Posts for Blechy


1 2
21 22
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
This might be known as I know IRC discussion is hot hot hot, but I believe one improvement is to enter the clown room @ 3:34 earlier, to start his walking pattern sooner.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
After a long hiatus from TASing this was a great reason to return. Brilliant idea. Can't wait to watch the video & hopefully see how I compared.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Still can't believe midair zips are a thing. Fun stuff.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Well I think it's a great idea and I'll be buying 2 subscriptions for every 1 person that doesn't.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Stumbled into something new today: if you die on the same frame you exit a level, you activate the zero health glitch just like at the end of 2-4. Unfortunately, this is probably useless in tas because health isn't an issue anywhere (I think) and it requires time to set up, and it's too difficult to reliably set up in real time. Link to video
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Found another type of zip that can't be used anywhere. Done by dropping down from a platform. For a split second I thought this could be extremely useful in 2-4 in the bottom middle room, but the door prevents it. http://tasvideos.org/userfiles/info/7928813407387583
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Ahhh I'm so stoked to see this in progress! Is it not possible to boost while you're jumping left along the set of purple blocks just above it? Can't wait, I'll watch every wip you post!
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Ohhh good idea! I dismissed this cause I thought you'd need more headroom, but I guess if you don't need to do a running jump, then that's a nonissue. 2-4 has a room like this in the new route as well. Too bad about 3-4. I noticed in one of your vids you were able to not fall through a 1-tile high row by looking up, but I guess you need to get stuck in it first. Also um, here's this. Turns out the fortress levels don't have pits I guess, so you're able to just jump around down there.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Awesome stuff! One thing that intrigued me is the way you zipped into the 1-3 wall without building up speed first. I tried the same inputs of 14 left+A, 3 left, 2 left +A, but I didn't go nearly as far into the wall as you did. Does it have to do with that first "failed" zip you did just before it? That 1-4 route is actually the one I suggested a few posts back :D I think the 2 most hopeful places right now are: 1) The 3-4 room that's just like the new 2-4 skip room (1 right & 1 up from entrance). This time however you'd be zipping from right to left. I wonder if it's possible to apply that 1-tile zip you found, that would be a huge improvement! I tried, but couldn't get it working. 2) The 2-4 room with two eggplant wizards (3 down & 1 right from entrance). I can't see how this would work since the door is preventing you from doing anything, but if you could screenwrap here you would avoid any extra screens in thsi level.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Cool! I looked around for an application of this and nothing immediately popped out to me. There's a room in 3-4 with the same design as the 2-4 skip room, where it would be nice to zip from the right side to left rather than vice versa. It even has a tile 3 spots above the floor, but there's no floor below it so if you try zipping you just fall through. There's also a 1-tile high spot in that room, but pit's minimum jump is so tall I don't see that working out. If Pit's x-speed while zipping is faster than his normal walking speed, there might be a bunch of tiny timesavers throughout the run by walking inside walls you'd normally walk on top of. But I don't think it is, plus even in a tas there would be minor setup time as you get stuck in the wall.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
So, a few new strats have come up, some of which are tas improvements. 1) Wallwraps. First noticed by Jorf, investigated by Krankdud & me. Movie file: http://tasvideos.org/userfiles/info/7660940499519359 You can fully clip into any wall which is 3 tiles high, and to the left of Pit. This works very similarly to the 1-4 skip, except you must be running at the wall before you jump. This gives you enough speed to clip in far enough to push you to the left, rather than simply push you back out of the wall. The easiest way to explain this is to rest Pit against the left wall, walk right for 22 frames, wait for him to stop, walk left for 12 frames, then hold left+A for 14 frames. You then have a large frame window to turn around, so that Pit "zips" slowly to the left, and wraps back to the right of the screen. You can do it without this elaborate setup - it's a 2 frame window to jump, and then a 2 frame window to release jump. From what I can tell, this is useful in 4 spots: - The 2-4 room seen in the movie. This saves up to ~58 seconds in realtime, probably closer to 45-50 in tas. - Very end of 1-3. You can wrap to the left of the final reaper, and jump into the door. Maybe saves 1-2 sec. - Allows an alternate route in 1-4, in the room 3 right & 2 down of the entrance. Cuts out 2 rooms in the route, probably saves like 8 sec. - You can use the head statue at the beginning of 2-2 in the upper left to screenwrap (you don't need a running start). If done in realtime, you eventually need to delay just a little bit so that you can dodge the flying enemies that come out from the side of the screen. 2) More overworld screenwraps. Found by Jorf. By jumping & ducking left into a ceiling (or snake urn) on the far left of the screen, Pit wraps around to the right side of the screen. Saves ~5 sec in 2-3 by not needing to wait for elevator positioning to set up the screenwrap. Possible in 2-1, but not helpful. Useless in 2-2 due to the faster wallwrap mentioned above. 3) Platform manipulation we've been discussing. This might lead to small improvements throughout, but if it's possible to have platforms move in the completely opposite direction, it would probably save several seconds each in the following spots: - 1-3 first platform (you actually already seem to manipulate this optimally in your tas but I can't tell how) - 1-3 last platform - 2-3 early platforms. You probably can't do anything here since there aren't any previous enemies or platforms, unless the addresses call back to the previous level's. - 2-3 platform we discussed previously, before narrow area.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Very odd. When I play out the level, as long as nothing goes terribly awry, that platform always moves left. In the movie files, I actually intentionally missed that very early jump to the right, as I couldn't get the final platform to move right if I did the first part optimally. I didn't try very long so maybe I'm wrong there. But it would make sense, as that first platform is the one which shares the same RAM address as the platform in question.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Ahh that makes total sense! So for the particular platform I was talking about in 1-2, since I didn't jump, I must have simply delayed its cycle so that it was moving right instead of left when I finally did despawn it. Thanks!! I'm still not positive on the platform in 3-2 though. In most cases if a platform starts moving right, it simply nudges a bit to the right, and then turns around and moves left instead. However, this 3-2 platform can continue moving right. Here are a couple fm videos illustrating it, made in fceux 2.2.0 using rom "Kid Icarus (UE).nes". I haven't uploaded videos like this before so hopefully it gives you what you need. Left: http://tasvideos.org/userfiles/info/7623580508089370 Right: http://tasvideos.org/userfiles/info/7623632946558730 Another weird thing, I've actually noticed that even the very first platform in the game (in 1-2) varies a bit. Do you know if the flying enemies share the x-speed RAM address with platforms, like they do for x/y position? If so, perhaps the platform is affected by the left/right direction of whatever enemy last occupied that address.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Thanks Randil! I monkeyed around with the addresses and this is some really cool stuff! I actually ended up investigating something totally different, which you probably already know - it looks like there are "categories" of enemies, and only 4 of a given category can be displayed at a time. If enemy names are confusing, here's a list: http://www.flyingomelette.com/kidicarus/kidicarusenemies.html Category A - Snakes Category B - Bouncing enemies, flying enemies (except reapettes), elevators, crawlers (ex. nettlers), castle crawlers (ex. kobils), collin/eeleye Category C - Reapettes Unknown - Floor enemies (ex. McGoo), Reapers. Sometimes reapers are category B enemies, and sometimes not. Mass enemy rooms (ex. Specknoses, or castles) allow 8 at a time, as they take up category B and I assume one other. In terms of respawns, category B is actually grouped in 4's, and a new group will not spawn until the entire group is defeated, and their hearts collected. Unlike snakes, which continue spawning as long as there aren't 4 on the screen, or reapettes, where only 4 spawn per "reaper call". This is extremely obvious just from playing the game, but it's cool that it's unique to each category. For the most part, this is all seamless due to how the levels are designed with regards to enemy placement. However, it's particularly apparent somewhere like 1-2, where a group of 4 flying enemies on the screen despawns one by one as you spawn platforms onto the screen. ----- SO, the reason this is all relevant is because I believe the position of a sprite as it spawns is affected by the position of whatever sprite it replaced in its category. Back to the particular moving platform I mentioned in my previous post. Here is a layout of all the nearby platforms: http://i.imgur.com/J8vsRjb.png To be clear, my ultimate goal here is to 1) spawn P2 as soon as possible to get it moving in its cycle, AND 2) position P3b in such a way that I can jump on it, then jump to the center block just to its right, thus skipping a cycle (as I do in the youtube vid I linked in my last post). To do #1, you just need to make P2 appear as early as possible. #2 is more confusing, and I believe is decided by when you jump to despawn P3a. If I jump to despawn P3a too early, P3b is moving left, and its x coordinates per frame are: -128, 127 (x2 frames), 126 (x2), and so on, which is the full speed of a moving platform. If I jump to despawn P3a a bit later, P3b is moving right (shown by underlined text), and its x coordinates per frame are: -128 (x2), -127 (x4), -126 (x8), -125 (x24), -126 (x4), -127 (x4), -128 (x2), 127 (x2), 126 (x2), etc. As you can see, it moves right all the way to x position -125, and it's not until the bolded text that it effectively "catches up" with the other cycle. This time is enough to jump on it, and then onto the central block. Thus, the spawn position of P3b definitely varies, and I think it's dependent on the last position of P3a. I approached the series of platforms in different ways, and if I despawned P3a when it was last at position -64 when moving right, P3b was in the ideal spot. (-64 was just a constant I chose, it's not tight to achieve the P3b positioning.) I hope this all makes sense, does this match your findings at all or is it possible I'm off base? ----- As for a platform moving in the "incorrect" direction, it's rare but I've definitely seen it. The only video I have of this phenomenon is here: http://www.twitch.tv/thesundaysequencebreak/b/377840856?t=3h17m42s Sorry for the weird format, but watch the upper left stream: when the last platform before the narrow part appears, it moves right instead of left.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Necro necro necro. A few folks in the speedrun scene are taking up this game, and I'm not too clear on what causes moving platforms to spawn differently. Sometimes they spawn in different locations, and sometimes they move in different directions altogether. Example: Second to last platform in 1-2. Here is a video of me approaching it in two different ways. http://www.youtube.com/watch?v=X9NOAW4XUh4 In attempt 1, I jump to spawn the 2nd platform in the video earlier. This allows me to progress quicker, but when the 3rd platform spawns, it is too far left to make the jump. In attempt 2, I don't spawn the 2nd platform early. This causes a longer wait for the 2nd platform, but spawns the 3rd platform differently so I can make the jump. Here is a picture illustrating where the platform is in the first & second clips, as it comes on screen at the top: http://i.imgur.com/AkPKLCC.jpg I was curious if Randil or anyone had any knowledge on what decides where platforms spawn/what direction they move in initially. The TAS actually doesn't make this jump, but I'm not sure if it's possible without losing time waiting for the platform which precedes it. (I haven't started practicing 2-2 yet, but meeting some of those cycles is tight enough that this might come into play there as well)
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Out of curiosity, does anyone know what the difficulty level changes in this game? It doesn't seem to affect default health, continue count, or enemy spawns. Is it maybe drop rates/boss health?
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Just played through this game. It seems that you can jump kick through the force fields which block your path at the end of levels; at the very least I did this for the final stage's. This should be a decent improvement if this game is ever revisited. e: Nevermind, looked into it further and I misunderstood how the level's triggers worked.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Possible ridiculously minor improvement if another version is ever made - after the first phoenix in the ashura fight, it might be faster to use warrior force and just give a barrage of punches with 1 final kick to finish him. This prevents him from ever going invulnerable. I was able to do it from a position where I was on a wall with him right next to me, and it saved a bit less than a second compared to just phoenixes. I didn't test it totally optimally, so might be off, but worth trying out. Not worth it realtime obvi. Also I tried this on dragophant boss refight and it's not useful, as spamming punches in this same way triggers his invulnerability timer.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Hiya. I've been playing the game a lot recently, and one of the few mysteries left is how to go about getting an ensured magic drop in the fortress, for use in the gateway. I tried going frame by frame, but it's going to take forever. Was any research ever put into finding a memory address for enemy drops?
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
I've actually been racing this game recently. Does anyone know how to raise the odds of getting particular powerups? At this point we know what powerups are available in what stages, and roughly from what blocks, but we don't know how to make it assured we'll get what we want.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Heya, any luck fixing the rng? If it helps, now that we've gotten a good time cyg and I are taking a bit of a break, partially in case you find any new timesavers. So we won't be randomly finding things that would save a few frames way early in the game. :]
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Randil wrote:
EDIT: Can you give details on how Goofy reaches the exit faster? I tried using the same strategy but exit the room with Goofy instead of Max, and it was 11 frames slower. I also tried picking up the pot with Max and immediately running to the exit with Goofy, but this was 8 frames slower.
Sure - Basically after picking up the pot, goofy takes the same path Max does, walking past the switch spot on the right. He should barely make it past the path of the block max kicks before it hits him. In a tas, Max may have to wait a few frames before kicking it.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Uh oh, found an improvement in early stage 1. In room 6 pass 1 on your map, goofy can actually make it out the right exit faster than Max can. Edit: Watched your videos, that boss is insane. We were thinking that double kills might still improve it further, but considering that everything gives you something to throw, might not improve by much. Again I think it's just a case of how far can you push the insane amount of luck in the game.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Cyg & I noticed in our runs that some pots reliably will never drop fruit. I was messing around just now and had a breakthrough as to why this is: it seems pots picked up via hook will never drop an item.
Experienced Forum User, Published Author, Former player
Joined: 6/27/2004
Posts: 550
Location: New York
Other wiki ideas. Dealing damage/enemy health lists (could use sprites at http://www.spriters-resource.com/community/showthread.php?tid=19734 to be fancy) I think bombs thrown by Max deal 5 damage to enemies and 4 damage to cannons/bosses. That seems weird though, there might be something behind the scenes we don't know. Bunch of other circumstances seem to deal 5 damage, but no way of testing further. This includes bombs shot by a cannon, bomb explosions, kicked blocks & their explosions, pots thrown by red pirates, mine-cart roadkill. Skinny blue pirates, snakes, porcupines, frogs, ghosts, bats, bees, boss 1 enemies all have 4 health. All non-pirate enemies can be killed by hook (untested on ghosts & bats). All other pirates have 5. Knights take 1 hit to be vulnerable, second to kill. Stage 2 cannons have 10 health, stage 3 & 5 cannons have 20 health (not fully tested) Boss 2 has 48 health. Not positive on boss 3/5 health. Boss 4 takes certain amount of hits. Other helpful info: Frame durations of pot pickup (different between goofy/max?), throw animation, point blank hook animation, damage animation, stun animation (from friendly barrel), death animation, and falling to death animation. I think death animation might always be the same no matter how far you have to travel to reach the edge of the screen. Information on z-axis. Goofy tends to be hit by a thrown object before Max is. Think he can also throw from behind max further than max can throw from behind him, without it hitting the other player. Can throw over some of the animal enemies, in a speedrun this is relevant for porcupines & frogs. Helpful info that requires extensive testing: Fruit spawns. All I know about this is that some pots will never ever spawn fruit, at least with the path cyg & I take. Determining enemy AI Interesting info that's not particularly helpful Kicked block speed Shovel details!!
1 2
21 22