment 3
Miniboss segment. T4's run.
The instant the miniboss loads, we fire the laser. That's 20/25 right there in two hits. We finish with charged Spikeball as that's the only other weapon we've got that does enough damage, and this one doesn't lock us in place. We'll swat 5 HP with these 8 damage attacks like this a few more times. It's a pretty fast win, and obvious enough for three teams to match the time.
s1 Segment 4
Right segment. T4 used as base.
We don't plant ourselves on the right edge of the screen. The biggest reason being the timing of the flames. By selecting this spot, we can mostly just charge straight through with little delay. Otherwise, we take damage multiple times, and that won't help us. Has a side-effect of making it easier to deal with enemies in this segment, too.
Probably should look at T3, maybe use them as base.
For manipulation, we primarily use basic shots, but switching weapons with basic shots on-screen has a strange effect on their hitboxes, and also makes them look... uh, colorful. Equip the spikeball, and they are absolutely huge. Mutating these hitboxes like this makes it much easier to manipulate better ammo drops from enemies on the way. Equip laser, and the basic shots become piercing. Shame they still only deal 1 damage.
s1 Segment 5
Up segment. T4 used as base.
We start with charged spikeball, 8 damage, against a bat, 2 HP. If there are weapons with a better attack angle, I would use them, but I don't have them, so Charged Spikeball it is. The next enemy on the wall is taken out with a regular spikeball shortly after. Actually, we were delayed by this enemy a bit, so that our regular spikeball can reach, but it's better than wasting 16 frames of ouch. Note that a spikeball travels at 4 pixels/frame, but we scale walls at 6 pixels/frame, and this is about the only way to get the spikeball to unload off-screen.
There's a fairly open area with ladders above. Normally, we'd have to climb all the way up that ladder only to fall back down to the next platform. On the ground, I tap left to begin the Dash Ready timer, jump up to the ladder, catch on, then dash left from there before the timer ends. Our second Ladder Dash is a lot more glorious than the first, as now you see us fly through the air. Even nicer, this Ladder Dash lets us jump out of it, so we effectively get a mid-air jump from it, and it's enough to make the shortcut to the platform. Yes, it is glorious.
Not sure what's up with the last wall jump. Apparently I got better height from exact same input as T4. 4-frame animation rule? We use charged spikeball to swat another 2 HP enemy, and go down to 1 ammo left. We need spikeballs for boss, but at least manipulation for ammo is viable with mutating our basic bullets by simply swapping weapons.
s1 Segment 6
Right segment. T4 used as base.
First thing we do, we dash straight into the ledge ahead, then jump apparently rather late as we're blocked by it. This seemingly sub-optimal jump was to prevent a flame from spawning, and trust me, we can't avoid that flame if it spawned. Later, there's another flame which spawns just fine, so refer to that one to see how much trouble it would be to dodge it.
There's not a lot to this segment, really. Mostly just precision jumping, manipulating item drops with glitched basic bullets, and getting ideas on self-immolation by taking damage on the fires for when we finally get this stage's weapon. Trust me, once we start setting ourselves on fire with the actual weapon, it will be a whole lot more effective.
s1 Segment 7
Basically this really short segment starts after we pass the save platform. Like other such segments, I don't have much to say. Oh, the door didn't seem to spawn right away. Apparently, it does have special spawning code that spawns it outside the usual regions if it hasn't already spawned.
Charge Laser on door. Since this is the second time you're seeing it, I will remind you it's to break the lock on the door. No, it's not that the door refuses to open without a key, but rather that it locks our control, and we're breaking that. The Laser Lock Break is less interesting now that it is used almost exclusively to regain control on these doors for the rest of the run, but it's still frames to save.
s1 Segment 8
We smack boss with charged Spikeballs before moving on to weaker weaponry. This is the strategy you will see for every boss for the remainder of the run. Charged Spikeball does 8 damage per hit, and usually gets in 2 hits with optimal timing. Charge Storage off of dialog, and we typically open with 3 or 4 hits for 24 or 32 damage right off the bat. Since we lack weapons, I do one more Charge Storage by taking damage (twice, to use invincibility to help position the second hurt), fire one more charged Spikeball, and drop straight to negative ammo for 6 hits at 48 damage. Honestly, at this point, I sort of am kind of doing that for show.
Then again, no matter how much damage you do, each hit can only remove zero or one bar of boss HP, out of 11 each boss gets. Thanks to all the overflow damage, my basic buster is enough for the remaining 5 hits we need to do with 7 HP of damage to deal. Two charged shots is all that is needed at that point. At no point does the boss spend one frame vulnerable without a projectile lodged in its hitbox. At this point, I'm sure you see why the above strategy is repeated basically forever. At that point, we just try to keep lag low and still beat those bosses.
|Seg 0||231||232||234||232||234||233||233||239||215||246||T1 took auto-scroller|
|Seg 2 (mB)||245||245||264||376||374||286||382||399||802||456|
|Seg 3||151||152||155||92||90||482||461||218||477||144||Setting up Ghost Fall|
This stage has an auto-scroller, which would waste thousands of frames scrolling slowly. Thanks to doing Stages 8 & 1 first, we can bypass all that. The weapon from Stage 4 gives us a barrier when charged, offering us the last major timesaver when we can mostly ignore enemies.
Team 4 still leads DTC6 here, but this time our advancement over them is general optimizations rather than any skips they didn't do. Our gap between this run and Team 4 is wider than the gap between Teams 4 & 3, the beginnings of that edge becoming lost.
s4 Segment 0
Right segment. T4 used as base.
Now that we have the flame weapon, the first thing we do is catch fire. In an underwater stage. We are not letting this self-immolation thing be stopped by common sense.
I've got things to explain. The stage starts off with the auto-scroller vehicle, which is supposed to be mandatory, but Flame Charge is enough to let us skip that. Flame Charge itself lasts around 64 frames, sends you forward through air (or fluids) at 3 pixels/frame, and has a few useful glitches. With the lack of terrain in some places, it is supposed to be impossible to traverse this stage normally, but that clearly didn't stop us.
We're faster than Team 4 due to the fact we triggered our Flame Charge with our dash counter at 39, not 40. When the Flame Charge ends, it continues whatever state we had, and the ground dash at 39 is just another frame of 3 pixels movement. The ground dash at 40 is a stopped frame. Team 4 missed this detail. Meaning I, FatRatKnight, should be embarrassed about it.
s4 Segment 1
Auto-scroller segment. At least, it's supposed to be. T4 used as base.
This is a long, long segment. Seriously, over 1700 frames with dashing and Flame Charge? That's longer than even most of our boss fights! Please, do not try to imagine what it's like at 1 pixel/frame, for when one does take the auto-scroller vehicle.
Also of note, this segment, and the other "auto-scroller" later in the stage, have broken slopes. Not seriously broken, just that most were misplaced one tile down. This is enough to stop us from just simply dashing on the slope and onto flat floor. If you're curious as to why the short jumps at the top of some slopes, this is the reason. In one case, the actual slope is backwards from what is seen there, and trying to walk on it in the visual downhill direction will have it look like the player is slowly rising to the air before suddenly snapping downward a bunch of pixels to the slope below, then rise again from there.
The first broken slope, a downhill, was taken just fine without a jump using a Flame Charge. This is because a Flame Charge interacting with a floor tile will snap the player to the top of the floor tile, and if this was done from the ground state, will continue moving forward. Jumps are slow, we like ground dashing more, and 5 damage is sure convenient for enemies in the way. We have taken damage twice in this segment, and did a Charge Storage off them both times. The lack of decent ground to dash on generally meant the extra Flame Charge we got out of the deal probably makes up for the slow damage.
This segment is so long and complicated it's hard to come up with any other particular points to talk about. It's just not fun.
s4 Segment 2
Miniboss segment. T4's run.
Considering this miniboss was supposed to be fought in an auto-scroller vehicle, there is absolutely no floor below us. With no place to stand, we would surely fall to our deaths. It's a good thing this miniboss doesn't last long enough to make that a problem!
We enter with Flame Charge, with upward momentum from our recent jump. We timed it so the Flame Charge ends as soon as it dealt 5 damage. Switching to laser, we used its charged shot to blast the miniboss. Two hits, 10 damage each, plus the 5 from the Flame Charge, and down goes the miniboss! Then we land on what looks like nothing, and dash on our way.
The laser's timing was chosen to minimize lag, probably due to laggy animation crud lining up with other laggy animation crud to overlap the lag frames. We've taken damage, intentionally, before meeting the miniboss, so that we would remain invincible until we fire the laser, so that the 16 frames of ouch is spent moving us (slowly) rather than flatly wasted, and secondly to make the jump to the invisble terrain easier.
The fact survival at all is even possible wasn't obvious, as a few teams did die, but they still killed the miniboss, so they showed up in the next segment, still having broken the auto-scroller as a whole.
s4 Segment 3
Save segment. Mix of T4 and T3.
Wondering about that invisible terrain? When we defeat the miniboss, we are moved straight into the next segment. Usually, the segments have their latest block coincide with the earliest block of the next one. As the game expected us to fight the miniboss in the auto-scroller vehicle, it would transition and look nice from there. Since we aren't, the standard routine of warping you to the start of the next segment is instead called, and the miniboss' and next segment's blocks do not coincide. As a result, we actually have the terrain warp in around us. My terrain script shows this effect just fine, even if the game's graphics don't. We landed on the save platform here, as you can clearly not see.
Short segment, yet I've swapped between T4's and T3's strats. T4 miniboss kill, T3 save platform location, T3's pixel perfect spot on ladder, T4's ladder headbonk. The pixel perfect spot is so we avoid a bullet later, by the way. In the meantime, we're starved for Flame ammo, and manipulate a whole bunch from a convenient 1 HP enemy.
There's two frames of inserted delay before jumping off the ladder. For some reason, removing the delay causes lag frames later, and I opted to reduce lag count. Part of the reason was because this delay happens to line up the item manipulation I did earlier, which meant less work. Removing lag is basically guaranteed savings, but removing normal frames might clash with frame rules and item drops.
s4 Segment 4
Down segment. T3 used as base.
Ghost Fall! Whee! Some teams used the Infinite Climb, which is where you climb on a ladder and misalign yourself with it, confusing the game into never finding a ladder tile and causing you to ascend or descend forever, with no respect for any terran. You only get out of it if there is another ladder on the way that's lined up with you, and thus a ladder tile is detected and you are free to move. I didn't do that this run, since it's only 2 pixels/frame, and Ghost Fall uses the falling speed of 4 pixels/frame with zero speed once every 9 frames. Triggering Ghost Fall on this ladder simply meant picking a particular camera position where traversing the thickness of the ladder is enough to trigger this segment.
Some shots were made early this segment to damage a few enemies. This is to reduce the object count as they don't fire bullets, and thus less lag. Flame shot was needed to destroy an enemy on the way, as otherwise we take damage, and the fact we got an item drop is a bonus. An enemy shoots a bullet with a rather nasty timing, needing us to be as far right as possible, one pixel short isn't enough. One more shot to an enemy so we can sneak around the right side of it damage free, and finally pick the left-most position to enter the next segment without causing an enemy to self-destruct and lag the game to pieces.
About that pixel perfect bullet dodge. When I say as far right as possible, we had to have an odd X position. An even position would not have worked, as we travel in steps of 2 pixels, and wall detection is all-or-nothing. We can't get to an odd position during a fall, so this had to be adjusted from a dash while in the previous segment.
s4 Segment 5
Our Flame Charge has us crashing into a wall and sticking there in an ugly fashion for a few frames. Yes, I would like to avoid that. No, I can't think of any useful way. Flame Charge checks our bottom pixel for horizontal collision, and a floor tile gets in the way. Yes, if we loaded this segment earlier and were a few pixels higher, we would continue straight forward, but if we fire it a frame early, we fail to load the segment and collide with camera edge. Yes, the camera would be free to go down if we were a few pixels higher relative to the camera to load this segment with us higher up, but then we don't get the Ghost Fall. Finally, we are falling to our deaths thanks to Ghost Fall, with the only hope of escaping is to Flame Charge. No, there is no way to cut a Flame Charge short. If only that floor ahead was just one tile down!
s4 Segment 6
Auto-scroller segment. At least, it's supposed to be.
s4 Segment 7
This is supposed to be a separate segment meant to catch us after the auto-scroller, but since we broke that, it just looks like an extension of the previous segment.
s4 Segment 8
s4 Segment 9
s4 Segment 10
In a shocking twist, I use charged Spikeballs four times, for 7 hits totalling 56 damage! Sure, I'm at -5 ammo after the fact, but then Charge Storage doesn't know when to quit. Actually, I couldn't do it more if I wanted to, as I need 2 ammo so the game would allow the B press to do something.
56 damage or not, the boss still has 4 hits remaining with -1 HP to go. The no damage Muzzle Flash is enough to defeat the boss at this point, as we just need the hits, not the damage.
|Seg 1||479||477||477||477||716||477||477||579||477||493||Charge Swap Bubble Barrier|
|Seg 3||123||119||91||122||108||110||122||105||115||102||Miniboss skip set-up|
|Seg 4 (mB)||77||268||241||282||315||281||271||373||294||381|
|Seg 11||1407||1406||1408||1408||1544||1459||1468||1493||1462||1538||Frame rule|
Team 3 leads DTC6! About time Team 4 fell behind somewhere, but they still were very competent compared to other teams. Our advantage over them is the Miniboss Skip and spamming Charge Swap Bubble Barrier.
s5 Segment 0
Right segment. Inspired by Team 2.
We need to immediately activate the barrier, then immediately use a Flame Charge, with no break in between. This actually makes it important that we do Stage 5 before Stages 2 or 3, as we'd otherwise have a weapon in between flame and bubble, which will slow us down 2 frames to move the cursor an extra step. Since none of the remaining weapons are even worth the 2 frames, here I am, selecting Stage 5 after 4, and if those 2 frames get me a frame rule, it's more than any other route can give.
We're still not done with talking about the first frame of the stage. In addition to Bubble Barrier Charge Swap to Flame Charge, we also don't immediately dash. If we do, we're in screen position X 82. By walking two frames first, we can shift this to screen position X 80. This is yet another example of the importance of the camera, as it is that which loads segments, not our position. And it is vital at the end of this segment that the camera is in a better spot.
So, with the
invincibility cheat Charge Swap Bubble Barrier, this gives more freedom to simply dash forward whenever the ground presents itself, and to Flame Charge in spots where I would do less ground dashing. I would smash enemies more, but lag issues mean some of them get to live.
s5 Segment 1
Ghost Fall. I'm sorry, but I can't convince gravity to pull me down any faster. The ladder, aside from its top tile, doesn't interfere with my movement, so I can get into position to trigger the Ghost Fall easily enough. Most teams began falling at screen position X 82, but since we did so at X 80, we fell for a shorter time before triggering the segment. So we're on a different fall cycle when the camera begins moving down, and as a result, we manage to reach the next segment just before we hit the part of the fall cycle with zero speed. I suppose we do manage to fall faster than other teams, by a single frame.
The reason we lose 3 frames is because we needed another Charge Swap Bubble Barrier. Pause, switch weapons, unpause process takes 3 frames, if we get zero lag. Technically 4, but the gameplay advances one frame on the moment we press Start to pause. Since we have a barrier that doesn't interfere with our attacks, the plan then becomes to avoid enemies to preserve the barrier, rather than having to get rid of its 6 hits quickly. Enemies are enough in the way that I leave with 4 HP left on the barrier.
s5 Segment 2
Remember that Screen Position X 80 I was talking about earlier? It hasn't finished giving us gifts. Because we got Ghost Fall after falling a shorter time, we're higher up relative to camera. Since the camera itself is needed to load segments, we're higher up when loading this segment. This additional height is just enough to reach the first ledge with a lower jump, and we can begin dashing earlier. After that, the screen position from the start of the stage is finally done helping us. Screen position and the camera is extremely important for optimizations!
s5 Segment 3
This short segment is the spot just before the miniboss. Basically an extension of the right-going segment before this one. I don't have much to say.
I did try shooting the painfully slow bubbles to the right, but haven't had much success in getting it to destroy the missile at the right moment. Would have saved us from going left for a few more frames, but so it goes.
s5 Segment 4
Miniboss segment. New stuff here.
There really should be this huge, massive golden robot guardian here, I swear! You won't believe me because I skip it thanks to a missile I shot to the left.
s5 Segment 5
s5 Segment 6
s5 Segment 7
s5 Segment 8
s5 Segment 9
s5 Segment 10
s5 Segment 11
Standard procedure: Spikeball first, hit every frame enemy becomes vulnerable using overflow damage. I hope you're used to seeing this since two bosses ago. There's about nothing here that's any problem.
|Seg 3||105||89||100||110||106||102||100||109||99||107||Miniboss Skip set-up|
|Seg 4 (mB)||61||199||210||216||61||215||263||355||333||220|
|Seg 7||1008||1034||1013||1052||1004||1063||1073||1067||1066||1053||Probably tiny frame stuff|
|Seg 9||123||124||123||121||111||121||117||99||110||133||Camera optimization|
|Seg 11||1675||1676||1672||1698||1706||1698||1692||1787||1688||1694||How'd they manage?|
This volcanic stage has a few barriers that require us to shoot to get through. There are also many inconvenient enemies in the way, so the barrier would be very helpful, if it weren't for those breakable blocks stopping us. Well, they would, except the Charge Swap lets us get around that restriction while still enjoying our ability to ignore stuff.
s3 Segment 0
Right segment. T3 used as base.
s3 Segment 1
Down segment. T4 used as base.
We're too lazy to reach the ladder on the far side, instead we'll just fall down past this ledge. We looked for ways to trigger Ghost Fall, but it is not happening this segment. We also make use of Charge Swap Bubble Barrier here, since there's a lot of enemies in the way, and there's a breakable block in the way as well. We get to ignore attacks and break through, the best of having a barrier and being able to attack freely. Since we're immediately stopped by flat ground, we need to take the ladder on the left. This segment basically snakes back and forth from there, with no uneven terrain to Ledge Clip through other than starting off this segment and leaving it again.
This segment is an excellent example of the assymetry of left-side drops vs. right-side drops. We dash straight off ledges on the right, but prefer to jump on the left. This is because, whenever we stop our dash, the game only checks the right-side pixel. If that is over air, we fall. Since it's in a convenient position on the right side, we're closer to the ledge when we stop dashing, and we're in an ideal spot to turn around and go the other way. On the left side, we're stuck waiting until the right-side pixel is over the air. Then we're pretty far from the ledge when we do fall, and we're not in a good position to go back the other way.
s3 Segment 2
s3 Segment 3
Save segment. T7's inspiration.
s3 Segment 4
Miniboss segment. T7's inspiration.
We skip the miniboss. It would be some guy sitting in a drill car. You did not need to see it anyway.
We do carefully tweak the camera, by the way. If we dash blindly, the CamX might be 15 or even 14, but with proper pixel positioning, we get it to 16. As for our screen position, we begin from 100, as that lets us Ledge Clip into a Ghost Fall readily enough.
s3 Segment 5
We ignore the floor by doing a Ledge Clip, then continue to ignore more floors with Ghost Fall later. Understandably, many teams have taken the same 333 frames on this segment.
s3 Segment 6
I'm falling as fast as I can. What? Is ignoring all terrain not good enough? Yeesh.
s3 Segment 7
s3 Segment 8
s3 Segment 9
s3 Segment 10
s3 Segment 11
Charged Spikeball. Frame perfect hits. Easy, easy to do on 60 frames charging or 12 frames shot cooldown against 96 frames invinciblity.
I've noted this boss acts based on when you clear the dialog, as opposed to when the boss loads. There's a distinction -- If we delay the boss, this one will act later, exactly matching the delay. Pointing this out because this boss has an invincibility move that takes a while, and there's no sneaking in a shot earlier by delaying the dialog and changing our first attack relative to this invincibility move.
|Seg 0||935||964||933||987||987||969||993||949||973||1082||Segment transitions, mainly|
|Seg 1||896||870||925||966||989||1005||909||1007||970||915||Team 4 took the left path|
|Seg 2||93||154||95||103||90||92||154||92||152||160||Um, camera fixings?|
|Seg 3 (mB)||195||220||223||264||246||219||285||230||295||263|
|Seg 6||249||242||248||244||239||254||240||228||240||293||We're fixing the camera|
|Seg 7||252||268||254||247||299||254||273||296||293||294||Remind me to look at Team 6|
|Seg 8||1788||1788||1785||1806||1815||1809||1807||1820||1818||1807||Probably how we took the door|
Team 3 did a rather stellar job. Most of my stuff was based on copying their stuff then butchering their work. No, nothing against them for being so close to winning or anything. I just want the most optimal Zook Man ZX4 run so there's less of it to see.
s6 Segment 0
s6 Segment 1
We cheat some more- I mean, use Charge Swap Bubble Barrier again. Seriously, there's no visuals by the game itself why we're invincible, so I have to remind you of this. The leading teams use Bubble Barrier, but Charge Swap takes a few frames and in return frees our ability to attack, which is snapping useful.
There's a rather interesting spot here, that takes no fewer than three tricks in the span of a second to pull off. First, we land on the ledge below the ladder, and tap left to initiate Dash Ready. Then we jump onto the ladder, get ourselves to a pixel perfect hieght, and dash left before the Dash Ready expires. The pixel perfect height lets us go past the wall and later walljump up it on the wrong side. Finally, the walljumps allow us to rise through the ceiling in that awkward spot. All this so we can save having to climb a ladder only to fall back down.
Even before this segment is done, we call up another Charge Swap Bubble Barrier.
s6 Segment 2
s6 Segment 3
The Miniboss Skip can theoretically work here, but one set-up likely requires we first beat this stage, which is a lovely catch-22, or have us visit another miniboss and then leave its stage, which takes far too long. Beating this thing up normally, it is.
5 damage from Flame Charge as our opener, then two hits from Laser Charge for 10 damage each, totals 25 HP. Me > Miniboss.
I walk in with 5 barrier HP, and walk out with 2 remaining. Yes, the 3 HP was best spent on this one enemy. Would have been 4 HP, but I adjusted my position to have the laser last slightly longer, which is enough to keep me invincible that little bit longer, and avoid wasting that one extra barrier HP as now the miniboss disappears before the barrier timer runs out. The 2 HP is important in the next segment, as I'll be too busy with Flame Charge to have a chance to charge the barrier.
Trivia: After beating this miniboss, it is possible to almost soft-lock the game, very easily. Just charge up your bubble weapon and use the Bubble Barrier. There are no enemies before you are stopped by a breakable block. With a normal barrier up, you can't shoot, swap weapons, or pause. The barrier will never go away on its own. Good luck finding a way out! (Hint: Find a spot Ledge Clip to escape)
s6 Segment 4
s6 Segment 5
s6 Segment 6
s6 Segment 7
s6 Segment 8
Oh, this guy's hitbox is very small. Don't worry, it still fits him fine, just that he's smaller than other bosses. Small enough that our opener Spikeball only hits once. Actually, it'll still hit twice just fine, just that we won't have a projectile embedded in him on the frame he loses invincibility. So we fire our second spikeball early, for 3 hits of 24 damage, and no chance for Charge Storage for another spikeball hit. This works just fine, if for no other reason than the fact we can't really win faster than 11 hits anyway.
|Seg 0||224||224||224||224||220||224||224||183||184||224||Don't take the autoscroller|
|Seg 1||1599||1595||1554||1660||1656||1653||1689||4101||4101||1619||Setting up Miniboss Skip|
|Seg 2 (mB)||61||265||265||340||61||242||532||427||422||240|
|Seg 5||194||189||204||224||198||200||211||211||129||216||T1 took the autoscroller again|
|Seg 10||272||237||237||272||234||249||248||251||237||297||Need to fix camera|
This is the other stage that contains an auto-scroller. I am still really glad there are segment triggers we can use beyond the auto-scroller vehicles.
s2 Segment 0
Dash straight off the platform into the wild blue yonder! Flame Charge into the open! I believe I can fly! Well, thanks to Flame Charge preserving upward momentum as well as allowing mid-air jumps, we can get pretty far, indeed. Not forever, but at least far enough to use floating platforms ahead to save ourselves. We don't take the auto-scroller vehicle, as auto-scrollers have no style.
Team 7 opts to use Flame Charge before landing on that little ledge up there, which is primarily why they're ahead on this segment by a few frames. This puts them in an awkward spot that, ultimately, loses them a few frames.
s2 Segment 1
Auto-scroller segment. At least, it's supposed to be.
s2 Segment 2
Miniboss segment. Inspired by T7.
Some kind of flying ship miniboss would be here. Well, it would be. Just take my word for it, it's yet another miniboss I skip.
s2 Segment 3
s2 Segment 4
s2 Segment 5
s2 Segment 6
Auto-scroller segment. At least, it's supposed to be.
s2 Segment 7
s2 Segment 8
s2 Segment 9
s2 Segment 10
s2 Segment 11
Boss segment. T3's run
At last, there's a small variation in our standard procedure. This boss starts in the air. We use another laser here, before the boss dialog starts, and break the lock there as well. It's all address 15F0, I say. This was because this lets us use Charged Spikeball on the earliest frame possible instead of waiting until we had a chance to jump up.
After that initial hit, standard procedure. Smack boss with enough overflow damage early, and lodge bullets in boss on frames that invincibility wears off. There is a moment the boss uses whirlwind attacks, but there's enough objects to lag us even if we stand still. Taking damage stops this lag, probably because the game's not drawing us, or perhaps the player hitbox is disabled for important frames. Not sure. Just that taking damage removes the lag.
The final hit isn't standard procedure. This boss just is interesting start and finish! We jump up the wall a little ways and shoot the boss with the star weapon. The hit itself is delayed a few frames, and this was to bring the boss further up. We're on the wall to get higher up so the boss does not warp to its next position when it is hit. When the boss dies, it stays up there to explode. Far enough that the explosion particles instantly unload. As a result, we beam out earlier.
Stage 9 (end)
This is a series of small stages that end in all the bosses you've already fought. After you've finished fighting the eight all over again... The game ends, and you didn't really get to see anything you haven't already. Bah.
Unlike the stage select, each transition does not refill ammo. Basically, you come in with full tanks of ammo on each of your weapons, and they had better last, as the only refills are from item drops off of enemies, and whatever ammo tanks you walked in with. If it's any consolation, you do get free health refills in the transitions.
So the game starts sending re-fights at us. The only thing that makes this slightly more interesting is that our ammo doesn't automatically refill when we move on to the next segment of the end stages.
On a side note, we tried skipping this boss like we do later ones. We can spawn these normal enemies by the door, but they are exactly one pixel away from remaining in existence when the scrolling stops. We're not aware of any method to move these enemies one single pixel, but if you do get that one pixel, it will be enough to skip this boss.
Oh great, another re-fight. Guess we'll- No, don't feel like fighting this boss. Hey, I can spawn and subsequently shoot this enemy! Let's see what happens when we do after the boss loads! So impressed the boss literally exploded. Please see my comments on the intro boss as to what meaning of 'literally' I'm actually using. I've spent my quota of rants today.
There will be three more of these. These Boss Skips work because we shoot an enemy while in boss mode, which attacks the boss HP, and in turn this is zero before the dialog finishes. Only in the end stages are there spawns in the segment as we scroll. Although it automatically scrolls the camera to the right, moving left will still trigger spawns. In order to move left, we have to use the charged laser to break ourselves out of the door lock. So laser at door, move left when it would spawn something close enough, shoot at it before the boss speaks. Charged Star weapon is basically the only weapon that will reach far enough off-screen.
The Frozen Laser was once used to dramatically shorten the input file. Now that the boss skip works on this guy, we don't have to wait until end of dialog. And since this skip ends the game much sooner, I am happy to see that I don't have to make the choice of shortest input or shortest to ending.