As it is no longer April 1st anywhere in the world, I have modified the submission text in order to describe this movie rather than the already published one.

Emulator used: Gens 9z

Emulator settings:

  • Controller 1: 3 button
  • Controller 2: 6 button
  • Allow Up+Down: Yes
  • Sound: Enabled, 44100KhZ Stereo, YM2612 High Quality enabled

Categories

  • Does not use warps or passwords
  • Abuses programming errors in the game
  • Manipulates luck
  • Takes damage to save time
  • Aims to entertain
  • Ignores semi-important goals in the game

About the Game

In Ristar, you control a star with similarly stretchy arms, who grabs enemies and bonks them with his rock-hard head. Interestingly enough, Ristar actually uses a heavily modified version of the game engine used in Sonic the Hedgehog.

About the Movie

This movie uses the Japanese version of Ristar, because it has a faster menu, harder hard mode, and fewer cut-scenes. This movie makes extensive use of programming glitches to regain control faster than is normally possible, and help maintain full speed as much as possible. This movie manipulates luck to prevent bosses from using their longer attacks. This movie takes damage to save time, and ignores bonus-stages. This movie takes a small amount of time in several places to perform additional tricks, most notably, getting the maximum height bonus at the end of every round. This costs some time, but makes for a much more entertaining run.
This movie does not challenge "Super Mode" for the following reason: The only relevant difference between hard mode and super mode is that in Super Mode, it is impossible to have more than one hit point. As a result, the only difference for playing on Super Mode would be a few added periods of standing around, waiting.
This movie contains speed/entertainment tradeoffs. In many places, I took a more entertaining approach even when it was a few frames slower. For instance, the first miniboss can be defeated 12 frames sooner if I land inbetween the first and second hits and most levels can be completed several seconds faster by completely ignoring the poles at the end, or using them for a quick horizontal speed boost.

Suggested Screenshot

Improvement Explanations

Velocity control

  • Jumping is slower than walking -- sort of. Jumping will often cause the player object not to move for a frame, and velocity is reset to 0 upon landing. Several frames have been saved by minimizing jumps.
  • There is no angle at which bouncing of a normal enemy will result in a speed higher than normal walking speed. Much time has been saved by not attempting to exploit this nonexistant trick.
  • By grabbing an enemy at full arm extension, Ristar can gain several frames versus walking to that enemy. Though this is somewhat negated by the rebound, it still saves some frames in many places.
  • Swinging across trees/vertical poles provides a speed 0.625 pixels per frame higher than normal walking. Some time has been saved by grabbing these poles at the highest point possible, where applicable.
  • Hopping across monkey bars is faster than walking. Time has been saved by grabbing ceiling bars at the first possible frame, rather than the last one.
  • Better optimized hopping up ladders.
  • Better optimized use of poles. It's about 75% accurate to say that in the first version of this movie, if there was a pole, I used it wrong.
  • Better optimized diagonal grabs used in wall climbing.
  • Much better momentum preservation after gaining speed from slopes, poles, and the like.
  • Optimized swimming for fastest acceleration, and most time spent at maximum speed.
  • Made use of jumping before ledges to save time on long falls.
  • On ice, velocity raises by 8 per frame while below 896, and drops by 3 per frame while above this value, time is saved both by bunny hopping when moving quickly, and landing early on ice when moving slowly.

Luck improvements

  • Manipulated miniboss on Leatow to bring each wave down as quickly as possible.
  • Better patterns on the nautili in Leatow's Lost Palace allowed for less dodging between hits.
  • Better manipulation of Ohsat, the shark boss, for faster appearance in foreground.
  • Pattern manipulation on the Onaclove memory game miniboss.
  • Pattern manipulation on Awauek Sing.
  • Manipulation of the initial position of the Elykiki snowball monkey.
  • Manipulation of Itamor's landing zone so that he only takes two bounces to get back to his place between rounds.
  • Better pattern on Rewope's orb seargent miniboss.
  • Better pattern on Greedy's machine gun orbs.
  • Better manipulation of Greedy's attack pattern.

Level specific notes

Planet Neer

  • Not a lot special here. However, in the first portion, I swing across a root/vine that I had taken pains to avoid in the first version.

Dark Forest

  • Better grabbing of branches, roots, and bars (as explained above) saved about 2 seconds in this level.

Riho is Coming

  • Ah, Riho. The simplest of all the bosses. I saved about 9 frames from not jumping like a moron on the path up to his appearance, I saved about 6 frames from grabbing the old man's monkey better, and I saved about 3 frames getting the old man's monkey exposed. Not a lot of savings, but at least it was easy. Also I get to say that I optimized beating off an old man's monkey.

Planet Leatow

  • Hugely improved swimming route -- in the first version, I swam straight when I should have been going diagonally, and diagonally when I should have been going straight.

Lost Palace

  • Better use of grabbing frogs allowed me to spend less time on ladders to destroy the high up spawners.
  • Better patterns for destroying the sets of nautilus spawn allowed me to get to attacking the nautili much faster.
  • More efficient rebounds and grabs saved quite a bit of time in the drowned seahorse chamber.

Ohsat Attack

  • Sometimes he scrolls with the camera, sometimes he scrolls opposite the camera, and at nearly all times you're trying to get him to swim off the screen as soon as possible. But at the same time, your position affects the pattern he chooses. Manipulating this boss more or less came down to manually brute forcing it for nearly 8 hours, and I'm still not sure he's optimized, but it would take Nitsuja or a robot to do a better job.

Planet Onaclove

  • The first version really suffered from my misconception that I could get a higher than walking speed from bouncing off of enemies, and chests.
  • Also a slight spinning-pole improvement saves about half a second, but the major improvement comes from
  • Manipulating the miniboss -- For some reason it hadn't occured to me to attempt to manipulate the pattern on the memory machine. This was a fairly major oversight considering how incredibly easy this thing is to manipulate.

Under Factory

  • Here is where momentum preservation from poles truly shines.
  • There's a minor route change in this level -- hopping across monkey bars was a lot faster than dropping down and climbing back up. I hadn't taken this route before because I had (stupidly) thought it was impossible, and the bars were solely there so players could backtrack.
  • There's also a unique and bizarre glitch displayed in the minor route change. This glitch allows me to walk on air for about a second, until the chest I bounced off disappears. I have two theories about possible cause, but neither of them were possible to duplicate anywhere else. It occurs to me that this is likely actually the infamous "ice cap glitch" which plagues games with the sonic engine, but partially fixed so it doesn't last past the object being deleted.

Adahan Fall

  • Not a lot to say about this boss. He's simple, and boring, and there were minimal savings. Savings came from the same place as Riho.

Planet Neuos

  • I saved a bit of time before getting to the first fan+ladder segment. Here my savings dropped from about 50 frames to about 8. Oh well.
  • I was able to throw the third metronome clear across over the fan with the bomb, but in order for that to save any time, I had to take damage from the bomb to get it out of my way.
  • There is a route change here, some impressive pole use allows me to skip about 20 seconds of carrying the fourth metronome.

Dance Dance

  • Some better momentum preservation allows me to swing down much further to the right than I originally did. This saves me having to fall down and lose all my horizontal speed for a second time a few frames after landing initially.
  • By grabbing the fat dancing bird woman from farther away, I'm able to hit her before she's able to use her dance partner as a shield, so I don't need to bounce off the ceiling for invincibility in order to get past.
  • On the lighted dance floor, if you jump the frame you land, your speed is not lost, but unless you jump that frame, it gets reset to 0. So I bunny hop rather than showing off the annoying dancing guitars.
  • I grab the pole lower before the first long fall onto the drums, so that there's less distance to fall.
  • I don't bounce off the ceiling for invincibility time to land on the spikes after bonking the orb out of my way -- I have invincibility time from bouncing off the orb.

Awauek Sing

  • After falling from his perch, Awauek will bounce all the way to the left edge of the screen before regaining his composure and attacking Ristar. I save some time by keeping the screen scrolled all the way right.
  • I also save some time from better luck manipulation.

Planet Elykiki

  • Lots of time saved from bunny hopping on the ice
  • I don't grab the star like I did in the prior version, because I don't take damage on the miniboss.
  • I was able to manipulate the miniboss so that he appears far enough to the left that I am able to grab him before he picks up the bomb. This saves time from having to get the star, and time spent recovering from damage.
  • For the last snowball, I don't do one of those frozen fast balls, because they take longer to set up, and the monkey thing takes longer to recover from them, and the camera doesn't start scrolling again until after he waves.

Splash Snow

  • The swimming in this area wasn't nearly as improvable as I expected, I actually swam properly for the most part the first time around. Still, some time was saved by a better exit from the first water part, and optimizing the acceleration to hit top speed sooner.
  • I also save time by not doing a bunch of crazy grabs on the santa bomber and his parabombs. I actually have second thoughts about that decision, because those grabs were fairly awesome.
  • The other time saver is that I was able to turn around faster on each platform in the final slippery jump sequence.

Itamor Lunch

  • When actually playing, I love this boss, he's very creative, and difficult without being overly cheap. Unfortunately, his frost breath makes him a close second to Ohsat, the shark boss, for being a pain to TAS. Luckily, it only happens once.
  • I gain some frames from not jumping like a moron before the boss is loaded.
  • I gain some frames from making sure Itamor doesn't bounce more than necessary in between rounds.
  • I probably lost some frames to the old version during the frost breath segment. Out of forty or fifty tries, this was the first "pattern" that it was possible to carry the soup through in order to throw it at him when he starts sucking. I would like to note that both the horizontal and the vertical position of Ristar affect the remaining pattern of ice clouds, even after the breath segment has started.

Planet Rewope

  • Better strategy for the dividing tube monster.
  • These poles change their formation every time the grab button is pressed. On the first version I thought they only did it when Ristar was actually able to grab. This saved a bit of time.
  • Hopping across the monkey bars was much faster than bouncing across the enemies.
  • Landed further left on the icy slope simulator in order to get a higher speed when I jump.
  • Grabbed this pole further down before the long fall to the drum simulator
  • Actually hopped up this ladder instead of climbing it normally
  • Better swimming in the invisible water simulator.
  • Better velocity control on the floating things that carry ristar across the spikes.
  • Incidentally, when I grab that one but don't swing on it, it resets my falling speed to 0, allowing me to catch the one further down.

Brain Maze

  • In Room one, there is a single frame where it is possible to give input before the controls are locked out for the screen fade in. Bouncing of the wall to the left like I did is the one that gave him the position furthest right by the time controls are unlocked.
  • The teleporters can be activated 2 pixels sooner by facing the "wrong" way. This can save either 1, 0, or -1 frames. I used this when it saved 1 frame, but not at other times.
  • The special enemies here have a rebound speed 1.5 pixels per frame faster than Ristar's walking speed, I was able to use this to save time.
  • I saved time on the vertical spiked room by letting go of the pole earlier.
  • I lost some time in the torn up room because the crystal is in the one furthest left, and I was unable to manipulate this placement. I do make some of that time back, though, by swinging into the teleporter after placing the crystal.
  • In the room with four missle launchers, was able to grab the first and third ladders one rung higher to start with.
  • I was able to see Ristar's position in the background by disabling graphic layer 3. This helped a bit to speed up skipping the layer switcher in order to remain "invisible" into the teleporter.
  • Rebounding off this wall allows me to release the pole faster and still get the max height bonus.

Uranim Power

  • Savings here were much higher than expected, because I was able to get inside the boss and hit him repeatedly while being rebounded downwards.

Greedy's Spaceship

  • I saved some time on wall climbs
  • I was able to release the pole much sooner and still get the necessary height.

Inonis Snake

  • This is another annoying one to TAS... Always hitting the forcefield on the first possible frame will eventually result in death. Hitting the forcefield too many times will actually delay the machine becoming vulnerable, etc...
  • I saved a lot of time by not hitting the forcefield any more than necessary.
  • This level change doesn't come when the teleporter brings you to a certain position, but a certain number of frames after it activates. As a result, jumping before entering results in the transition occuring one frame later.

Greedy

  • Better manipulation lead to a savings of nearly 200 frames.

Comparison Table

LevelStartEndOld StartOld EndLevel SavingsTotal Savings
Planet Neer11144488111445829494
Dark Forest4664859047578837154247
Riho is Coming87871113290331139618264
Planet Leatow12318174821257918124381642
Lost Palace17711229581835323810210852
Ohsat Attack231572623324010274323461199
Planet Onaclove274023319528601346792761484
Under Factory334103582234894376503441828
Adahan Fall35991402473781942087121840
Planet Neuos4140847672432485095214403280
Dance Dance478755165751155551682313511
Awauek Sing51817553765532858944573568
Planet Elykiki565376056960106645023643933
Splash Snow607936441264726686442994132
Itamor Lunch64592691636882473467724304
Planet Rewope703367626374640808883214625
Brain Maze764668532781093902703164943
Uranim Power855268885090470939671735117
Greedy's Rocket89934912699504996483995214
Inanis Snake913359423296549996832375451
Greedy9433297272997841032875636015
Input09624901018425593(6015)
Note: For the purposes of this table, a level is considered "Started" when the in-level timer RAM address first changes from 0, and is considered "Ended" when the end of level value is replaced with 0, or any other value.

Truncated: Accepted as an improvement to the current version.

adelikat: Processing.


TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15576
Location: 127.0.0.1
This topic is for the purpose of discussing #1506: upthorn's Genesis Ristar in 26:44.15
Banned User
Joined: 12/23/2004
Posts: 1850
Is this an actual improvement or a joke submission? I can't tell; the text itself makes it seem like it's just the first level, copypastaed with the rest of the run.
Perma-banned
upthorn
He/Him
Emulator Coder, Active player (391)
Joined: 3/24/2006
Posts: 1802
Xkeeper wrote:
Is this an actual improvement or a joke submission? I can't tell; the text itself makes it seem like it's just the first level, copypastaed with the rest of the run.
The idea is that people will have to watch to find out. It will all be cleared up in 8 and a half hours, though
How fleeting are all human passions compared with the massive continuity of ducks.
Banned User
Joined: 12/23/2004
Posts: 1850
Normally I wouldn't mind, since it wouldn't take long to find out. Except Gens (and it's forced-resolution change) sucks. So boooooo.
Perma-banned
Senior Moderator
Joined: 8/4/2005
Posts: 5777
Location: Away
Whether this is a joke submission or not will be a mystery. What exactly did I vote, however, won't. :) (yes)
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Editor, Active player (297)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
$ cmp -l upthorn.gmv UpthornRistarV2.gmv |wc -l cmp: EOF on UpthornRistarV2.gmv 62574 At least it's not a clone.
Player (81)
Joined: 3/11/2005
Posts: 352
Location: Oregon
I'm glad that someone finally used the phrase "more optimal monkey grabbing" in a submission. Voting "yes".
ideamagnate| .seen aqfaq <nothing happens> DK64_MASTER| .seen nesvideoagent * DK64_MASTER slaps forehead
Experienced player (828)
Joined: 11/18/2006
Posts: 2426
Location: Back where I belong
Fun run, with a game that looks like it's really fun to play. All that and an improvement too? Yes, duh.
Living Well Is The Best Revenge My Personal Page
Banned User
Joined: 12/23/2004
Posts: 1850
Ristar is still one of the most beautiful, but insane, games I've ever seen. It's really fun to watch, especially since it has water levels that aren't slow or boring! Voting no since it killed my lappy's battery, though.
Perma-banned
Joined: 2/1/2007
Posts: 245
Location: Israel
I loved the game, I like this run, I'd like to vote 'yes', so I'll do it any moment now.
Player (81)
Joined: 3/11/2005
Posts: 352
Location: Oregon
upthorn wrote:
<various changes to submission text>
even better!
ideamagnate| .seen aqfaq <nothing happens> DK64_MASTER| .seen nesvideoagent * DK64_MASTER slaps forehead
upthorn
He/Him
Emulator Coder, Active player (391)
Joined: 3/24/2006
Posts: 1802
Does anyone have any other questions or comments at all?
How fleeting are all human passions compared with the massive continuity of ducks.
Player (206)
Joined: 5/29/2004
Posts: 5712
I don't get what the purpose of the height bonuses is.
put yourself in my rocketpack if that poochie is one outrageous dude
upthorn
He/Him
Emulator Coder, Active player (391)
Joined: 3/24/2006
Posts: 1802
Bag of Magic Food wrote:
I don't get what the purpose of the height bonuses is.
The height bonuses are very difficult to achieve in real time play, and seemed like the least boring option to end levels with.
How fleeting are all human passions compared with the massive continuity of ducks.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15576
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [861] Genesis Ristar by upthorn in 26:44.15
Joined: 7/26/2006
Posts: 53
"grabbing the old man's monkey better" Gets my yes vote just for that, lovely run
Joined: 7/5/2007
Posts: 1
Impressive run. I was only introduced to tool-assisted videos today, and I just had to see the video of my favorite Genesis game. The way that you beat Greedy was stupidly impressive. If there's one thing wrong, though, it's the time to beat Riho. I have a method where I hit the monkey thing right off the guy's back, and I'm pretty sure it's faster than even the tool-assisted method. Other than that, that's about as perfect as you can get. :)
Site Admin, Skilled player (1254)
Joined: 4/17/2010
Posts: 11475
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
HD encode arrived. Playlist Part 1 Part 2
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.