Hello! This game is Battle Kid 2, I made it "100%" run. It including open Portal ABCDE, get double power shot and x,y.
I watched caitsith2's TAS, nEilfox's TAS and some videos like SGDQ, and I think it can be improved.
I compared nEilfox's TAS, changed the route. Get pink cave Orrexium and open Portal B later. Manipulates luck like in ice cave to get Orrexium. Changed bosses fright way and so on. I improved it 3156 frames.
Skill
When in the corner and no get the X-sludge obtained, don't put left or right in polluted water, it can be faster.
When fight bosses, to be as close to him (her), it can be faster.
When jump in to water and get the feather fall obtained, put up in first frame it can be faster.
Treble jump faster move than double jump, of course, double jump can faster move than one jump.
Use jump breakpoint in front of the next scene can be faster move.
When get the feather fall obtained, behind of jump put up can faster drop.
As much as possible jump high, it can get more acceleration.
Static scenes put jump in first frame maybe can faster.
All in all, I remade this TAS saved about 3 seconds.
Thanks
caitsith2 Thanks your lua and encode videos, them help me a lot.
Masterjun Thanks your idea about odd pixel values can go around enemies, also can faster drop in some places and do something danger.
Noxxa: Updated movie file with a 21-frame improvement.
Noxxa: There are known small improvements here and there, but nothing major or particularly noticeable to the viewer. That aside, this is still a fun, fast-paced, action-packed TAS. Accepting for Moons.
Nice stuff, especially at the bosses, I think I'm not too opposed to all teleports in the end. Questions:
1. Can't you stand right in front of the cancer boss using odd pixel positioning?
2. Can't you keep progressing left while still shooting at the parrot boss?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Joined: 7/25/2015
Posts: 123
Location: Republic of China
feos wrote:
Nice stuff, especially at the bosses, I think I'm not too opposed to all teleports in the end. Questions:
1. Can't you stand right in front of the cancer boss using odd pixel positioning?
2. Can't you keep progressing left while still shooting at the parrot boss?
Thank you feos.
1. Excuse me… Which is the cancer boss?
2. Penguin boss jump right can hit him with 30 Hz, although keep away left, it is faster than in left 13 frames.
Link to video
Working on:
[NES] Downtown Special - Kunio-kun no Jidaigeki Dayo Zenin Shuugou! (J) ''2 players 100%''
Plan:
[SNES] Kenyuu Densetsu Yaiba (Japan)
_________________
Ｍｙ Ｅｎｇｌｉｓｈ ｉｓ ｐｏｕｒ．
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Joined: 7/25/2015
Posts: 123
Location: Republic of China
Thank you feos! Your idea save 15 frames! (becasue I waste 1 frame in ice cave…)
By the way, question 1, the Lobster boss can't stand top by using odd pixel positioning.
Working on:
[NES] Downtown Special - Kunio-kun no Jidaigeki Dayo Zenin Shuugou! (J) ''2 players 100%''
Plan:
[SNES] Kenyuu Densetsu Yaiba (Japan)
_________________
Ｍｙ Ｅｎｇｌｉｓｈ ｉｓ ｐｏｕｒ．
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Alright, a submission.
So let me start this by saying: Good job at finishing this! I bet it took a lot of time considering how long this game is.
This room-based game is really fast-paced and you get a lot of different movement powerups while progressing. This means that this game is almost perfect for speedrunning and especially TASing and it looks really nice.
Though, having TASed this game myself (and posted a WIP), it's not too straightforward for a platformer. The improvements are usually only a few frames, but this game has a lot of rooms, so those frames add up.
Basically, optimizing this game is hard, but you can't easily see those slight errors.
And this is exactly the problem I have with this submission.
Of course I compared this movie to mine, and let me quickly list a few things that aren't optimized.
But first of all, this:
fcxiaopengyou wrote:
to be frank, it influences randomness, so I wasted 1 frame in ice cave.
RNG is advanced by simply adding xposition + yposition + some music byte.
You can easily manipulate the RNG by changing your yposition... so by jumping. You shouldn't be wasting a frame for RNG except for complicated situations.
You lose a frame in the menu at the start (...seriously?).
You lose a frame in room 4 and room 5.
Then like 2 in later rooms. 4 frames in the second boss.
A few frames again, and then (for some reason) 4 frames here.
(As you can tell, this game is suitable for frame wars... if only it wouldn't be so long)
This is around where my current WIP ends, so I can't compare further. I didn't even get to the other movement powerups so that might be an important part.
I'm sure I'm biased towards not liking this too much, considering how much effort I put into my movie myself and how there are a bunch of places that aren't too optimal. And I don't see a problem with accepting this for now, but this game deserves a really good, optimized movie and I think this just isn't the one.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Thanks to fcxiaopengyou for the submission, and thanks to caitsith2 for the encode. I've been looking forward to a Battle Kid 2 run for a long time and I found this movie very entertaining. The movie may be improvable, but no unoptimizations stood out to the naked eye, so I don't see the harm in publishing this for now. If anything, it'll make for an easy obsoletion down the line.
Yes vote.
Yeah, even though it fails to even match the existing records, it's not losing seconds, only frames here and there. So why not publish it?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
